[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"cms-page:\u002Fabout\u002Four-team\u002Fharry-brace\u002Fpage\u002F3":3},{"name":4,"url":5,"description":6,"avatar_urls":7,"acf":9,"social":197,"head_links":199,"breadcrumbs":185,"posts":203,"pages":180,"head":1474},"Harry Brace","","I'm the Director, Head of Solutions at Impression. I spend my time debugging and implementing tracking using Google Tag Manager, managing data warehousing platforms and building Looker Studio reports!",{"24":8,"48":8,"96":8},"https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress&fit=fit&fm=jpg&h=96&q=90&w=96",{"full_bio":10,"job_title":11,"specialist_service_links":12,"photo":49,"page_content_modules":105,"hero":181},"\u003Cp>Director, Head of Solutions. Joined Impression in November 2017.\u003C\u002Fp>\n\u003Cp>I gained 4 years of experience as a web developer before moving to Impression as a digital account manager in 2017. I moved across to the Analytics department in 2019 and I’ve been strengthening and growing our Analytics offering ever since.\u003C\u002Fp>\n\u003Cp>Day to day, I spend my time debugging and implementing tracking using Google Tag Manager, managing data warehousing platforms and building Looker Studio reports!\u003C\u002Fp>\n\u003Cp>\u003Cstrong>Something that I’m proud of:\u003C\u002Fstrong>\u003Cbr \u002F>\nBuilding the analytics department over the last couple of years with the help of Aaron and now Alex H.\u003C\u002Fp>\n\u003Cp>\u003Cstrong>Outside of work:\u003C\u002Fstrong>\u003Cbr \u002F>\nFind me socialising, skiing or gaming.\u003C\u002Fp>\n","Director, Head of Solutions",[13,30,40],{"page":14,"link_text":29},{"ID":15,"post_author":16,"post_date":17,"post_date_gmt":17,"post_content":5,"post_title":18,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":21,"to_ping":5,"pinged":5,"post_modified":22,"post_modified_gmt":22,"post_content_filtered":5,"post_parent":23,"guid":24,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":28},374,"98","2021-01-11 16:25:32","Analytics","publish","closed","analytics","2026-09-28 11:36:16",0,"\u002F?page_id=374","page","0","raw","\u002Fanalytics\u002F","Web Analytics",{"page":31,"link_text":34},{"ID":32,"post_author":16,"post_date":33,"post_date_gmt":33,"post_content":5,"post_title":34,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":35,"to_ping":5,"pinged":5,"post_modified":36,"post_modified_gmt":36,"post_content_filtered":5,"post_parent":37,"guid":38,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":39},17916,"2022-06-20 08:56:55","Tag Management","tag-management","2026-09-28 11:05:27",17552,"\u002F?page_id=17916","\u002Fmedia-solutions\u002Ftag-management\u002F",{"page":41,"link_text":44},{"ID":42,"post_author":16,"post_date":43,"post_date_gmt":43,"post_content":5,"post_title":44,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":45,"to_ping":5,"pinged":5,"post_modified":46,"post_modified_gmt":46,"post_content_filtered":5,"post_parent":37,"guid":47,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":48},17140,"2022-05-27 14:22:19","Google Marketing Platform","google-marketing-platform","2026-09-25 10:43:13","\u002F?page_id=17140","\u002Fmedia-solutions\u002Fgoogle-marketing-platform\u002F",{"ID":50,"id":50,"title":51,"filename":52,"filesize":53,"url":54,"link":55,"alt":5,"author":56,"description":5,"caption":5,"name":57,"status":58,"uploaded_to":23,"date":59,"modified":59,"menu_order":23,"mime_type":60,"type":61,"subtype":62,"icon":63,"width":64,"height":65,"sizes":66},25135,"Harry-Brace2","Harry-Brace2.png",306919,"https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress%2Cformat&q=90","\u002Fharry-brace2\u002F","12","harry-brace2","inherit","2023-10-03 16:03:03","image\u002Fpng","image","png","\u002Fwp\u002Fwp-includes\u002Fimages\u002Fmedia\u002Fdefault.png",696,600,{"thumbnail":67,"thumbnail-width":68,"thumbnail-height":68,"medium":69,"medium-width":70,"medium-height":71,"medium_large":72,"medium_large-width":73,"medium_large-height":74,"large":75,"large-width":76,"large-height":77,"1536x1536":78,"1536x1536-width":79,"1536x1536-height":80,"2048x2048":81,"2048x2048-width":82,"2048x2048-height":83,"post-thumbnail":84,"post-thumbnail-width":85,"post-thumbnail-height":86,"12_col":87,"12_col-width":88,"12_col-height":89,"default_landscape":90,"default_landscape-width":91,"default_landscape-height":65,"default_portrait":92,"default_portrait-width":65,"default_portrait-height":91,"6_col_and_margin_short":93,"6_col_and_margin_short-width":85,"6_col_and_margin_short-height":89,"11_col_and_margin":94,"11_col_and_margin-width":95,"11_col_and_margin-height":89,"9_col_rectangle":96,"9_col_rectangle-width":97,"9_col_rectangle-height":98,"11_col":99,"11_col-width":100,"11_col-height":89,"staff_image_696_600":101,"staff_image_696_600-width":64,"staff_image_696_600-height":65,"rectangular_thumbnail":102,"rectangular_thumbnail-width":103,"rectangular_thumbnail-height":104},"https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress%2Cformat&fit=crop&h=640&q=90&w=640","640","https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress%2Cformat&fit=scale&h=172&q=90&w=200",200,172,"https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress%2Cformat&fit=scale&h=662&q=90&w=768",768,662,"https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress%2Cformat&fit=crop&h=1000&q=90&w=1892","1892","1000","https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress%2Cformat&fit=scale&h=1324&q=90&w=1536",1536,1324,"https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress%2Cformat&fit=scale&h=1766&q=90&w=2048",2048,1766,"https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress%2Cformat&fit=crop&h=1180&q=90&w=1860",1860,1180,"https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress%2Cformat&fit=crop&h=870&q=90&w=2880",2880,870,"https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress%2Cformat&fit=crop&h=600&q=90&w=800",800,"https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress%2Cformat&fit=crop&h=800&q=90&w=600","https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress%2Cformat&fit=crop&h=870&q=90&w=1860","https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress%2Cformat&fit=crop&h=870&q=90&w=3080",3080,"https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress%2Cformat&fit=crop&h=1400&q=90&w=2150",2150,1400,"https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress%2Cformat&fit=crop&h=870&q=90&w=2640",2640,"https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress%2Cformat&fit=crop&h=600&q=90&w=696","https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress%2Cformat&fit=crop&h=300&q=90&w=750",750,300,[106],{"acf_fc_layout":107,"reduced":108,"top_overlap":-1,"bottom_underlap":-1,"disable_observer":-1,"title":109,"posts":110},"author_posts",true,"Harry Brace’s posts",{"data":111,"pages":180},[112,126,136,146,158,168],{"type_key":113,"title":114,"excerpt":115,"to":116,"postedAt":117,"readingTime":118,"image":119,"authors":120,"category":123},"post","The importance of calibration in Marketing Mix Modelling","\u003Cp>What is calibration in Marketing Mix Modelling? Calibration in Marketing Mix Modelling (“MMM”) is a crucial step that involves fine-tuning the model to ensure its predictions align closely with the incrementality and\u002For lift estimates derived from controlled experiments. This process is essential to enhance the model’s reliability and applicability to the real-world, making it more […]\u003C\u002Fp>\n","\u002Fblog\u002Fthe-importance-of-calibration-in-marketing-mix-modelling\u002F","2024-06-12T09:41:17",7,"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fries-bosch-SSREiOm8tMs-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=640&q=90&w=640",[121],{"name":4,"image":122},"https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress&fit=crop&fm=jpg&h=150&q=80&w=150&bg=CFE6F9",{"to":124,"name":125},"\u002Fblog\u002Fcategory\u002Fdigital-marketing\u002F","Digital Marketing",{"type_key":113,"title":127,"excerpt":128,"to":129,"postedAt":130,"readingTime":131,"image":132,"authors":133,"category":135},"Data requirements for Marketing Mix Modelling","\u003Cp>Marketing Mix Modelling (MMM) is a powerful statistical tool which aims to understand the impact of various marketing activities on sales and other key performance metrics. At the heart of MMM lies the need for high-quality, accurate data containing multiple variables such as advertising spend, promotions, and external marketing factors. Therefore, the reliability and completeness […]\u003C\u002Fp>\n","\u002Fblog\u002Fdata-requirements-for-marketing-mix-modelling\u002F","2024-05-24T10:48:39",8,"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpath-digital-tR0jvlsmCuQ-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=640&q=90&w=640",[134],{"name":4,"image":122},{"to":124,"name":125},{"type_key":113,"title":137,"excerpt":138,"to":139,"postedAt":140,"readingTime":141,"image":142,"authors":143,"category":145},"The Ultimate Guide to Incrementality Testing","\u003Cp>What does ‘incrementality’ mean in advertising? Incrementality refers to the causal improvement in revenue, or the KPI of interest, from an isolated activity.  Incrementality testing aims to measure the additional sales generated exclusively as a result of the advertising efforts. Incrementality testing provides a more precise understanding of the direct impact of advertising efforts on […]\u003C\u002Fp>\n","\u002Fblog\u002Fincrementality-testing\u002F","2024-10-14T13:51:39",35,"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpietro-jeng-n6B49lTx7NM-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=640&q=90&w=640",[144],{"name":4,"image":122},{"to":124,"name":125},{"type_key":113,"title":147,"excerpt":148,"to":149,"postedAt":150,"readingTime":151,"image":152,"authors":153,"category":155},"Key questions to consider when implementing marketing mix modelling","\u003Cp>Before diving headfirst into marketing mix modelling implementation, it’s essential to consider several crucial questions. From understanding the intricacies of variable selection to aligning modelling objectives with your business goals, thoughtful consideration of the following points can pave the way for more accurate and actionable insights. Below, you’ll find a list of key questions to […]\u003C\u002Fp>\n","\u002Fblog\u002Fnavigating-mmm\u002F","2024-04-18T13:51:27",2,"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Fjon-tyson-hhq1Lxtuwd8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=640&q=90&w=640",[154],{"name":4,"image":122},{"to":156,"name":157},"\u002Fblog\u002Fcategory\u002Fmeasurement\u002F","Measurement",{"type_key":113,"title":159,"excerpt":160,"to":161,"postedAt":162,"readingTime":163,"image":164,"authors":165,"category":167},"Bayesian Statistics in Media Mix Modelling: A real-world analogy","\u003Cp>Bayesian statistics is a powerful tool for tackling the complexities inherent in media effectiveness analysis, making it an attractive approach for practitioners in the field. Bayesian statistics particularly shines in media mix modelling (MMM) for its ability to incorporate prior knowledge and enhance model robustness with limited or missing data. Additionally, Bayesian methods facilitate the […]\u003C\u002Fp>\n","\u002Fblog\u002Fbayesian-statistics-in-media-mix-modelling-a-real-world-analogy\u002F","2024-03-22T10:21:02",3,"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002Fjakub-zerdzicki-ip7GFn5JqX8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=640&q=90&w=640",[166],{"name":4,"image":122},{"to":156,"name":157},{"type_key":113,"title":169,"excerpt":170,"to":171,"postedAt":172,"readingTime":173,"image":174,"authors":175,"category":177},"A beginners’ guide to Incrementality Testing and Media Mix Modelling","\u003Cp>Upon researching Media Effectiveness you might have stumbled across the following terms: media mix modelling, econometric modelling, incrementality testing, synthetic data, multiple linear regression, Bayesian, frequentist, R-squared, geo-experiments and multi-touch attribution. You’re not alone if you read all this and find the world of Media Effectiveness a little intimidating. For a discipline based on transparency […]\u003C\u002Fp>\n","\u002Fblog\u002Fguide-to-media-effectiveness-measurement\u002F","2024-02-06T13:57:02",15,"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Furiel-sc-11KDtiUWRq4-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=640&q=90&w=640",[176],{"name":4,"image":122},{"to":178,"name":179},"\u002Fblog\u002Fcategory\u002Fbeginners-guides\u002F","Beginners' Guides","6",{"type":182,"style":183,"title":184,"heading":4,"intro":10,"services":12,"breadcrumbs":185,"photo":67,"social":197},"author","dark","Meet the Team",[186,189,192,195],{"path":187,"name":188},"\u002F","Home",{"path":190,"name":191},"\u002Fabout\u002F","About Us",{"path":193,"name":194},"\u002Fabout\u002Four-team\u002F","Meet the team",{"path":196,"name":4},null,{"twitter":5,"linkedin":198},"https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fharrison-brace-137b8482\u002F",[200],{"rel":201,"href":202},"canonical","https:\u002F\u002Fwww.impressiondigital.com\u002Fabout\u002Four-team\u002Fharry-brace\u002F",[204,527,716,906,1093,1281],{"id":205,"date":206,"date_gmt":206,"guid":207,"modified":117,"modified_gmt":117,"slug":209,"status":19,"type":113,"link":116,"title":210,"content":211,"excerpt":214,"author":215,"featured_media":216,"comment_status":20,"ping_status":20,"sticky":213,"template":5,"format":217,"meta":218,"categories":221,"collections":224,"coauthors":225,"class_list":227,"acf":236,"head_links":237,"breadcrumbs":238,"url_path":116,"search_description":220,"featured_image_urls":245,"authors":280,"type_nicename":293,"reading_minutes":118,"category":294,"yoast_title":219,"yoast_meta":297,"yoast_json_ld":353,"_links":477},28980,"2024-06-12T09:39:50",{"rendered":208},"\u002F?p=28980","the-importance-of-calibration-in-marketing-mix-modelling",{"rendered":114},{"rendered":212,"protected":213},"\n\u003Cdiv class=\"table-of-contents\">\u003Cul>\u003Cli class=\"heading-h2\">\u003Ca href=\"#what-is-calibration-in-marketing-mix-modelling\">What is calibration in Marketing Mix Modelling?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#using-experiments-to-inform-marketing-mix-modelling\">Using experiments to inform Marketing Mix Modelling\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#using-marketing-mix-modelling-to-inform-experimentation-strategy\">Using Marketing Mix Modelling to inform experimentation strategy\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#how-does-calibration-work\">How does calibration work?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h3\">\u003Ca href=\"#calibration-through-the-prior-distributions\">Calibration through the prior distributions\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h3\">\u003Ca href=\"#calibration-through-the-likelihood\">Calibration through the likelihood\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#why-is-it-important\">Why is it important?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h3\">\u003Ca href=\"#introduces-causality\">Introduces causality\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h3\">\u003Ca href=\"#reduces-uncertainty\">Reduces uncertainty\u003C\u002Fa>\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdiv>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"what-is-calibration-in-marketing-mix-modelling\">What is calibration in Marketing Mix Modelling?\u003C\u002Fh2>\n\n\n\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"1242\" height=\"798\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fmmmincrementality-testing.png?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-28998\"\u002F>\u003C\u002Ffigure>\u003C\u002Fdiv>\n\n\n\u003Cp>\u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fincrementality-testing\u002F#how-does-incrementality-testing-complement-marketing-mix-modelling\">Calibration\u003C\u002Fa> in \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fmedia-mix-modelling\u002F\">Marketing Mix Modelling\u003C\u002Fa> (“MMM”) is a crucial step that involves fine-tuning the model to ensure its predictions align closely with the \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fincrementality-testing\u002F\">incrementality\u003C\u002Fa> and\u002For lift estimates derived from controlled experiments. This process is essential to enhance the model’s reliability and applicability to the real-world, making it more effective in allocating the impact of various marketing activities on sales. Calibration adjusts the model parameters based on \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fdata-requirements-for-marketing-mix-modelling\u002F\">historical data\u003C\u002Fa> and experiments, helping to account for vast uncertainty and\u002For discrepancies that may arise from incomplete or noisy data. \u003C\u002Fp>\n\n\n\n\u003Cp>By calibrating your MMM, you can ensure that the MMM accurately reflects real-world scenarios, thus providing more dependable and casual insights for decision-making.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"using-experiments-to-inform-marketing-mix-modelling\">Using experiments to inform Marketing Mix Modelling\u003C\u002Fh2>\n\n\n\n\u003Cp>Incrementality and lift tests measure the additional effect that a marketing action has on sales compared to a \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fincrementality-testing\u002F#weighted-averages\">baseline scenario\u003C\u002Fa> without that action. These tests help determine the true impact of marketing activities by isolating the effect of individual campaigns or channels. Incrementality tests, often conducted in the form of a \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fincrementality-testing\u002F#geo-experiments\">geo-experiment\u003C\u002Fa>, compares the performance of a group exposed to a marketing action with a control group that is not. This comparison helps establish a causal relationship between the marketing activity and observed changes in sales. By incorporating the results of these tests into MMM, you can better understand the causal impact of your marketing efforts and account for the unique complexities associated with your business, leading to more informed budget allocation and strategy decisions.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"using-marketing-mix-modelling-to-inform-experimentation-strategy\">Using Marketing Mix Modelling to inform experimentation strategy\u003C\u002Fh2>\n\n\n\n\u003Cp>In addition to using experimentation to inform MMM, MMM can be used to suggest which marketing channels or campaign types should be tested further. These are just two subsections of what is considered \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fincrementality-testing\u002F#triangulation-in-marketing\">triangulation in Marketing\u003C\u002Fa>. MMM identifies the channels and campaigns that have the most significant impact on sales and (in the case of \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fbayesian-statistics-in-media-mix-modelling-a-real-world-analogy\u002F\">Bayesian\u003C\u002Fa> MMM) their associated uncertainty. These insights help pinpoint which areas of your media mix are likely to yield the highest returns and how confident we can be with these insights.\u003C\u002Fp>\n\n\n\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"1204\" height=\"584\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fincrementality-testing-experiment-example.jpeg?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-28997\"\u002F>\u003C\u002Ffigure>\u003C\u002Fdiv>\n\n\n\u003Cp>For example, if MMM suggests that your paid social channel drives high sales, and is relatively low down on the \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fmedia-mix-modelling\u002F\">response curve\u003C\u002Fa>, you may be inclined to invest more into that channel as the MMM suggests room to grow. However, if the MMM reports an average ROAS of 8, with lower and upper estimates of 1 and 15, you might decide to conduct a detailed incrementality test in which budgets are only increased in one region and the additional impact on sales is observed. By doing this you’ll gain greater confidence in what to expect if you were to increase budgets across all regions and significantly reduce the chance of wasting any additional budget on an ineffective channel. This targeted approach ensures that marketing resources are invested in the most promising areas, maximising ROI.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"how-does-calibration-work\">How does calibration work?\u003C\u002Fh2>\n\n\n\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"816\" height=\"294\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fbayesian-statistics-calibration.png?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-28995\"\u002F>\u003C\u002Ffigure>\u003C\u002Fdiv>\n\n\n\u003Ch3 class=\"wp-block-heading\" id=\"calibration-through-the-prior-distributions\">Calibration through the prior distributions\u003C\u002Fh3>\n\n\n\n\u003Cp>\u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fguide-to-media-effectiveness-measurement\u002F#bayesian-vs-frequentist-statistics\">Bayesian Marketing Mix Modelling\u003C\u002Fa> introduces a probabilistic framework that allows for the incorporation of prior knowledge and the quantification of uncertainty in model estimates. Calibration in Bayesian MMM involves the fine-tuning of priors, which are essential components that influence the posterior distributions of the model parameters.\u003C\u002Fp>\n\n\n\n\u003Cp>Priors represent the initial beliefs about the parameters before observing any historical data. These beliefs can be informed by industry benchmarks, expert opinions, or previous studies. For instance, a non-negative prior distribution might be used to enforce the assumption that \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fwww.warc.com\u002Fnewsandopinion\u002Fopinion\u002Fles-binet-on-why-long-term-marketing-matters-in-the-age-of-short-termism\u002Fen-gb\u002F3307#:~:text=Any%20advertising%20or%20marketing%20campaign,short%2Dterm%20metrics%20with%20caution.\">advertising cannot have a negative impact on sales\u003C\u002Fa>. \u003C\u002Fp>\n\n\n\n\u003Cp>The calibration process involves iterative adjustments to the parameters of the prior distribution to improve the model’s generalisability. This can be done by comparing the model’s predictions to actual observed outcomes from a controlled experiment and adjusting the priors to reduce discrepancies. For example, if the initial priors lead to overestimating the impact of a particular channel, the mean of your prior distribution can be adjusted to reflect a lower expected impact.\u003C\u002Fp>\n\n\n\n\u003Cp>After adjusting the priors, the calibrated model is evaluated using techniques like \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fdataaspirant.com\u002Fleave-one-out-cross-validation-loocv\u002F#:~:text=Leave%2Done%2Dout%20cross%20validation%20(LOOCV)%20is%20a,to%20evaluate%20the%20model's%20performance.\">cross-validation\u003C\u002Fa>, \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fmc-stan.org\u002Fdocs\u002Fstan-users-guide\u002Fposterior-predictive-checks.html\">posterior predictive checks\u003C\u002Fa>, and comparing the model’s predictions with \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fincrementality-testing\u002F#hold-out-testing-for-the-synthetic-control\">holdout samples\u003C\u002Fa>. This evaluation ensures that the priors are set in a way that enhances the model’s ability to generalise to new data.\u003C\u002Fp>\n\n\n\n\u003Cfigure class=\"wp-block-image size-full\">\u003Ca rel=\"noreferrer noopener\" href=\"\u002Fmedia-solutions\u002Fmeasurement\u002F\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"3840\" height=\"540\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Ff0sKYLsk-Measurement_Banner_01.jpg?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-27925\"\u002F>\u003C\u002Fa>\u003C\u002Ffigure>\n\n\n\n\u003Ch3 class=\"wp-block-heading\" id=\"calibration-through-the-likelihood\">Calibration through the likelihood\u003C\u002Fh3>\n\n\n\n\u003Cp>In addition to incorporating prior knowledge, Bayesian MMM produces insights through a combination of the \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fbayesian-statistics-in-media-mix-modelling-a-real-world-analogy\u002F\">prior and the likelihood\u003C\u002Fa>. The likelihood simply describes the probability of observing the data given our priors. \u003C\u002Fp>\n\n\n\n\u003Cp>Calibration through the likelihood function allows for the adjustment of parameter influence over time by adding additional lift test observations into the likelihood. This method ensures that the model remains flexible and responsive to changes in the data. The main benefit of this approach to calibration is that the parameter adjustment isn’t fixed throughout the entire data set and it is updated at the time of the lift test, enabling more realistic and dynamic modelling. However, it is important to note that this approach does not always reduce uncertainty, as the variability inherent in the data can still persist.\u003C\u002Fp>\n\n\n\n\u003Cp>In the context of Bayesian \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002F1749.io\u002Flearn\u002Ff\u002Fbayesian-hierarchical-models-marketing-mix-models-mmm?blogcategory=Methodologies#:~:text=Marketing%20data%20is%20often%20fragmented,advertising%20channels%20to%20creative%20executions.\">hierarchical modelling\u003C\u002Fa>, in which numerous channels are associated with an “overall” hyperparameter, calibration can be particularly beneficial. The most common hierarchical structure in MMM involves parameter means. For instance, you may choose to model the performance of Meta ads, Pinterest ads and LinkedIn ads, with individual parameters that share an overall parameter called the “paid social mean”, capturing both the individual channel impacts and their collective influence on overall sales. \u003C\u002Fp>\n\n\n\n\u003Cp>When modelling with a hierarchical structure, a single lift test can inform multiple channels, reducing the need for extensive testing across all channels and minimising disruption to business as usual. In this case, one incrementality test for Meta might influence the contribution of both Pinterest and Linkedin. This approach leverages the hierarchical structure to share information across different levels, improving overall model performance and reducing the operational impact of testing.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"why-is-it-important\">Why is it important?\u003C\u002Fh2>\n\n\n\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"1920\" height=\"1080\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fcalibration-importance.png?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-28996\"\u002F>\u003C\u002Ffigure>\u003C\u002Fdiv>\n\n\n\u003Cp>Calibration in MMM is crucial for ensuring the reliability and applicability of the model’s predictions. By fine-tuning the model’s parameters to align more closely with observed data from lift and incrementality tests, calibration enables the model to better capture any unique complexities associated with your business. This process ensures that the MMM can better predict the true effects of different marketing activities, making the insights derived from the model more actionable and trustworthy. Without proper calibration, the model’s outputs could be significantly off, leading to poor decision-making and suboptimal allocation of marketing resources.\u003C\u002Fp>\n\n\n\n\u003Ch3 class=\"wp-block-heading\" id=\"introduces-causality\">Introduces causality\u003C\u002Fh3>\n\n\n\n\u003Cp>Calibration introduces \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002F1749.io\u002Flearn\u002Ff\u002Fquantifying-retail-media-impact-with-causalpy-causal-inference\">causality\u003C\u002Fa> into MMM. By incorporating lift tests and incrementality tests into the calibration process, you can more accurately isolate the effects of individual marketing activities and establish a clearer causal relationship between these activities and sales outcomes. This approach allows for the differentiation between correlation and causation, providing more robust insights into which marketing tactics are genuinely driving sales. Consequently, you can make more informed decisions about where to invest their marketing budgets, optimising your media mix for maximum impact.\u003C\u002Fp>\n\n\n\n\u003Ch3 class=\"wp-block-heading\" id=\"reduces-uncertainty\">Reduces uncertainty\u003C\u002Fh3>\n\n\n\n\u003Cp>Calibration also plays a vital role in reducing the \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002F1749.io\u002Flearn\u002Ff\u002Funcertainty-in-marketing-mix-modeling-optimizing-sales\">uncertainty\u003C\u002Fa> associated with marketing effectiveness and enhancing the generalisability of MMM. By continuously adjusting the model to reflect new data, calibration ensures that the model remains relevant and accurate over time. This adaptability helps to account for external factors and fluctuations, reducing the margin of error in the model’s predictions\u003C\u002Fp>\n\n\n\n\u003Cp>As previously mentioned, calibration in hierarchical modelling can inform multiple channels. As a result, we might find that the results of one incrementality or lift test not only decrease the uncertainty of one particular but any channels associated with the same hyperparameter. Again, this reduces the need for extensive testing and disruptions to business operations. This shared learning across channels enhances the model’s overall robustness and applicability, making it a more powerful tool for strategic decision-making.\u003C\u002Fp>\n\n\n\n\u003Cp>As we wrap up this exploration into the importance of calibration in MMM, it’s clear that incorporating insights derived from incrementality tests in your model is not just a technical necessity, but a strategic advantage. Incorporating causality into your model via incrementality tests and lift tests transforms your MMM into a robust mechanism for understanding true cause and effect. By embracing the principles and practices of calibration in MMM, you position your marketing efforts to achieve greater accuracy, accountability, and effectiveness.\u003C\u002Fp>\n\n\n\n\u003Cfigure class=\"wp-block-image size-full\">\u003Ca rel=\"noreferrer noopener\" href=\"\u002Fget-in-touch\u002F\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"3840\" height=\"540\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002FGeneric_Banner_01.jpg?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-27527\"\u002F>\u003C\u002Fa>\u003C\u002Ffigure>\n",false,{"rendered":115,"protected":213},29,29001,"standard",{"_yoast_wpseo_focuskw":5,"_yoast_wpseo_title":219,"_yoast_wpseo_metadesc":220,"inline_featured_image":213,"footnotes":5},"The importance of calibration in Marketing Mix Modelling | Impression","Calibration in Marketing Mix Modelling is a crucial step that fine-tunes the model to ensure its predictions align closely with the incrementality.",[222,223],39,41,[],[226],120,[228,113,229,230,231,232,233,234,235],"post-28980","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-digital-marketing","category-measurement",[],[],[239,240,243,244],{"path":187,"name":188},{"path":241,"name":242},"\u002Fblog\u002F","Blog",{"path":156,"name":157},{"path":196,"name":114},{"thumbnail":246,"medium":248,"medium_large":250,"large":252,"1536x1536":256,"2048x2048":258,"post-thumbnail":260,"12_col":262,"default_landscape":264,"default_portrait":266,"6_col_and_margin_short":268,"11_col_and_margin":270,"9_col_rectangle":272,"11_col":274,"staff_image_696_600":276,"rectangular_thumbnail":278},{"url":119,"width":247,"height":247},640,{"url":249,"width":70,"height":70},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fries-bosch-SSREiOm8tMs-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=139&q=90&w=200",{"url":251,"width":73,"height":23},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fries-bosch-SSREiOm8tMs-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=532&q=90&w=768",{"url":253,"width":254,"height":255},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fries-bosch-SSREiOm8tMs-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=1000&q=90&w=1892",1892,1000,{"url":257,"width":79,"height":79},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fries-bosch-SSREiOm8tMs-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=1064&q=90&w=1536",{"url":259,"width":82,"height":82},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fries-bosch-SSREiOm8tMs-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=1418&q=90&w=2048",{"url":261,"width":85,"height":86},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fries-bosch-SSREiOm8tMs-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=1180&q=90&w=1860",{"url":263,"width":88,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fries-bosch-SSREiOm8tMs-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=2880",{"url":265,"width":91,"height":65},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fries-bosch-SSREiOm8tMs-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=600&q=90&w=800",{"url":267,"width":65,"height":91},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fries-bosch-SSREiOm8tMs-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=800&q=90&w=600",{"url":269,"width":85,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fries-bosch-SSREiOm8tMs-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=1860",{"url":271,"width":95,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fries-bosch-SSREiOm8tMs-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=3080",{"url":273,"width":97,"height":98},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fries-bosch-SSREiOm8tMs-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=1400&q=90&w=2150",{"url":275,"width":100,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fries-bosch-SSREiOm8tMs-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=2640",{"url":277,"width":64,"height":65},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fries-bosch-SSREiOm8tMs-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=600&q=90&w=696",{"url":279,"width":103,"height":104},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fries-bosch-SSREiOm8tMs-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=300&q=90&w=750",[281],{"id":215,"path":282,"display_name":4,"job_title":11,"description":6,"avatar_url":122,"acf":283},"\u002Fabout\u002Four-team\u002Fharry-brace\u002F",{"full_bio":10,"job_title":11,"specialist_service_links":284,"photo":291},[285,287,289],{"page":286,"link_text":29},{"ID":15,"post_author":16,"post_date":17,"post_date_gmt":17,"post_content":5,"post_title":18,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":21,"to_ping":5,"pinged":5,"post_modified":22,"post_modified_gmt":22,"post_content_filtered":5,"post_parent":23,"guid":24,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":28},{"page":288,"link_text":34},{"ID":32,"post_author":16,"post_date":33,"post_date_gmt":33,"post_content":5,"post_title":34,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":35,"to_ping":5,"pinged":5,"post_modified":36,"post_modified_gmt":36,"post_content_filtered":5,"post_parent":37,"guid":38,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":39},{"page":290,"link_text":44},{"ID":42,"post_author":16,"post_date":43,"post_date_gmt":43,"post_content":5,"post_title":44,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":45,"to_ping":5,"pinged":5,"post_modified":46,"post_modified_gmt":46,"post_content_filtered":5,"post_parent":37,"guid":47,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":48},{"ID":50,"id":50,"title":51,"filename":52,"filesize":53,"url":54,"link":55,"alt":5,"author":56,"description":5,"caption":5,"name":57,"status":58,"uploaded_to":23,"date":59,"modified":59,"menu_order":23,"mime_type":60,"type":61,"subtype":62,"icon":63,"width":64,"height":65,"sizes":292},{"thumbnail":67,"thumbnail-width":68,"thumbnail-height":68,"medium":69,"medium-width":70,"medium-height":71,"medium_large":72,"medium_large-width":73,"medium_large-height":74,"large":75,"large-width":76,"large-height":77,"1536x1536":78,"1536x1536-width":79,"1536x1536-height":80,"2048x2048":81,"2048x2048-width":82,"2048x2048-height":83,"post-thumbnail":84,"post-thumbnail-width":85,"post-thumbnail-height":86,"12_col":87,"12_col-width":88,"12_col-height":89,"default_landscape":90,"default_landscape-width":91,"default_landscape-height":65,"default_portrait":92,"default_portrait-width":65,"default_portrait-height":91,"6_col_and_margin_short":93,"6_col_and_margin_short-width":85,"6_col_and_margin_short-height":89,"11_col_and_margin":94,"11_col_and_margin-width":95,"11_col_and_margin-height":89,"9_col_rectangle":96,"9_col_rectangle-width":97,"9_col_rectangle-height":98,"11_col":99,"11_col-width":100,"11_col-height":89,"staff_image_696_600":101,"staff_image_696_600-width":64,"staff_image_696_600-height":65,"rectangular_thumbnail":102,"rectangular_thumbnail-width":103,"rectangular_thumbnail-height":104},"Article",[295,296],{"id":222,"name":125,"path":124},{"id":223,"name":157,"path":156},[298,300,302,305,307,310,313,315,317,319,322,325,328,331,334,337,340,342,343,346,348,351],{"property":299,"content":219},"title",{"property":301,"content":220},"description",{"name":303,"content":304},"robots","index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1",{"property":201,"content":306},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fthe-importance-of-calibration-in-marketing-mix-modelling\u002F",{"name":308,"content":309},"og:locale","en_GB",{"property":311,"content":312},"og:type","article",{"property":314,"content":219},"og:title",{"property":316,"content":220},"og:description",{"property":318,"content":306},"og:url",{"property":320,"content":321},"og:site_name","Impression",{"property":323,"content":324},"article:publisher","https:\u002F\u002Fwww.facebook.com\u002Fimpression",{"property":326,"content":327},"article:published_time","2024-06-12T09:39:50+00:00",{"property":329,"content":330},"article:modified_time","2024-06-12T09:41:17+00:00",{"property":332,"content":333},"og:image","https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fopen-graph-incrementality-calibration.png?auto=compress%2Cformat&q=90",{"property":335,"content":336},"og:image:width",1200,{"property":338,"content":339},"og:image:height",630,{"property":341,"content":60},"og:image:type",{"property":182,"content":4},{"property":344,"content":345},"twitter:card","summary_large_image",{"property":347,"content":333},"twitter:image",{"property":349,"content":350},"twitter:creator","@impressiontalk",{"property":352,"content":350},"twitter:site",{"@context":354,"@graph":355},"https:\u002F\u002Fschema.org",[356,369,378,382,395,405,471],{"@type":293,"@id":357,"isPartOf":358,"author":359,"headline":114,"datePublished":327,"dateModified":330,"mainEntityOfPage":361,"wordCount":98,"publisher":362,"image":364,"thumbnailUrl":366,"articleSection":367,"inLanguage":368},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fthe-importance-of-calibration-in-marketing-mix-modelling\u002F#article",{"@id":306},{"name":4,"@id":360},"https:\u002F\u002Fadmin.impressiondigital.com\u002F#\u002Fschema\u002Fperson\u002Fa18b79c97abbcb2444ace02aa88250ae",{"@id":306},{"@id":363},"https:\u002F\u002Fadmin.impressiondigital.com\u002F#organization",{"@id":365},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fthe-importance-of-calibration-in-marketing-mix-modelling\u002F#primaryimage","https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F06\u002Fries-bosch-SSREiOm8tMs-unsplash.jpg?auto=compress%2Cformat&q=90",[125,157],"en-GB",{"@type":370,"@id":306,"url":306,"name":219,"isPartOf":371,"primaryImageOfPage":373,"image":374,"thumbnailUrl":366,"datePublished":327,"dateModified":330,"description":220,"breadcrumb":375,"inLanguage":377},"WebPage",{"@id":372},"https:\u002F\u002Fwww.impressiondigital.com\u002F#website",{"@id":365},{"@id":365},{"@id":376},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fthe-importance-of-calibration-in-marketing-mix-modelling\u002F#breadcrumb","en",{"@type":379,"inLanguage":368,"@id":365,"url":366,"contentUrl":366,"width":380,"height":381},"ImageObject",2400,1662,{"@type":383,"@id":376,"itemListElement":384},"BreadcrumbList",[385,389,391,393],{"@type":386,"position":387,"name":188,"item":388},"ListItem",1,"https:\u002F\u002Fwww.impressiondigital.com\u002F",{"@type":386,"position":151,"name":242,"item":390},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002F",{"@type":386,"position":163,"name":157,"item":392},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fcategory\u002Fmeasurement\u002F",{"@type":386,"position":394,"name":114},4,{"@type":396,"@id":372,"url":388,"name":321,"description":5,"publisher":397,"inLanguage":368,"potentialAction":399},"WebSite",{"@id":398},"https:\u002F\u002Fwww.impressiondigital.com\u002F#organization",{"@type":400,"target":401,"query-input":404},"SearchAction",{"@type":402,"urlTemplate":403},"EntryPoint","https:\u002F\u002Fwww.impressiondigital.com\u002Fsearch?q={search_term_string}","required name=search_term_string",{"@type":406,"@id":398,"name":321,"url":388,"logo":407,"image":412,"sameAs":413,"@context":417,"email":418,"foundingDate":419,"legalName":420,"vatID":421,"description":422,"telephone":423,"address":424,"location":431,"areaServed":464},"Organization",{"@type":379,"inLanguage":368,"@id":408,"url":409,"contentUrl":409,"width":410,"height":411,"caption":321},"https:\u002F\u002Fwww.impressiondigital.com\u002F#\u002Fschema\u002Flogo\u002Fimage\u002F","https:\u002F\u002Fcdn.impression.co.uk\u002F2021\u002F06\u002FA5LuTDrq-Impression_Wordmark.png",920,176,"https:\u002F\u002Fcdn.impression.co.uk\u002F2021\u002F04\u002FFAVxDeOS-Nottingham.jpg",[414,415,416],"https:\u002F\u002Fwww.linkedin.com\u002Fcompany\u002Fimpression-digital-limited","https:\u002F\u002Fwww.facebook.com\u002Fimpression\u002F","https:\u002F\u002Fwww.instagram.com\u002Fimpressiontalk\u002F","https:\u002F\u002Fwww.schema.org\u002F","hello@impressiondigital.com","2012-11-07","Impression Digital Limited","GB186363774","Impression is an award-winning digital marketing agency based in the UK and USA. We offer SEO agency, PPC agency, Digital PR agency and Analytics services to companies of all sizes across the UK, North America, Europe and beyond. Get in touch with us if you think we can work together.","01158 242 212",{"@type":425,"streetAddress":426,"addressLocality":427,"addressRegion":428,"postalCode":429,"addressCountry":430},"PostalAddress","Fothergill House, 16 King Street","Nottingham","Nottinghamshire","NG1 2AS","United Kingdom",[432,441,453],{"@type":433,"name":434,"geo":435,"address":439,"photo":412,"telephone":440},"Place","Impression Digital - Nottingham",{"@type":436,"latitude":437,"longitude":438},"GeoCoordinates","52.95501401025914","-1.1494887460285232",{"@type":425,"streetAddress":426,"addressLocality":427,"addressRegion":428,"postalCode":429,"addressCountry":430},"+44 01158 242 212",{"@type":433,"name":442,"geo":443,"address":446,"photo":451,"telephone":452},"Impression Digital - London",{"@type":436,"latitude":444,"longitude":445},"51.50237208377361","-0.09229841534284115",{"@type":425,"streetAddress":447,"addressLocality":448,"addressRegion":449,"postalCode":450,"addressCountry":430},"201 Borough High St","Southwark","London","SE1 1JA","https:\u002F\u002Fcdn.impression.co.uk\u002F2021\u002F04\u002FwD0pPVWz-London.jpg","+44 020 7112 8148",{"@type":433,"name":454,"geo":455,"address":458},"Impression Digital - New York",{"@type":436,"latitude":456,"longitude":457},"40.726722746949","-74.00527778220652",{"@type":425,"streetAddress":459,"addressLocality":460,"addressRegion":461,"postalCode":462,"addressCountry":463},"160 Varick St","New York","NY","10013","USA",[465,468,470],{"@type":466,"name":467},"Country","Worldwide",{"@type":466,"name":469},"GBR",{"@type":466,"name":463},{"@type":472,"@id":473,"name":4,"image":474,"description":6,"sameAs":476,"url":202},"Person","https:\u002F\u002Fwww.impressiondigital.com\u002F#\u002Fschema\u002Fperson\u002Fa18b79c97abbcb2444ace02aa88250ae",{"@type":379,"inLanguage":368,"@id":475,"url":8,"contentUrl":8,"caption":4},"https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress&fit=fit&fm=jpg&h=96&q=90&w=9615847ead018a2de28b7216e39fc92f32",[198],{"self":478,"collection":488,"about":491,"author":494,"replies":497,"version-history":500,"predecessor-version":504,"wp:featuredmedia":508,"wp:attachment":511,"wp:term":514,"curies":523},[479],{"href":480,"targetHints":481},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F28980",{"allow":482},[483,484,485,486,487],"GET","POST","PUT","PATCH","DELETE",[489],{"href":490},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts",[492],{"href":493},"\u002Fwp-json\u002Fwp\u002Fv2\u002Ftypes\u002Fpost",[495],{"embeddable":108,"href":496},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fusers\u002F29",[498],{"embeddable":108,"href":499},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcomments?post=28980",[501],{"count":502,"href":503},11,"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F28980\u002Frevisions",[505],{"id":506,"href":507},29005,"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F28980\u002Frevisions\u002F29005",[509],{"embeddable":108,"href":510},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fmedia\u002F29001",[512],{"href":513},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fmedia?parent=28980",[515,518,521],{"taxonomy":516,"embeddable":108,"href":517},"category","\u002Fwp-json\u002Fwp\u002Fv2\u002Fcategories?post=28980",{"taxonomy":519,"embeddable":108,"href":520},"collections","\u002Fwp-json\u002Fwp\u002Fv2\u002Fcollections?post=28980",{"taxonomy":182,"embeddable":108,"href":522},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcoauthors?post=28980",[524],{"name":525,"href":526,"templated":108},"wp","https:\u002F\u002Fapi.w.org\u002F{rel}",{"id":528,"date":529,"date_gmt":529,"guid":530,"modified":130,"modified_gmt":130,"slug":532,"status":19,"type":113,"link":129,"title":533,"content":534,"excerpt":536,"author":215,"featured_media":537,"comment_status":20,"ping_status":20,"sticky":213,"template":5,"format":217,"meta":538,"categories":541,"collections":542,"coauthors":543,"class_list":544,"acf":546,"head_links":547,"breadcrumbs":548,"url_path":129,"search_description":540,"featured_image_urls":553,"authors":585,"type_nicename":293,"reading_minutes":131,"category":597,"yoast_title":539,"yoast_meta":600,"yoast_json_ld":627,"_links":679},28804,"2024-05-24T10:39:36",{"rendered":531},"\u002F?p=28804","data-requirements-for-marketing-mix-modelling",{"rendered":127},{"rendered":535,"protected":213},"\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"1611\" height=\"611\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fdata-requirememnts-for-mmm-01.png?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-28806\"\u002F>\u003C\u002Ffigure>\u003C\u002Fdiv>\n\n\n\u003Cp>\u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fmedia-mix-modelling\u002F\">Marketing Mix Modelling\u003C\u002Fa> (MMM) is a powerful statistical tool which aims to understand the impact of various marketing activities on sales and other key performance metrics. At the heart of MMM lies the need for high-quality, accurate data containing multiple variables such as advertising spend, promotions, and external marketing factors. Therefore, the reliability and completeness of the data in MMM is crucial for generating actionable insights and making informed decisions in today’s competitive marketplace. \u003C\u002Fp>\n\n\n\n\u003Cp>Regardless of your objectives, all MMM data should meet the following specifications:\u003C\u002Fp>\n\n\n\n\u003Col class=\"wp-block-list\">\n\u003Cli>Of \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fwww.tableau.com\u002Flearn\u002Farticles\u002Ftime-series-analysis\">time series\u003C\u002Fa> format.\u003C\u002Fli>\n\n\n\n\u003Cli>Contains approximately 3 years of observations.\u003C\u002Fli>\n\n\n\n\u003Cli>As granular as possible, daily is preferred.\u003C\u002Fli>\n\n\n\n\u003Cli>As accurate as possible, not subject to errors and\u002For missing observations.\u003C\u002Fli>\n\n\n\n\u003Cli>Data boundaries must align, if you’re conducting a US-only MMM, make sure your data is also US-only.\u003C\u002Fli>\n\u003C\u002Fol>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"sales-data\">Sales data\u003C\u002Fh2>\n\n\n\n\u003Cp>In Marketing Mix Modeling, the decomposition of sales data serves as the focal point of analysis, representing the outcome that the model seeks to predict. The accuracy and reliability of your sales data is paramount, as any inaccuracies or inconsistencies can significantly impact the model’s predictive power. The following considerations are crucial when collecting your sales data:\u003C\u002Fp>\n\n\n\n\u003Col class=\"wp-block-list\">\n\u003Cli>Online only sales or overall sales. (Usually overall sales is preferred)\u003C\u002Fli>\n\n\n\n\u003Cli>B2B, B2C or both. (This will depend on the specific analysis) \u003C\u002Fli>\n\n\n\n\u003Cli>Product groups. (This will depend on the objectives of analysis) \u003C\u002Fli>\n\n\n\n\u003Cli>Sales volume or sales value. (Usually value is best for generating ROI insights) \u003C\u002Fli>\n\n\n\n\u003Cli>Source (we advise raw first party financial data).\u003C\u002Fli>\n\u003C\u002Fol>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"marketing-data\">Marketing data\u003C\u002Fh2>\n\n\n\n\u003Cp>Marketing Mix Models utilise marketing data to assess the influence of various marketing activities on sales. Consequently, the accuracy of your marketing data significantly affects the model’s estimation capabilities. In MMM, it is crucial to incorporate both your digital and offline marketing data.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"digital-marketing-data\">Digital marketing data\u003C\u002Fh2>\n\n\n\n\u003Cp>The most obvious data requirement for MMM is digital marketing data, but deciding which metric to use that best describes the performance of your marketing channel\u002Fcampaign can be challenging. All MMM’s require marketing spend data as, even if spend data isn’t used for parameter estimation, it’s required to produce ROI estimates and to help guide budget forecasts. \u003C\u002Fp>\n\n\n\n\u003Cp>Additionally, it is common for some MMM vendors to use engagement metrics such as impressions or clicks for parameter estimation, meaning that the contribution outputs produced by the model are derived using engagement metrics as opposed to spend. \u003C\u002Fp>\n\n\n\n\u003Cfigure class=\"wp-block-table\">\u003Ctable>\u003Cthead>\u003Ctr>\u003Cth>Metric\u003C\u002Fth>\u003Cth>Benefit\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd>Spend\u003C\u002Ftd>\u003Ctd>Offers a direct measure of investment in marketing channels, allowing for straightforward comparisons of resource allocation, saturation and ROI across different channels.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>Engagement metric (impressions, clicks…)\u003C\u002Ftd>\u003Ctd>Offers a more nuanced understanding of consumer behaviour, by capturing the level of interaction and interest generated by marketing activities.\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Ffigure>\n\n\n\n\u003Cp>Ultimately, the choice between engagement metrics and spend for parameter estimation depends on the specific objectives and priorities of the marketing analysis, with each approach offering unique insights into marketing performance.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"offline-marketing-activity\">Offline marketing activity\u003C\u002Fh2>\n\n\n\n\u003Cp>Incorporating offline marketing activity is essential in MMM because it provides a holistic view of how all your marketing efforts impact business outcomes. While digital marketing channels often receive significant attention due to their trackability and measurability, offline channels such as TV advertising, print media, outdoor campaigns and even affiliate activity still play a significant role in reaching and influencing consumers. Some examples and their associated metrics include:\u003C\u002Fp>\n\n\n\n\u003Cfigure class=\"wp-block-table\">\u003Ctable>\u003Cthead>\u003Ctr>\u003Cth>Variable\u003C\u002Fth>\u003Cth>Metrics\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd>TV\u003C\u002Ftd>\u003Ctd>Gross rating points (GRP’s).\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>Print\u003C\u002Ftd>\u003Ctd>Circulation, readership, gross spend\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>OOH (out of home)\u003C\u002Ftd>\u003Ctd>Net spend, potential reach.\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Ffigure>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"calibration-data\">Calibration data\u003C\u002Fh2>\n\n\n\n\u003Cp>Calibration in MMM refers to the process of fine-tuning the model parameters to ensure that the predicted outcomes align closely with actual observed data from \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fincrementality-testing\u002F\">incrementality tests\u003C\u002Fa>, thereby enhancing the model’s accuracy and reliability. Provided you’ve recently performed an incrementality, incorporating the following elements of your incrementality test can help the model to provide greater causal insights.\u003C\u002Fp>\n\n\n\n\u003Col class=\"wp-block-list\">\n\u003Cli>The time & date at which the test took place.\u003C\u002Fli>\n\n\n\n\u003Cli>The nature of the treatment (examples include switching advertising on\u002Foff, increasing\u002Freducing spend by x%).\u003C\u002Fli>\n\n\n\n\u003Cli>The observed impact.\u003C\u002Fli>\n\u003C\u002Fol>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"control-data\">Control data\u003C\u002Fh2>\n\n\n\n\u003Cp>In Marketing mix modelling, control data refers to the variables or factors that are not directly influenced by marketing activities and are used as benchmarks or reference points for comparison. These variables typically include factors such as seasonality, economic conditions, significant events, and \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fplanning-for-the-peak-selling-season\u002F\">peak selling seasons\u003C\u002Fa> that may impact sales. By incorporating control data into the model, we can isolate the impact of marketing efforts from other external influences. The following sections outline some common variables to include in your control data.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"pricing-information\">Pricing information\u003C\u002Fh2>\n\n\n\n\u003Cp>The aim of including a price-related control variable is to help account for changes in pricing dynamics and customer purchasing behaviour which may affect sales independently of marketing efforts. This ensures that changes in the sales attributed to marketing activities are not confounded by changes in price, providing a more accurate assessment of marketing effectiveness. The following are some commonly used pricing variables in MMM:\u003C\u002Fp>\n\n\n\n\u003Cfigure class=\"wp-block-table\">\u003Ctable>\u003Cthead>\u003Ctr>\u003Cth>Variable\u003C\u002Fth>\u003Cth>Benefit\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd>Average order value (AOV)\u003C\u002Ftd>\u003Ctd>Allows greater understanding of how changes in marketing strategies impact customer spending behaviour.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fwww.wiser.com\u002Fwp-content\u002Fuploads\u002F2018\u002F02\u002FWhat-is-Market-Price-Index.pdf\">Market price index\u003C\u002Fa> (MPI)\u003C\u002Ftd>\u003Ctd>Allows you to account for fluctuations in prices of goods or services, which may affect sales independently of marketing efforts. \u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Ffigure>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"seasonality\">Seasonality\u003C\u002Fh2>\n\n\n\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"1611\" height=\"611\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fdata-requirememnts-for-mmm-2.png?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-28825\"\u002F>\u003C\u002Ffigure>\u003C\u002Fdiv>\n\n\n\u003Cp>Seasonality refers to the recurring and predictable patterns or fluctuations in data that follow a regular and periodic cycle over time. \u003Cstrong>Seasonality data is usually \u003C\u002Fstrong>\u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fwhat-is-synthetic-data\u002F\">\u003Cstrong>synthetically\u003C\u002Fstrong>\u003C\u002Fa>\u003Cstrong> derived by the MMM practitioner \u003C\u002Fstrong>using various methods\u003Cstrong>.\u003C\u002Fstrong>\u003C\u002Fp>\n\n\n\n\u003Cp>Care must be taken when accounting for seasonality in MMM as marketing activity performs better when demand is high! In some cases, you find that controlling for seasonality will under-credit your marketing performance when your sales peak. If you find yourself in this position, you may wish to consider incorporating \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002F1749.io\u002Flearn\u002Ff\u002Fmarketing-mix-modeling-in-a-modern-era-time-varying-parameters\">time variant parameters\u003C\u002Fa> into your marketing mix model.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"events\">Events\u003C\u002Fh2>\n\n\n\n\u003Cp>Accounting for significant events such as promotions or seasonal holidays is essential in MMM because these events can impact consumer behaviour and purchasing patterns. By incorporating these events into the model, you can better understand the relationships between marketing activities and sales performance without the confounding influence of significant events. For this reason, it is essential that you keep track of any promotion data. Some examples of event data include the following:\u003C\u002Fp>\n\n\n\n\u003Cfigure class=\"wp-block-table\">\u003Ctable>\u003Cthead>\u003Ctr>\u003Cth>Event\u003C\u002Fth>\u003Cth>Data required\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd>Promotions\u003C\u002Ftd>\u003Ctd>Nature of the promotion (x% off), the start and end date.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>Public holidays\u003C\u002Ftd>\u003Ctd>The date of the public holiday.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>New product launches\u003C\u002Ftd>\u003Ctd>The date of the new product launch.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>Significant business changes\u003C\u002Ftd>\u003Ctd>Nature of the change and the relevant dates at which this change occurred.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>COVID-19\u003C\u002Ftd>\u003Ctd>Dates of lockdown periods.\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Ffigure>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"trend-baseline-sales\">Trend & Baseline sales\u003C\u002Fh2>\n\n\n\n\u003Cfigure class=\"wp-block-image size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"1611\" height=\"611\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fdata-requirememnts-for-mmm-3.png?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-28826\"\u002F>\u003C\u002Ffigure>\n\n\n\n\u003Cp>Trend in data refers to the long-term direction or pattern of change in a variable over time. It represents the overall tendency of the data to increase, decrease, or remain relatively stable over an extended period. Similarly to seasonality, \u003Cstrong>trend data is usually derived by the MMM practitioner\u003C\u002Fstrong> using a number of different methods.\u003C\u002Fp>\n\n\n\n\u003Cp>Baseline sales in MMM simply refers to how sales would behave in absence of all other variables. In MMM this output is usually referred to as the “intercept”. For most MMM’s, this intercept is modelled as constant, however at Impression we believe that baseline sales should be allowed to fluctuate over time if the data suggests so. To allow baseline sales to fluctuate overtime we treat the intercept as \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fmedium.com\u002F@nialloulton\u002Festimating-the-long-term-base-in-marketing-mix-models-864782323109\">time-variant\u003C\u002Fa>, in doing this we simultaneously aim to account for trends.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"macroeconomics\">Macroeconomics \u003C\u002Fh2>\n\n\n\n\u003Cp>It is essential to include \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fwww.worldbank.org\u002Fen\u002Ftopic\u002Fmacroeconomics#:~:text=Macroeconomics%20focuses%20on%20the%20performance,Overview\">macroeconomic\u003C\u002Fa> variables in MMM as they help account for external factors that can significantly influence consumer behaviour and marketing performance. Understanding the influence that purchasing power, disposable income and customer confidence has on demand enables you to isolate the impact of your marketing from broader economic trends, leading to more precise and actionable insights. The following are some examples of macroeconomic factors to include in MMM:\u003C\u002Fp>\n\n\n\n\u003Cfigure class=\"wp-block-table\">\u003Ctable>\u003Cthead>\u003Ctr>\u003Cth>Variable\u003C\u002Fth>\u003Cth>Benefit\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd>\u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fwww.investopedia.com\u002Fterms\u002Fc\u002Fconsumerpriceindex.asp\">Customer price index\u003C\u002Fa> (CPI)\u003C\u002Ftd>\u003Ctd>Captures shifts in consumer spending patterns\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>Unemployment rate\u003C\u002Ftd>\u003Ctd>Captures consumer confidence and disposable income.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>Exchange rates\u003C\u002Ftd>\u003Ctd>Captures the fluctuations in the cost of imported goods.\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Ffigure>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"literally-anything-else\">Literally anything else\u003C\u002Fh2>\n\n\n\n\u003Cp>Each business is different and has unique complexities associated with it. When thinking about MMM, if you’re modelling sales, make sure you consider as many factors as possible that could influence sales. Even if some factors were found to have very little or no impact by the MMM, providing as much information as possible prior to modelling is essential! \u003C\u002Fp>\n\n\n\n\u003Cp>Although we recommend providing as much data as possible, we do not recommend building an MMM and throwing all your available data at it. Including all of the above control data in less complex models might be overkill. Too many variables may not only lead to model overfitting, but also multicollinearity if many of these variables behave similarly. The art of a good MMM is therefore variable selection, which is where we as skilled practitioners step in to overcome this and many other \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fnavigating-mmm\u002F\">important considerations\u003C\u002Fa> for MMM.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"budgets\">Budgets\u003C\u002Fh2>\n\n\n\n\u003Cp>Provision of budgets is crucial for MMM because it provides essential context and constraints for evaluating the effectiveness of marketing activities. This is particularly useful when performing budget allocation in MMM, as knowledge of budgets will constrain the budget optimiser, enabling you to produce realistic forecasts to produce data-driven insights that you can actually implement.\u003C\u002Fp>\n\n\n\n\u003Cfigure class=\"wp-block-image size-full\">\u003Ca rel=\"noreferrer noopener\" href=\"\u002Fmedia-solutions\u002Fmeasurement\u002F\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"3840\" height=\"540\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Ff0sKYLsk-Measurement_Banner_01.jpg?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-27925\"\u002F>\u003C\u002Fa>\u003C\u002Ffigure>\n",{"rendered":128,"protected":213},28829,{"_yoast_wpseo_focuskw":5,"_yoast_wpseo_title":539,"_yoast_wpseo_metadesc":540,"inline_featured_image":213,"footnotes":5},"Data requirements for Marketing Mix Modelling | Impression","Marketing Mix Modelling is a powerful statistical tool which aims to understand the impact of various marketing activities on sales and performance metrics.",[222,223],[],[226],[545,113,229,230,231,232,233,234,235],"post-28804",[],[],[549,550,551,552],{"path":187,"name":188},{"path":241,"name":242},{"path":156,"name":157},{"path":196,"name":127},{"thumbnail":554,"medium":555,"medium_large":557,"large":559,"1536x1536":561,"2048x2048":563,"post-thumbnail":565,"12_col":567,"default_landscape":569,"default_portrait":571,"6_col_and_margin_short":573,"11_col_and_margin":575,"9_col_rectangle":577,"11_col":579,"staff_image_696_600":581,"rectangular_thumbnail":583},{"url":132,"width":247,"height":247},{"url":556,"width":70,"height":70},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpath-digital-tR0jvlsmCuQ-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=150&q=90&w=200",{"url":558,"width":73,"height":23},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpath-digital-tR0jvlsmCuQ-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=576&q=90&w=768",{"url":560,"width":254,"height":255},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpath-digital-tR0jvlsmCuQ-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=1000&q=90&w=1892",{"url":562,"width":79,"height":79},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpath-digital-tR0jvlsmCuQ-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=1152&q=90&w=1536",{"url":564,"width":82,"height":82},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpath-digital-tR0jvlsmCuQ-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=1536&q=90&w=2048",{"url":566,"width":85,"height":86},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpath-digital-tR0jvlsmCuQ-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=1180&q=90&w=1860",{"url":568,"width":88,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpath-digital-tR0jvlsmCuQ-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=2880",{"url":570,"width":91,"height":65},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpath-digital-tR0jvlsmCuQ-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=600&q=90&w=800",{"url":572,"width":65,"height":91},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpath-digital-tR0jvlsmCuQ-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=800&q=90&w=600",{"url":574,"width":85,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpath-digital-tR0jvlsmCuQ-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=1860",{"url":576,"width":95,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpath-digital-tR0jvlsmCuQ-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=3080",{"url":578,"width":97,"height":98},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpath-digital-tR0jvlsmCuQ-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=1400&q=90&w=2150",{"url":580,"width":100,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpath-digital-tR0jvlsmCuQ-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=2640",{"url":582,"width":64,"height":65},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpath-digital-tR0jvlsmCuQ-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=600&q=90&w=696",{"url":584,"width":103,"height":104},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpath-digital-tR0jvlsmCuQ-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=300&q=90&w=750",[586],{"id":215,"path":282,"display_name":4,"job_title":11,"description":6,"avatar_url":122,"acf":587},{"full_bio":10,"job_title":11,"specialist_service_links":588,"photo":595},[589,591,593],{"page":590,"link_text":29},{"ID":15,"post_author":16,"post_date":17,"post_date_gmt":17,"post_content":5,"post_title":18,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":21,"to_ping":5,"pinged":5,"post_modified":22,"post_modified_gmt":22,"post_content_filtered":5,"post_parent":23,"guid":24,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":28},{"page":592,"link_text":34},{"ID":32,"post_author":16,"post_date":33,"post_date_gmt":33,"post_content":5,"post_title":34,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":35,"to_ping":5,"pinged":5,"post_modified":36,"post_modified_gmt":36,"post_content_filtered":5,"post_parent":37,"guid":38,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":39},{"page":594,"link_text":44},{"ID":42,"post_author":16,"post_date":43,"post_date_gmt":43,"post_content":5,"post_title":44,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":45,"to_ping":5,"pinged":5,"post_modified":46,"post_modified_gmt":46,"post_content_filtered":5,"post_parent":37,"guid":47,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":48},{"ID":50,"id":50,"title":51,"filename":52,"filesize":53,"url":54,"link":55,"alt":5,"author":56,"description":5,"caption":5,"name":57,"status":58,"uploaded_to":23,"date":59,"modified":59,"menu_order":23,"mime_type":60,"type":61,"subtype":62,"icon":63,"width":64,"height":65,"sizes":596},{"thumbnail":67,"thumbnail-width":68,"thumbnail-height":68,"medium":69,"medium-width":70,"medium-height":71,"medium_large":72,"medium_large-width":73,"medium_large-height":74,"large":75,"large-width":76,"large-height":77,"1536x1536":78,"1536x1536-width":79,"1536x1536-height":80,"2048x2048":81,"2048x2048-width":82,"2048x2048-height":83,"post-thumbnail":84,"post-thumbnail-width":85,"post-thumbnail-height":86,"12_col":87,"12_col-width":88,"12_col-height":89,"default_landscape":90,"default_landscape-width":91,"default_landscape-height":65,"default_portrait":92,"default_portrait-width":65,"default_portrait-height":91,"6_col_and_margin_short":93,"6_col_and_margin_short-width":85,"6_col_and_margin_short-height":89,"11_col_and_margin":94,"11_col_and_margin-width":95,"11_col_and_margin-height":89,"9_col_rectangle":96,"9_col_rectangle-width":97,"9_col_rectangle-height":98,"11_col":99,"11_col-width":100,"11_col-height":89,"staff_image_696_600":101,"staff_image_696_600-width":64,"staff_image_696_600-height":65,"rectangular_thumbnail":102,"rectangular_thumbnail-width":103,"rectangular_thumbnail-height":104},[598,599],{"id":222,"name":125,"path":124},{"id":223,"name":157,"path":156},[601,602,603,604,606,607,608,609,610,611,612,613,615,617,619,620,621,622,623,624,625,626],{"property":299,"content":539},{"property":301,"content":540},{"name":303,"content":304},{"property":201,"content":605},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fdata-requirements-for-marketing-mix-modelling\u002F",{"name":308,"content":309},{"property":311,"content":312},{"property":314,"content":539},{"property":316,"content":540},{"property":318,"content":605},{"property":320,"content":321},{"property":323,"content":324},{"property":326,"content":614},"2024-05-24T10:39:36+00:00",{"property":329,"content":616},"2024-05-24T10:48:39+00:00",{"property":332,"content":618},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fopen-graph-data-requirements.png?auto=compress%2Cformat&q=90",{"property":335,"content":336},{"property":338,"content":339},{"property":341,"content":60},{"property":182,"content":4},{"property":344,"content":345},{"property":347,"content":618},{"property":349,"content":350},{"property":352,"content":350},{"@context":354,"@graph":628},[629,640,646,648,654,658,676],{"@type":293,"@id":630,"isPartOf":631,"author":632,"headline":127,"datePublished":614,"dateModified":616,"mainEntityOfPage":633,"wordCount":634,"publisher":635,"image":636,"thumbnailUrl":638,"articleSection":639,"inLanguage":368},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fdata-requirements-for-marketing-mix-modelling\u002F#article",{"@id":605},{"name":4,"@id":360},{"@id":605},1535,{"@id":363},{"@id":637},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fdata-requirements-for-marketing-mix-modelling\u002F#primaryimage","https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpath-digital-tR0jvlsmCuQ-unsplash.jpg?auto=compress%2Cformat&q=90",[125,157],{"@type":370,"@id":605,"url":605,"name":539,"isPartOf":641,"primaryImageOfPage":642,"image":643,"thumbnailUrl":638,"datePublished":614,"dateModified":616,"description":540,"breadcrumb":644,"inLanguage":377},{"@id":372},{"@id":637},{"@id":637},{"@id":645},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fdata-requirements-for-marketing-mix-modelling\u002F#breadcrumb",{"@type":379,"inLanguage":368,"@id":637,"url":638,"contentUrl":638,"width":380,"height":647},1800,{"@type":383,"@id":645,"itemListElement":649},[650,651,652,653],{"@type":386,"position":387,"name":188,"item":388},{"@type":386,"position":151,"name":242,"item":390},{"@type":386,"position":163,"name":157,"item":392},{"@type":386,"position":394,"name":127},{"@type":396,"@id":372,"url":388,"name":321,"description":5,"publisher":655,"inLanguage":368,"potentialAction":656},{"@id":398},{"@type":400,"target":657,"query-input":404},{"@type":402,"urlTemplate":403},{"@type":406,"@id":398,"name":321,"url":388,"logo":659,"image":412,"sameAs":660,"@context":417,"email":418,"foundingDate":419,"legalName":420,"vatID":421,"description":422,"telephone":423,"address":661,"location":662,"areaServed":672},{"@type":379,"inLanguage":368,"@id":408,"url":409,"contentUrl":409,"width":410,"height":411,"caption":321},[414,415,416],{"@type":425,"streetAddress":426,"addressLocality":427,"addressRegion":428,"postalCode":429,"addressCountry":430},[663,666,669],{"@type":433,"name":434,"geo":664,"address":665,"photo":412,"telephone":440},{"@type":436,"latitude":437,"longitude":438},{"@type":425,"streetAddress":426,"addressLocality":427,"addressRegion":428,"postalCode":429,"addressCountry":430},{"@type":433,"name":442,"geo":667,"address":668,"photo":451,"telephone":452},{"@type":436,"latitude":444,"longitude":445},{"@type":425,"streetAddress":447,"addressLocality":448,"addressRegion":449,"postalCode":450,"addressCountry":430},{"@type":433,"name":454,"geo":670,"address":671},{"@type":436,"latitude":456,"longitude":457},{"@type":425,"streetAddress":459,"addressLocality":460,"addressRegion":461,"postalCode":462,"addressCountry":463},[673,674,675],{"@type":466,"name":467},{"@type":466,"name":469},{"@type":466,"name":463},{"@type":472,"@id":473,"name":4,"image":677,"description":6,"sameAs":678,"url":202},{"@type":379,"inLanguage":368,"@id":475,"url":8,"contentUrl":8,"caption":4},[198],{"self":680,"collection":685,"about":687,"author":689,"replies":691,"version-history":694,"predecessor-version":697,"wp:featuredmedia":701,"wp:attachment":704,"wp:term":707,"curies":714},[681],{"href":682,"targetHints":683},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F28804",{"allow":684},[483,484,485,486,487],[686],{"href":490},[688],{"href":493},[690],{"embeddable":108,"href":496},[692],{"embeddable":108,"href":693},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcomments?post=28804",[695],{"count":502,"href":696},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F28804\u002Frevisions",[698],{"id":699,"href":700},28834,"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F28804\u002Frevisions\u002F28834",[702],{"embeddable":108,"href":703},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fmedia\u002F28829",[705],{"href":706},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fmedia?parent=28804",[708,710,712],{"taxonomy":516,"embeddable":108,"href":709},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcategories?post=28804",{"taxonomy":519,"embeddable":108,"href":711},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcollections?post=28804",{"taxonomy":182,"embeddable":108,"href":713},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcoauthors?post=28804",[715],{"name":525,"href":526,"templated":108},{"id":717,"date":718,"date_gmt":718,"guid":719,"modified":140,"modified_gmt":140,"slug":721,"status":19,"type":113,"link":139,"title":722,"content":723,"excerpt":725,"author":215,"featured_media":726,"comment_status":20,"ping_status":20,"sticky":213,"template":5,"format":217,"meta":727,"categories":730,"collections":731,"coauthors":732,"class_list":733,"acf":735,"head_links":736,"breadcrumbs":737,"url_path":139,"search_description":729,"featured_image_urls":742,"authors":774,"type_nicename":293,"reading_minutes":141,"category":786,"yoast_title":728,"yoast_meta":789,"yoast_json_ld":816,"_links":868},28273,"2024-05-01T15:30:40",{"rendered":720},"\u002F?p=28273","incrementality-testing",{"rendered":137},{"rendered":724,"protected":213},"\n\u003Cdiv class=\"table-of-contents\">\u003Cul>\u003Cli class=\"heading-h2\">\u003Ca href=\"#what-does-incrementality-mean-in-advertising\">What does ‘incrementality’ mean in advertising?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#what-is-incrementality-testing\">What is incrementality testing?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h3\">\u003Ca href=\"#incrementality-testing-example\">Incrementality testing example\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#how-contribution-based-measurement-compares-to-attribution-based-measurement\">How contribution-based measurement compares to attribution-based measurement\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#triangulation-in-marketing\">Triangulation in marketing\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#when-should-you-consider-incrementality-testing\">When should you consider incrementality testing?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#which-digital-advertising-channels-are-well-suited-to-incrementality-testing\">Which digital advertising channels are well suited to incrementality testing?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#what-if-the-insights-from-an-incrementality-test-dont-align-with-insights-from-multi-touch-attribution\">What if the insights from an incrementality test don’t align with insights from multi-touch attribution?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#how-to-calculate-incrementality-and-lift\">How to calculate incrementality and lift\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#aims-of-incrementality-testing\">Aims of incrementality testing\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#incrementality-test-designs\">Incrementality test designs\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h3\">\u003Ca href=\"#geo-experiments\">Geo experiments\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h3\">\u003Ca href=\"#user-testing\">User testing\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h3\">\u003Ca href=\"#product-incrementality-testing\">Product incrementality testing\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h3\">\u003Ca href=\"#synthetic-controls\">Synthetic controls\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#incrementality-testing-frameworks\">Incrementality testing frameworks\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#deriving-the-control-groups\">Deriving the control group(s)\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h3\">\u003Ca href=\"#a-b-testing\">A\u002FB testing\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h3\">\u003Ca href=\"#weighted-averages\">Weighted averages\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h3\">\u003Ca href=\"#time-series-methods\">Time series methods\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#example-of-an-incrementality-test\">Example of an incrementality test\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h3\">\u003Ca href=\"#background\">Background\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h3\">\u003Ca href=\"#methodology\">Methodology\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h3\">\u003Ca href=\"#pre-treatment\">Pre-Treatment\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h3\">\u003Ca href=\"#during-treatment\">During treatment\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h3\">\u003Ca href=\"#post-treatment\">Post-treatment\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h3\">\u003Ca href=\"#results-and-interpretation\">Results and interpretation\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#common-considerations-of-geo-testing\">Common considerations of geo-testing\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#how-to-interpret-the-results-of-your-incrementality-test\">How to interpret the results of your incrementality test\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#what-if-your-channels-are-not-incremental\">What if your channels are not incremental?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#next-steps-if-your-channels-are-incremental\">Next steps if your channels are incremental?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#how-long-should-an-incrementality-test-last\">How long should an incrementality test last?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#how-to-overcome-challenges-with-incrementality-testing\">How to overcome challenges with incrementality testing\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#hold-out-testing-for-the-synthetic-control\">Hold out testing for the synthetic control\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#how-does-incrementality-testing-complement-marketing-mix-modelling\">How does incrementality testing complement Marketing Mix Modelling?\u003C\u002Fa>\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdiv>\n\n\n\n\u003Chr class=\"wp-block-separator has-alpha-channel-opacity is-style-wide\"\u002F>\n\n\n\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"796\" height=\"396\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fincrementality-graph.png?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-28281\"\u002F>\u003C\u002Ffigure>\u003C\u002Fdiv>\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"what-does-incrementality-mean-in-advertising\">What does ‘incrementality’ mean in advertising?\u003C\u002Fh2>\n\n\n\n\u003Cp>Incrementality refers to the causal improvement in revenue, or the KPI of interest, from an isolated activity. \u003C\u002Fp>\n\n\n\n\u003Cp>Incrementality testing aims to measure the additional sales generated exclusively as a result of the advertising efforts. Incrementality testing provides a more precise understanding of the direct impact of advertising efforts on driving consumer behaviour, allowing you to discover the true effectiveness of campaigns in driving desired outcomes. \u003C\u002Fp>\n\n\n\n\u003Cp>In short, the outcomes from Incrementality testing can go a long way to helping you answer “where should I put my next pound\u002Fdollar of marketing budget to get the best returns”. \u003C\u002Fp>\n\n\n\n\u003Cp>Incrementality is not the same as measurement methods which rely on digital, multi-touch attribution. Digital attribution typically claims credit for sales following any ‘touch-point’ in the customer’s journey, not just the sales which were truly motivated by the advertising exposure. Digital attribution does not measure incrementality, as digital attribution will also cover sales which would have happened anyway. \u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"what-is-incrementality-testing\">What is incrementality testing?\u003C\u002Fh2>\n\n\n\n\u003Cp>Incrementality testing refers to a method used to evaluate the true impact of specific marketing initiatives or channels on desired outcomes, such as sales or conversions, through controlled experimentation. \u003C\u002Fp>\n\n\n\n\u003Cp>Unlike digital, multi-touch attribution, which assigns credit to multiple touchpoints along the customer journey, incrementality testing isolates the causal effect of a marketing intervention by comparing the behaviour of a treatment group against a control group.\u003C\u002Fp>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Cstrong>Treatment Group: \u003C\u002Fstrong>Exposed to advertising, used to measure the causal impact of advertising.\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Control Group: \u003C\u002Fstrong>Not exposed to advertising, serves as a baseline for comparison.\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cp>By measuring the differences in desired outcomes between the treatment and control groups, incrementality testing provides you with valuable insights into the true effectiveness of marketing strategies and helps optimise budget allocation for maximum ROI.\u003C\u002Fp>\n\n\n\n\u003Cp>Incrementality testing can do this by starting, stopping, increasing or decreasing investment into a particular marketing activity for a measurable treatment group. For incremental marketing activities, even when the experiment is controlled for externalities, the total sales values in the treatment group should be affected by the change in marketing investments there. \u003C\u002Fp>\n\n\n\n\u003Cp>If a change isn’t detectable, then the marketing investment is offering no incremental value. However, this doesn’t necessarily mean the channel or investment type is entirely ineffective – just the specific execution measured in the test.  \u003C\u002Fp>\n\n\n\n\u003Ch3 class=\"wp-block-heading\" id=\"incrementality-testing-example\">\u003Cstrong>Incrementality testing example\u003C\u002Fstrong>\u003C\u002Fh3>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Cstrong>Step 1: \u003C\u002Fstrong>Select a market or region in which you’d like to examine the effectiveness of advertising. Call this the treatment group. \u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Cstrong>Step 2: \u003C\u002Fstrong>Select a market or region, or derive an artificial market or region, which serves as a baseline sales for comparison. Call this the control group.\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Step 3: \u003C\u002Fstrong>Run ads in the treatment group only, ensuring that everything in the control group remains as constant as possible. \u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Step 4: \u003C\u002Fstrong>Once the test has finished, analyse the results and calculate the incremental uplift and its associated ROI.\u003Cbr>\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"how-contribution-based-measurement-compares-to-attribution-based-measurement\">How contribution-based measurement compares to attribution-based measurement\u003C\u002Fh2>\n\n\n\n\u003Cp>Contributed-based measurement refers to the impact made by advertising on the top line of the business’ results — on its revenue. Attribution-based measurement could measure plenty of sales, but unless these sales are net new to the business, they would make no additional contribution. \u003C\u002Fp>\n\n\n\n\u003Cp>Understanding these differences in measurement is critical, as these can sometimes be confused. Incrementality testing (as well as marketing mix modelling for that matter) seeks to measure incremental improvements in contribution from marketing, for the business. \u003C\u002Fp>\n\n\n\n\u003Cp>A contribution-based approach to measurement uses historical data, utilises statistical techniques and doesn’t require knowledge of the customer journey. An attribution-based approach tracks touchpoints in the customer journey from the point of viewing an ad and measures every sale associated with ad exposure. \u003C\u002Fp>\n\n\n\n\u003Cp>To truly understand the impact of your marketing you need to understand the differences in the measurement techniques available to you, and the pros and cons of each method. \u003C\u002Fp>\n\n\n\n\u003Cp>An attribution-based approach to measurement has two main flaws: \u003C\u002Fp>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Cstrong>Generosity\u003C\u002Fstrong>: The first flaw is that it tends to measure marketing performance through a slightly generous lens. As this method measures all sales associated with ad exposure, it will also record those sales which would’ve happened anyway regardless of the user interaction with the ad. Additionally, if the user were to view the ad twice, using two different devices, an attribution-based approach would count this as two different user interactions, when in fact it was the same user. As a result of this, you might find that an attribution-based approach to measurement reports generously. \u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Accuracy\u003C\u002Fstrong>: The second flaw is that, with the depreciation of the third-party cookie, tracking touch points within the customer journey is becoming more difficult and less accurate. \u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cp>Although an attribution-based approach to measurement has its flaws, it does provide user and campaign-level insights. User and campaign-level insights offer a granular understanding of your audience’s behaviour and the performance of individual marketing campaigns. By analysing user-level data, you can gain insights into customer preferences, behaviours, and purchase patterns, allowing for more personalised and targeted marketing efforts. Campaign-level insights also provide visibility into the effectiveness of specific marketing initiatives, helping you to identify which campaigns are driving the most engagement, conversions, and ROI.\u003C\u002Fp>\n\n\n\n\u003Cp>Incrementality testing on the other hand adopts a contribution-based approach to measurement and avoids these flaws. The control group of the incrementality test is designed to act as a proxy for what would’ve happened to the treatment group in absence of the marketing effect. \u003C\u002Fp>\n\n\n\n\u003Cp>As an incrementality test measures the difference between the treatment and control group, the sales which would’ve happened anyway are accounted for by the control group. Furthermore, incrementality testing doesn’t require knowledge of the customer journey because it measures the total revenue impact to the business, not web analytics software, and hence is not affected by the depreciation of the third-party cookie.\u003C\u002Fp>\n\n\n\n\u003Cp>Although incrementality testing might seem like the solution to attribution-based measurement, an incrementality test is only as good as the control group used to mimic the treatment group in absence of the treatment.\u003C\u002Fp>\n\n\n\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Ca rel=\"noreferrer noopener\" href=\"\u002Fmedia-solutions\u002Fmeasurement\u002F\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"3840\" height=\"540\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Ff0sKYLsk-Measurement_Banner_01.jpg?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-27925\"\u002F>\u003C\u002Fa>\u003C\u002Ffigure>\u003C\u002Fdiv>\n\n\n\u003Cdiv class=\"wp-block-columns is-layout-flex wp-container-core-columns-is-layout-9d6595d7 wp-block-columns-is-layout-flex\">\n\u003Cdiv class=\"wp-block-column is-layout-flow wp-block-column-is-layout-flow\" style=\"flex-basis:66.66%\">\n\u003Ch2 class=\"wp-block-heading\" id=\"triangulation-in-marketing\">Triangulation in marketing\u003C\u002Fh2>\n\n\n\n\u003Cp>Although both contribution-based and attribution-based approaches aim to measure the performance of your marketing, they do so in different ways. This makes comparison of the two methods impractical. Instead of comparing these methods, you may gain more insights into your marketing performance, or at least be able to remain agile in your media buying, by combining them.\u003C\u002Fp>\n\u003C\u002Fdiv>\n\n\n\n\u003Cdiv class=\"wp-block-column is-vertically-aligned-center is-layout-flow wp-block-column-is-layout-flow\" style=\"flex-basis:33.33%\">\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"896\" height=\"794\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fmta-testing-modelling.png?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-28278\"\u002F>\u003C\u002Ffigure>\u003C\u002Fdiv>\u003C\u002Fdiv>\n\u003C\u002Fdiv>\n\n\n\n\u003Cp>\u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fmodern-measurement-series-3-the-trio-of-attribution-and-contribution\u002F\">Triangulation in marketing\u003C\u002Fa> simply refers to the process of gaining insights into marketing performance from the three most popular techniques used in media effectiveness measurement. The aim of triangulation is to extract insights from all sources and combine them to estimate the “true” performance of your marketing. The three most popular sources in marketing measurement are multi-touch attribution (MTA), \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fmedia-mix-modelling\u002F\">marketing mix modelling\u003C\u002Fa> (MMM) and incrementality testing, and each are important as they offer differing levels of insights which are useful at different time intervals. \u003C\u002Fp>\n\n\n\n\u003Cp>Multi-touch attribution (MTA) focuses on allocating credit to various touch points along the customer journey to understand how each touch point attributes to sales or conversions, and often reports on this in near-real-time. MMM, on the other hand, uses statistical modelling to analyse historical data and quantify the impact of different marketing channels on business outcomes. MMM is often run periodically, on refresh intervals which can range between weeks and months. Incrementality testing involves conducting experiments where a subset of the target audience is exposed to a marketing intervention, while another subset serves as a control group. Incrementality testing, and testing in general should really be an ongoing programme of work in a business, but each campaign or platform might only be tested periodically also. \u003C\u002Fp>\n\n\n\n\u003Cp>Triangulation is really a new industry term, which acknowledges these three different levels of insight. From MTA you’ll be able to extract campaign and user level insights, from MMM you’ll be able to extract channel and source level insights and from incrementality testing you’ll be able to extract causal impact estimates and activity level insights. Triangulation aims to cancel out any limitations of all three methods by taking advantage of their individual strengths, allowing you to extract the best insights from all three methods.\u003C\u002Fp>\n\n\n\n\u003Cp>In practice, with measurement triangulation, approximations will need to be manually elicited from the insights, however, one of the benefits in this approach for teams is that it can support a gradual movement towards contribution-based measurement reporting, without wholesale moves away from attribution-modelled reporting.  \u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"when-should-you-consider-incrementality-testing\">When should you consider incrementality testing?\u003C\u002Fh2>\n\n\n\n\u003Cp>If you’re considering a new marketing channel, or changing your marketing mix, then incrementality testing is crucial to understand the impacts. You can use incrementality testing to determine the causal relationship between marketing activities and desired outcomes, such as sales, conversions, or brand awareness. \u003C\u002Fp>\n\n\n\n\u003Cp>This testing methodology is essential when you need reliable and unbiased insights to optimise marketing budgets, allocate resources effectively, and maximise the return on investment from your marketing spend. \u003C\u002Fp>\n\n\n\n\u003Cp>In addition to incrementality testing, businesses with complex and multi-channel marketing strategies are better suited to \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fmedia-solutions\u002Fmeasurement\u002F\">Marketing Mix Modeling (MMM)\u003C\u002Fa>. MMM is also more effective for businesses with longer sales cycles, where there are multiple touchpoints across various channels influencing customer behaviour. \u003C\u002Fp>\n\n\n\n\u003Cp>On the other hand, MTA can be more suitable for earlier-stage businesses, those with shorter sales cycles, or those particularly those heavily reliant on digital channels and online advertising platforms for new customer acquisition and growth. These businesses can benefit from the granular insights provided by MTA, which allows them to optimise digital advertising campaigns in real time and measure the immediate impact of their marketing efforts.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"which-digital-advertising-channels-are-well-suited-to-incrementality-testing\">Which digital advertising channels are well suited to incrementality testing?\u003C\u002Fh2>\n\n\n\n\u003Cp>While all digital advertising channels can benefit from incrementality testing, some are particularly well-suited due to their inherent characteristics, and others are popular targets due to the potential to see conflicting digitally attributed performance reported against them. \u003C\u002Fp>\n\n\n\n\u003Cp>For most marketers, coming at this with experience with digital attribution, we see broadly two challenges: \u003C\u002Fp>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Cstrong>Short-term\u003C\u002Fstrong> – Advertising designed to to capture customers at the point of need, or advertise a sale promotion, or just generally drive short-term results is over-attributed in MTA reporting. Channels such as Microsoft or Google paid search and paid shopping ads (PLAs) are good examples here.\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cp>Tests on this type of activity are useful to demonstrate the true incrementality to the business’ top line revenue numbers, rather than self-serving platform reporting metrics. Testing for the impact of paid search brand bidding is a good example here, among others.\u003C\u002Fp>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Cstrong>Long-term\u003C\u002Fstrong> – Advertising designed to build mental availability and brand awareness for customers ahead of their point of need, often called “above the line”, is typically under-attributed in digital attribution. Channels such as TV, Radio, out-of-home, and some digital social (Facebook, Instagram, TikTok) and video (YouTube) advertising are good examples here.\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cp>Tests on this type of activity are useful to demonstrate that despite the under-reporting in digital attribution, these channels can contribute over the short and medium term.\u003C\u002Fp>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"what-if-the-insights-from-an-incrementality-test-dont-align-with-insights-from-multi-touch-attribution\">What if the insights from an incrementality test don’t align with insights from multi-touch attribution?\u003C\u002Fh2>\n\n\n\n\u003Cp>Differences in the measurement reporting from multi-touch attribution, media mix modelling and incrementality testing are to be expected. Incrementality testing and digital attribution measure similar things but in different ways. MTA will measure all sales associated with ad exposure, whereas incrementality testing aims to measure only the additional causal sales as a result of advertising. You may often find, but not always, that incrementality testing reports lower ROI estimates than MTA. This is because MTA will generally attribute more sales to an ad than what testing will. \u003Cs> \u003C\u002Fs> The Return On Ad Spend (ROAS) reported as a result of incrementality testing is often referred to as the \u003Cem>incremental\u003C\u002Fem> Return On Ad Spend (iROAS).\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"how-to-calculate-incrementality-and-lift\">How to calculate incrementality and lift\u003C\u002Fh2>\n\n\n\n\u003Cfigure class=\"wp-block-table\">\u003Ctable>\u003Cthead>\u003Ctr>\u003Cth class=\"has-text-align-center\" data-align=\"center\">\u003Cstrong>Key\u003C\u002Fstrong>\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd class=\"has-text-align-center\" data-align=\"center\">Σ = total sum\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"has-text-align-center\" data-align=\"center\">| . | = absolute value\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Ffigure>\n\n\n\n\u003Cfigure class=\"wp-block-image size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"968\" height=\"282\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002FIT-causal-impact.png?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-28288\"\u002F>\u003C\u002Ffigure>\n\n\n\n\u003Cp>In the context of testing, lift is equivalent to the positive difference between treatment sales and baseline sales. Baseline sales are simply an approximation to what sales would’ve been in absence of the treatment or marketing intervention. If the objective of your test is to observe an increase in sales, this may often be referred to as “lift” or “uplift”.\u003C\u002Fp>\n\n\n\n\u003Cp class=\"has-text-align-center has-medium-font-size\">\u003Cstrong>Lift =|Treatment sales – Baseline sales|\u003C\u002Fstrong>\u003C\u002Fp>\n\n\n\n\u003Cp>Incrementality refers to the incremental lift that advertising spend provides to overall sales or conversions. Incrementality provides the percentage of sales or conversions received as a direct result of an advertising campaign. Incremental lift analysis looks specifically at whether an individual advertising campaign was effective and how effective, by determining what sales revenue would have been without the ad campaign.\u003C\u002Fp>\n\n\n\n\u003Cp class=\"has-text-align-center has-medium-font-size\">\u003Cstrong>Incrementality = Causal Impact \u002F Σ\u003C\u002Fstrong> \u003Cstrong>Sales = Σ\u003C\u002Fstrong> \u003Cstrong>Lift \u002F Σ\u003C\u002Fstrong> \u003Cstrong>Sales\u003C\u002Fstrong>\u003C\u002Fp>\n\n\n\n\u003Cp>Overall, lift, causal impact and incrementality are similar concepts with different meanings. Lift tests and incrementality tests are often used interchangeably, the reason for this is that both tests offer the same thing. A lift test will provide both lift and incrementality estimates and similarly an incrementality test will provide both lift and incrementality estimates. The three key questions you should be asking before conducting yourself, or hiring a vendor to conduct a lift test are as follows: \u003C\u002Fp>\n\n\n\n\u003Col class=\"wp-block-list\">\n\u003Cli>\u003Cem>Can you estimate the total lift?\u003C\u002Fem>\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cem>Can you estimate the incrementality?\u003C\u002Fem>\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cem>How do you estimate base sales?\u003C\u002Fem>\u003C\u002Fli>\n\u003C\u002Fol>\n\n\n\n\u003Cp>The estimation of base sales is the key characteristic of an incrementality test. A poor approximation to base sales will yield unrealistic results. As both lift and incrementality are dependent upon this approximation, it’s essential to get it right!\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"aims-of-incrementality-testing\">Aims of incrementality testing\u003C\u002Fh2>\n\n\n\n\u003Cp>In short, incrementality testing aims to estimate the true performance of your marketing efforts, which in turn allows you to increase revenue through the best long term allocation of budgets and resources. These aims become particularly evident when testing the performance of newer advertising channels, or if as a brand you are moving from a performance focus to a \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fmodern-measurement-series-4-demonstrating-the-value-of-brand-media-vs-performance-media\u002F\">balanced brand and performance split\u003C\u002Fa>. \u003C\u002Fp>\n\n\n\n\u003Cp>Without a solid testing methodology, marketers might roll out a new channel across all advertising regions in hopes that the channel will perform well. Usually the budget required to do this is relatively large, and if a solid measurement plan isn’t in place, or multi-touch attribution is used to measure success and this fails to capture it, then the conclusion may be that the spend was ineffective. There are absolutely risks of this opinion being formed when activating on new or brand-based advertising channels or campaigns, as returns may not be immediately obvious. . \u003C\u002Fp>\n\n\n\n\u003Cp>By conducting an incrementality test, in which only one advertising region is exposed to the new channel, you can gain insights into the performance of this channel before rolling out the channel across all geographies. \u003C\u002Fp>\n\n\n\n\u003Cp>If the new channel, creative or campaign doesn’t perform well in one region, it’d be sensible to assume it won’t across more regions and hence it’d be time to re-evaluate the advertising strategy for this channel. Alternatively, if the new channel does perform well, you’d be rolling out this channel with much more confidence than you’d have if you didn’t conduct a test.\u003C\u002Fp>\n\n\n\n\u003Cp>Alternatively you may have a hypothesis that your Google Ads paid search budget for pure brand terms is too high and that your net sales won’t be affected by reducing budget. As opposed to just reducing budget across all advertising regions and hoping sales don’t take a drastic hit, you could conduct an incrementality test to prove or disprove this hypothesis. By conducting an incrementality test, in which only one advertising region is exposed to a reduced Google Ads paid search budget, you can gain insights into how this affects sales in one region before rolling out the channel across all regions. \u003C\u002Fp>\n\n\n\n\u003Cp>If reducing budget doesn’t significantly affect sales, then you’d have evidence to suggest this won’t significantly affect sales in other regions. Alternatively, if sales are significantly affected in one region, then you’d have evidence to suggest reducing budget in all regions will significantly affect sales. If this is the case, the amount of sales you’d have lost by reducing budget in all advertising regions would significantly outweigh the sales lost by reducing budget in one region first. \u003C\u002Fp>\n\n\n\n\u003Cp>In essence, incrementality testing aims to reduce the reliance on guesswork when implementing changes to your marketing strategies. Additionally, if your business advertises using multiple channels, you may consider \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fmedia-mix-modelling\u002F\">Marketing mix modelling\u003C\u002Fa>. If you do, incrementality testing helps to inform MMM and in turn aims to reduce any uncertainty associated with the outputs of an MMM.\u003C\u002Fp>\n\n\n\n\u003Cdiv class=\"wp-block-columns is-layout-flex wp-container-core-columns-is-layout-9d6595d7 wp-block-columns-is-layout-flex\">\n\u003Cdiv class=\"wp-block-column is-layout-flow wp-block-column-is-layout-flow\">\n\u003Ch2 class=\"wp-block-heading\" id=\"incrementality-test-designs\">Incrementality test designs\u003C\u002Fh2>\n\n\n\n\u003Cp>The test design simply refers to how you choose to separate your treatment and control groups. There are many different ways to implement incrementality testing, from the test design to the statistical techniques used, there is no universally accepted way of conducting an incrementality test. Each type of test will have its own strengths and limitations associated with it. In this section, we’ll discuss various different types of test designs and consider their strengths and weaknesses.\u003C\u002Fp>\n\u003C\u002Fdiv>\n\n\n\n\u003Cdiv class=\"wp-block-column is-vertically-aligned-center is-layout-flow wp-block-column-is-layout-flow\">\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"1920\" height=\"1440\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fincremental-geo-test.jpg?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-28291\"\u002F>\u003C\u002Ffigure>\u003C\u002Fdiv>\u003C\u002Fdiv>\n\u003C\u002Fdiv>\n\n\n\n\u003Ch3 class=\"wp-block-heading\" id=\"geo-experiments\">Geo experiments\u003C\u002Fh3>\n\n\n\n\u003Cp>One of the most frequently used forms of test design is geo-experiments. \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fguide-to-media-effectiveness-measurement\u002F\">A geo-experiment, within the realm of incrementality testing\u003C\u002Fa>, involves conducting controlled experiments within specific geographical regions to measure the incremental impact of marketing initiatives. In a geo-experiment, different geographic areas are selected as treatment and control groups, where the treatment group is exposed to the marketing intervention while the control groups remain unaffected. By comparing the outcomes between these groups, you can estimate the true causal effect of your marketing efforts within each geographic region. \u003C\u002Fp>\n\n\n\n\u003Cp>One of the key strengths of geo-experiments is their ability to provide localised insights, allowing you to understand regional variations in consumer behaviour and tailor your marketing strategies accordingly. Geo-experiments are relatively cost-effective and less prone to external confounding factors compared to broader market-wide experiments. \u003C\u002Fp>\n\n\n\n\u003Cp>However, a major weakness of geo-experiments is the potential for spill-over effects, where marketing activities targeted at one region inadvertently impact neighbouring regions, leading to biased results. Additionally, geo-experiments may not always capture the full extent of complex consumer behaviours, particularly in markets with significant cross-regional interactions or where digital footprints transcend geographical boundaries.\u003C\u002Fp>\n\n\n\n\u003Ch3 class=\"wp-block-heading\" id=\"user-testing\">User testing\u003C\u002Fh3>\n\n\n\n\u003Cp>Another form of test design is by user. User-based incrementality tests involve dividing users into randomised treatment and control groups to measure the incremental impact of marketing efforts on individual users. By comparing the outcomes between these groups, you can estimate the causal effect of your marketing activities on user behaviour. \u003C\u002Fp>\n\n\n\n\u003Cp>One of the key strengths of user-based incrementality tests is their ability to provide insights into the impact of marketing on individual users, allowing for personalised targeting and messaging strategies. Additionally, \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fdigital-experience\u002Fuser-research\u002F\">user-based tests\u003C\u002Fa> are less susceptible to external factors that may influence regional or market-wide experiments, providing more accurate and reliable results. \u003C\u002Fp>\n\n\n\n\u003Cp>However, a major drawback of user-based tests is the potential for selection bias, where certain types of users may be more likely to respond to marketing interventions, leading to skewed results. Moreover, user-based tests may require larger sample sizes to detect statistically significant differences in user behaviour, especially for rare outcomes or niche market segments.\u003C\u002Fp>\n\n\n\n\u003Cp>Ad platforms with large identity graphs can offer such tests, as can organisations which have large audience data pools, however, geography-based experiments are more often chosen for testing media channels in the context of new customer acquisition as often third-party measurement vendors and platforms cannot access these user attributes. \u003C\u002Fp>\n\n\n\n\u003Ch3 class=\"wp-block-heading\" id=\"product-incrementality-testing\">Product incrementality testing\u003C\u002Fh3>\n\n\n\n\u003Cp>The last form of test design we’re going to talk about is product-based. Product-based incrementality tests involve dividing products into treatment and control groups to assess the incremental impact of marketing efforts on product sales. By comparing sales outcomes between these product groups, you can estimate the causal effect of marketing activities on product performance. One of the main strengths of product-based incrementality tests is their ability to evaluate the impact of marketing campaigns on specific products or product categories, enabling you to tailor your marketing strategies accordingly. \u003C\u002Fp>\n\n\n\n\u003Cp>Additionally, product-based tests provide insights into the differential effects of marketing across different product lines, helping you allocate resources effectively. However, a potential flaw of product-based tests is the need for sufficient product volume and diversity to generate meaningful results. In cases where certain products have low sales volume or limited variability, detecting significant differences between treatment and control groups may be challenging. Moreover, product-based tests may not account for cross-product effects or interactions, which could affect the interpretation of results.\u003C\u002Fp>\n\n\n\n\u003Ch3 class=\"wp-block-heading\" id=\"synthetic-controls\">Synthetic controls\u003C\u002Fh3>\n\n\n\n\u003Cp>Although \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fwhat-is-synthetic-data\u002F\">synthetic\u003C\u002Fa> controls are not unique or limited to any one of the above types of experiments, their consideration in the design of a test is crucial depending on the next best alternative for providing a control to an experiment. \u003C\u002Fp>\n\n\n\n\u003Cp>Synthetic data is simply machine-generated data which is designed to simulate real-world observed data. In the context of providing a test control, this is often a modelled expectation based on large volumes of historical data, which captures a number of variables including seasonality. \u003C\u002Fp>\n\n\n\n\u003Cp>Synthetic controls are most often found in geo experiments. As well as being a more popular and accessible media test approach, geo experiments can also fall foul of issues around control geography selection, making the case for synthetic controls stronger. \u003C\u002Fp>\n\n\n\n\u003Cp>In geo experiments, it’s our approach, and a standard industry one to model anticipated sales inside the treatment geography prior to exposure, and then to compare actual treatment results versus the estimated control. Of course, this approach creates additional considerations, many of which are controlled for: \u003C\u002Fp>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Cstrong>Externalities\u003C\u002Fstrong> occurring mid-test, for which the modelled data would not have previously accounted for. Synthetic controls consider a number of variables and are generated from numerous geographies, therefore they can take into account the effect of expected externalities at the point of generation. However, to capture ongoing real-world externalities, models can be refreshed as new data becomes available so that large impacts in the data which are not explainable by the experiment aren’t accidentally attributed to it. \u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Accuracy\u003C\u002Fstrong> of modelled data, as the control is synthetic. Back-testing is a common solution to evidencing the efficacy of a synthetic control – where some actual real-world observations are held back at the point of model generation, and then compared to the model prior to its use. For the observations which fall in this backtested period, the model can show its level of ‘fit’ and predictive accuracy. A common statistic used to quantify goodness of fit,  which we implement, is the \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fwww.statisticshowto.com\u002Fprobability-and-statistics\u002Fregression-analysis\u002Frmse-root-mean-square-error\u002F\">root-mean-square-error (RMSE)\u003C\u002Fa>. The smaller the RMSE the better the predictive accuracy of the synthetic control.\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cp>The alternative approach to synthetic controls is called “matched market” testing, whereby similar geographies are selected each as treatment and control and these are evaluated against one another. Challenges around audience leeching between markets are just one reason this is less favourable — as such challenges are difficult to mitigate. Using a synthetic control accounts for this to a certain degree, but care should still be taken.\u003C\u002Fp>\n\n\n\n\u003Cdiv class=\"wp-block-columns is-layout-flex wp-container-core-columns-is-layout-9d6595d7 wp-block-columns-is-layout-flex\">\n\u003Cdiv class=\"wp-block-column is-layout-flow wp-block-column-is-layout-flow\">\n\u003Ch2 class=\"wp-block-heading\" id=\"incrementality-testing-frameworks\">Incrementality testing frameworks\u003C\u002Fh2>\n\n\n\n\u003Cp>\u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fguide-to-media-effectiveness-measurement\u002F\">Frequentist and Bayesian incrementality tests\u003C\u002Fa> represent two distinct statistical approaches to assess the causal impact of marketing interventions.\u003C\u002Fp>\n\u003C\u002Fdiv>\n\n\n\n\u003Cdiv class=\"wp-block-column is-vertically-aligned-center is-layout-flow wp-block-column-is-layout-flow\">\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"640\" height=\"480\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F11\u002Fbayesian-plot.png?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-25571\"\u002F>\u003C\u002Ffigure>\u003C\u002Fdiv>\u003C\u002Fdiv>\n\u003C\u002Fdiv>\n\n\n\n\u003Cp>In a frequentist approach, the focus is on estimating parameters of interest based on observed data, without incorporating prior beliefs or uncertainties. Frequentist tests typically rely on hypothesis testing and confidence intervals to make inferences about the true effect of marketing activities. For example, in a frequentist test, the null hypothesis might assume that there is no difference in outcomes between the treatment and control groups. Statistical significance is then determined by a measure, known as the p-value, that helps determine whether the results of this test are likely to have occurred by random chance or by a direct result of the treatment. One major benefit to testing under a frequentist framework is that frequentist tests provide straightforward interpretations and are often perceived as more objective, as unlike a Bayesian test, they do not involve subjective priors. However, one major drawback to incrementality testing under a frequentist framework is that frequentist tests may struggle with complex models or small sample sizes. \u003C\u002Fp>\n\n\n\n\u003Cp>Alternatively, \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fbayesian-statistics-in-media-mix-modelling-a-real-world-analogy\u002F\">Bayesian\u003C\u002Fa> incrementality tests explicitly incorporate prior beliefs or information about parameters of interest, allowing for the integration of prior knowledge with observed data to make inferences. Bayesian methods provide posterior distributions (a combination of the prior distribution and the data) that represent the updated beliefs about parameters after observing data, enabling a more nuanced understanding of uncertainty and variability. \u003C\u002Fp>\n\n\n\n\u003Cp>Incrementality testing under the Bayesian framework tackles the lack of flexibility as seen in frequentist tests. Bayesian methods can handle complex models, small sample sizes, and hierarchical structures more effectively than frequentist approaches because of this ability to incorporate prior understanding. \u003C\u002Fp>\n\n\n\n\u003Cp>We sometimes explain priors as “guardrails” in parameter estimation. The wider we make these “guardrails” the more we allow the media data to influence estimation. For an advertising channel with incrementality test data, it’s sometimes sensible to make these “guardrails” narrower for an appropriate estimate based on causal results.  However, one major drawback of the Bayesian approach to incrementality testing is that it can be very complex and computationally expensive to conduct compared to the frequentist alternative.\u003C\u002Fp>\n\n\n\n\u003Cp>In summary, frequentist and Bayesian incrementality tests offer distinct advantages and limitations, and the choice between them depends on the specific requirements of the analysis, the available data, and the researcher’s preferences regarding uncertainty and prior knowledge incorporation.\u003C\u002Fp>\n\n\n\n\u003Cp>It’s our preference to utilise a \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fbayesian-statistics-in-media-mix-modelling-a-real-world-analogy\u002F\">Bayesian\u003C\u002Fa> approach to media effectiveness measurement at Impression. We take forward learnings from incrementality testing into our marketing mix models, which the flexibility of Bayesian statistics allows us to do.\u003C\u002Fp>\n\n\n\n\u003Cdiv class=\"wp-block-columns is-layout-flex wp-container-core-columns-is-layout-9d6595d7 wp-block-columns-is-layout-flex\">\n\u003Cdiv class=\"wp-block-column is-layout-flow wp-block-column-is-layout-flow\">\n\u003Ch2 class=\"wp-block-heading\" id=\"deriving-the-control-groups\">Deriving the control group(s)\u003C\u002Fh2>\n\n\n\n\u003Cp>In incrementality testing, the derivation of the control group is a key characteristic. When deriving the control group, you could take an A\u002FB test approach, use weighted averages or conduct some time series analysis.\u003C\u002Fp>\n\u003C\u002Fdiv>\n\n\n\n\u003Cdiv class=\"wp-block-column is-vertically-aligned-center is-layout-flow wp-block-column-is-layout-flow\">\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"731\" height=\"491\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Frevenue-by-location.png?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-28301\"\u002F>\u003C\u002Ffigure>\u003C\u002Fdiv>\u003C\u002Fdiv>\n\u003C\u002Fdiv>\n\n\n\n\u003Ch3 class=\"wp-block-heading\" id=\"a-b-testing\">A\u002FB testing\u003C\u002Fh3>\n\n\n\n\u003Cp>An \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fdigital-experience\u002Fcro\u002Fab-testing\u002F\">A\u002FB test\u003C\u002Fa> requires a comparison of a treatment group (A) with a control group (B). The key difference between an A\u002FB test and the following alternatives is that the control group isn’t artificially designed. In the context of a geo-based A\u002FB test, the region exposed to the marketing intervention will be compared with a different region, which wasn’t exposed. Therefore the control group in A\u002FB testing is simply another group which hasn’t been specifically designed to mimic the treatment group.\u003C\u002Fp>\n\n\n\n\u003Cp>The major drawback with this approach, in the context of a geo-experiment, is that you won’t be able to separate the percentage of lift generated by the treatment from that generated as a result of regional differences. Choosing a region which behaves most similarly to the treatment region is one way to control for this, however, the influence of regional differences simply can’t be quantified in a geo-based A\u002FB test.\u003C\u002Fp>\n\n\n\n\u003Ch3 class=\"wp-block-heading\" id=\"weighted-averages\">Weighted averages\u003C\u002Fh3>\n\n\n\n\u003Cp>Weighted averages offer an alternative, in which the control group is artificially designed to mimic the treatment group in absence of the treatment or marketing intervention. To conduct a test using weighted averages, you require several control groups. A weighted average of these control groups is then taken and the weights are chosen such that this average best mimics the treatment group during the pre-treatment phase. Post-treatment, you must continue to observe the control groups in order to generate new forecasts of what the treatment group would’ve been in the absence of the treatment. One drawback to this approach is that, in some cases, you may find there does not exist weights such that the weighted average of the given control groups mimics the treatment group very well. If this is the case, it is recommended that you turn to an A\u002FB test approach or implement time series methods.\u003C\u002Fp>\n\n\n\n\u003Ch3 class=\"wp-block-heading\" id=\"time-series-methods\">Time series methods\u003C\u002Fh3>\n\n\n\n\u003Cp>Similarly to weighted averages, time series methods generate a synthetic control group. However, time series methods don’t necessarily require several additional control groups. Time series methods involve modelling the treatment group during the pre-treatment phase and using this model to forecast what the treatment group would’ve been in absence of the treatment. There are various ways to model time series data such as \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fwww.ibm.com\u002Ftopics\u002Flinear-regression#:~:text=the%20next%20step-,What%20is%20linear%20regression%3F,is%20called%20the%20independent%20variable.\">linear regression\u003C\u002Fa> and \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fmachinelearningmastery.com\u002Farima-for-time-series-forecasting-with-python\u002F\">ARIMA\u003C\u002Fa>. The main benefit of using this model over weighted averages is that this method often requires much less data. However, this method is often harder to implement as it forces you to account for trends and seasonality and other externalities, which without experience in statistical modelling can be quite difficult. \u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"example-of-an-incrementality-test\">Example of an incrementality test\u003C\u002Fh2>\n\n\n\n\u003Ch3 class=\"wp-block-heading\" id=\"background\">Background\u003C\u002Fh3>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>A global company specialising in garden appliances, called “Under the Hedge”, is considering investing in advertising, direct to consumer for the first time, on Meta in the US. \u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Ch3 class=\"wp-block-heading\" id=\"methodology\">Methodology\u003C\u002Fh3>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>To inform this decision, Under the Hedge decided to conduct a geo-based incrementality test, using a synthetic control group, to estimate the incremental impact of advertising on Meta.\u003C\u002Fli>\n\n\n\n\u003Cli>Treatment group – A region exposed to Meta advertising. Under the Hedge decided New York was to be the treatment group.\u003C\u002Fli>\n\n\n\n\u003Cli>Control groups – Region(s) not exposed to Meta advertising, Under the Hedge decided to use other mainland US states within reasonable proximity to New York, such that variance in consumer behaviour across these states is assumed minimal.\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Ch3 class=\"wp-block-heading\" id=\"pre-treatment\">Pre-Treatment\u003C\u002Fh3>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>3 years of historical revenue data for the relevant US states, where possible, is used for analysis. A weighted average of all other control regions is then taken to model the revenue generated in New York. The weights are chosen such that New York revenue is best mimicked.\u003C\u002Fli>\n\n\n\n\u003Cli>Let us propose the following weights are optimal for mimicking New York revenue.\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cfigure class=\"wp-block-table\">\u003Ctable>\u003Cthead>\u003Ctr>\u003Cth class=\"has-text-align-center\" data-align=\"center\">\u003Cstrong>Region\u003C\u002Fstrong>\u003C\u002Fth>\u003Cth class=\"has-text-align-center\" data-align=\"center\">Pennsylvania\u003C\u002Fth>\u003Cth class=\"has-text-align-center\" data-align=\"center\">Vermont\u003C\u002Fth>\u003Cth class=\"has-text-align-center\" data-align=\"center\">Massachusetts\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd class=\"has-text-align-center\" data-align=\"center\">\u003Cstrong>Weight\u003C\u002Fstrong>\u003C\u002Ftd>\u003Ctd class=\"has-text-align-center\" data-align=\"center\">0.4\u003C\u002Ftd>\u003Ctd class=\"has-text-align-center\" data-align=\"center\">0.3\u003C\u002Ftd>\u003Ctd class=\"has-text-align-center\" data-align=\"center\">0.3\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Ffigure>\n\n\n\n\u003Ch3 class=\"wp-block-heading\" id=\"during-treatment\">During treatment\u003C\u002Fh3>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Meta advertising is then switched on in New York and revenue generated in New York, Pennsylvania, Vermont and Massachusetts is to be observed.\u003C\u002Fli>\n\n\n\n\u003Cli>It is beneficial to monitor the test at equidistant time intervals.\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Ch3 class=\"wp-block-heading\" id=\"post-treatment\">Post-treatment\u003C\u002Fh3>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>The observed revenue data for the control groups is then used to approximate what New York revenue should’ve been in absence of the Meta advertising.\u003C\u002Fli>\n\n\n\n\u003Cli>Suppose the following data, for a given month, was observed:\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cfigure class=\"wp-block-table\">\u003Ctable>\u003Cthead>\u003Ctr>\u003Cth class=\"has-text-align-center\" data-align=\"center\">Region\u003C\u002Fth>\u003Cth class=\"has-text-align-center\" data-align=\"center\">New York\u003C\u002Fth>\u003Cth class=\"has-text-align-center\" data-align=\"center\">Pennsylvania\u003C\u002Fth>\u003Cth class=\"has-text-align-center\" data-align=\"center\">Vermont\u003C\u002Fth>\u003Cth class=\"has-text-align-center\" data-align=\"center\">Massachusetts\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd class=\"has-text-align-center\" data-align=\"center\">\u003Cstrong>Revenue\u003C\u002Fstrong>\u003C\u002Ftd>\u003Ctd class=\"has-text-align-center\" data-align=\"center\">$12,000\u003C\u002Ftd>\u003Ctd class=\"has-text-align-center\" data-align=\"center\">$16,000\u003C\u002Ftd>\u003Ctd class=\"has-text-align-center\" data-align=\"center\">$10,000\u003C\u002Ftd>\u003Ctd class=\"has-text-align-center\" data-align=\"center\">$11,500\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Ffigure>\n\n\n\n\u003Cfigure class=\"wp-block-table\">\u003Ctable>\u003Ctbody>\u003Ctr>\u003Ctd class=\"has-text-align-center\" data-align=\"center\">\u003Cstrong>New York \u003Cem>predicted\u003C\u002Fem> ≅ 0.4 x 12,000 + 0.3 x 10,000 + 0.3 x 11,500\u003C\u002Fstrong>\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"has-text-align-center\" data-align=\"center\">\u003Cstrong>New York \u003Cem>predicted\u003C\u002Fem> ≅ $4,800 + $3,000 + $3,450 = $11,250\u003C\u002Fstrong>\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Ffigure>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Calculate the uplift.\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cp class=\"has-text-align-center has-medium-font-size\">\u003Cstrong>\u003Cem>Uplift\u003C\u002Fem> = $12,000 – $11,250 = $750\u003C\u002Fstrong>\u003C\u002Fp>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Extract insights from results.\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Continue to monitor for an extended period after the test to assess any sustained effects and account for potential lagged impacts.\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Ch3 class=\"wp-block-heading\" id=\"results-and-interpretation\">Results and interpretation\u003C\u002Fh3>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Results suggest that switching on Meta ads has caused an increase in sales of $750 this month in New York alone.\u003C\u002Fli>\n\n\n\n\u003Cli>Insights gained from the incrementality test guide decision-making processes, informing future marketing investments and resource allocations based on evidence-based evaluations of channel effectiveness.\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cdiv class=\"wp-block-columns is-layout-flex wp-container-core-columns-is-layout-9d6595d7 wp-block-columns-is-layout-flex\">\n\u003Cdiv class=\"wp-block-column is-layout-flow wp-block-column-is-layout-flow\">\n\u003Ch2 class=\"wp-block-heading\" id=\"common-considerations-of-geo-testing\">Common considerations of geo-testing\u003C\u002Fh2>\n\n\n\n\u003Cp>If you find yourself in a scenario where you can’t simply “switch off” advertising in a specific geo, campaign targeting is something to watch out for. When reducing advertising spend, it is recommended that you do this on a geography level and not just reduce the spend immediately. For platforms like Meta, you’ll often find that users who are “most likely” to convert will be prioritised over anyone else. Although this is strategic for the campaign in question, for the purposes of measurement it can often warrant insignificant results.\u003C\u002Fp>\n\u003C\u002Fdiv>\n\n\n\n\u003Cdiv class=\"wp-block-column is-vertically-aligned-center is-layout-flow wp-block-column-is-layout-flow\">\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"1920\" height=\"1440\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fvariable-vs-control-uk.jpg?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-28313\"\u002F>\u003C\u002Ffigure>\u003C\u002Fdiv>\u003C\u002Fdiv>\n\u003C\u002Fdiv>\n\n\n\n\u003Cp>For example, say you wish to reduce TikTok spend by approximately 50% and observe the incremental loss in sales, you might find that reduction in spend will decrease the advertising reach and only focus on exposing those most likely to convert as opposed to exposing new users. To counter this, reduce spend by only exposing 50% of the region in question. If advertising spend isn’t uniformly distributed across that region, then only expose parts of that region such that spend is approximately reduced by 50%. \u003C\u002Fp>\n\n\n\n\u003Cp>Lastly, when conducting a geo-based incrementality test you should consider contamination between geographic regions. This occurs when a large number of users commute between the treatment and control regions for long enough to become exposed. Using a synthetic control accounts for this to a certain degree, but care should still be taken. \u003C\u002Fp>\n\n\n\n\u003Cdiv class=\"wp-block-columns is-layout-flex wp-container-core-columns-is-layout-9d6595d7 wp-block-columns-is-layout-flex\">\n\u003Cdiv class=\"wp-block-column is-layout-flow wp-block-column-is-layout-flow\" style=\"flex-basis:66.66%\">\n\u003Ch2 class=\"wp-block-heading\" id=\"how-to-interpret-the-results-of-your-incrementality-test\">How to interpret the results of your incrementality test\u003C\u002Fh2>\n\n\n\n\u003Cp>Interpreting the results of an incrementality test involves understanding the limitations of statistical testing. Rather than providing definitive proof, these tests aim to evaluate the strength of evidence for or against a hypothesis. Therefore, it’s essential to recognise that testing outcomes offer estimates rather than exact values. In Bayesian analysis, credible probabilities, such as 95% or 99%, indicate the confidence level that the observed results are indeed incremental and not merely due to chance. On the other hand, frequentist analysis typically involves reporting p-values, which provide insights into the statistical significance of the findings but do not directly quantify the degree of confidence in the results’ accuracy.\u003C\u002Fp>\n\u003C\u002Fdiv>\n\n\n\n\u003Cdiv class=\"wp-block-column is-vertically-aligned-center is-layout-flow wp-block-column-is-layout-flow\" style=\"flex-basis:33.33%\">\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"535\" height=\"377\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fimpact-of-incremental-testing.jpeg?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-28317\"\u002F>\u003C\u002Ffigure>\u003C\u002Fdiv>\u003C\u002Fdiv>\n\u003C\u002Fdiv>\n\n\n\n\u003Cp>It’s crucial for practitioners to approach test interpretation with a nuanced understanding of statistical inference. While statistical tests offer valuable insights into the effectiveness of marketing strategies, they inherently involve uncertainty and assumptions. By acknowledging these factors and carefully considering the context of the test, analysts can derive meaningful conclusions and make informed decisions regarding resource allocation and campaign optimisation strategies.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"what-if-your-channels-are-not-incremental\">What if your channels are not incremental?\u003C\u002Fh2>\n\n\n\n\u003Cp>When it comes to incrementality testing, if a channel isn’t significant now, it doesn’t mean it never will be! When faced with insignificant results, there are multiple different options to consider. First, seasonal fluctuations may render certain channels ineffective at the specific time of the test. In such cases, it might be prudent to temporarily suspend advertising on those channels and reallocate resources to more opportune avenues. \u003C\u002Fp>\n\n\n\n\u003Cp>Alternatively, if a channel consistently underperforms, experimentation with different advertising strategies could be beneficial. By conducting another round of testing with modified approaches, you can gain insights into consumer preferences and tailor strategies accordingly. \u003C\u002Fp>\n\n\n\n\u003Cp>Specifically, if a test is inconclusive and\u002For a channel appears to be ineffective, the factors to consider are: \u003C\u002Fp>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Cstrong>Execution\u003C\u002Fstrong> – are you advertising something with a short-term sales focus or a long-term brand-building focus, and is this appropriate for the channel?\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Offer\u003C\u002Fstrong> – if you are offering a sales promotion, or highlighting your ordinary pricing, are you competitive? \u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Creative\u003C\u002Fstrong> – visuals, video and copy all have an impact – have these been tested to ensure you are putting your best effort forward? \u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Inventory\u003C\u002Fstrong> – for some channels, you may be able to specify with a greater deal which inventory you are buying, or for which audience. Have these factors been considered and tested also? \u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cp>Finally, in situations where determining the root cause of channel ineffectiveness proves arduous, reallocating resources to alternative channels with proven incremental value may be the most efficient solution. Ultimately, by remaining agile and responsive to performance data, you can adapt your marketing strategies to achieve your campaign objectives. \u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"next-steps-if-your-channels-are-incremental\">Next steps if your channels are incremental?\u003C\u002Fh2>\n\n\n\n\u003Cp>In the context of conducting a geo-based incrementality test to measure uplift in a new potential channel’s effectiveness by increasing budgets, the obvious next step is to expand your marketing efforts to more geographies. Instead of walking in blind, you’ll now have an idea of what to expect from each geography. \u003C\u002Fp>\n\n\n\n\u003Cp>In the context of reducing budget for an existing channel, if you observe a significant decrease in lift you should plan to increase that budget again. You’ll have to take into account the delayed effects of advertising here, as increasing the budget again is unlikely to take effect instantly. \u003C\u002Fp>\n\n\n\n\u003Cp>On the other hand, if you observe no significant change in lift, it may well be in your best interest to consider reducing budget in more than just the treatment geography, and to consider re-deploying this budget into different activity elsewhere – via another experiment if possible. \u003C\u002Fp>\n\n\n\n\u003Cp>Lastly, if conducting a geo-based incrementality test to measure uplift in an existing potential channel’s effectiveness by increasing budgets, you may want to consider gradually increasing the budget for this channel across all geographies. Implementing MMM will further develop your understanding of channel saturation, which describes the diminishing return on investment available as you continue to over-invest in a particular channel and\u002For audience.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"how-long-should-an-incrementality-test-last\">How long should an incrementality test last?\u003C\u002Fh2>\n\n\n\n\u003Cp>In incrementality testing, the duration of the test plays a critical role in ensuring the reliability and robustness of the results. While there’s no definitive timeline, aiming for a duration of 6 to 12 weeks is often recommended for optimal insights. This time frame allows for the observation of consistent trends, offering greater confidence in the validity of the findings. \u003C\u002Fp>\n\n\n\n\u003Cp>Moreover, tests designed to measure declines in revenue should include periodic risk assessments to evaluate whether the potential loss outweighs the insights gained. If necessary, tests can be halted prematurely, but such decisions should be accompanied by a careful consideration of the test’s consistency and potential future use for calibration in MMM. \u003C\u002Fp>\n\n\n\n\u003Cp>Even after the cessation of treatment, continued observation can provide valuable insights into any lagged effects, allowing for a more comprehensive understanding of the channel’s impact over time.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"how-to-overcome-challenges-with-incrementality-testing\">How to overcome challenges with incrementality testing\u003C\u002Fh2>\n\n\n\n\u003Cp>In the realm of marketing, incrementality tests, while valuable, come with inherent limitations that may negatively impact the accuracy of your media effectiveness insights. One common challenge is the inadequate fit of synthetic controls, which can skew results and lead to misinterpretation. \u003C\u002Fp>\n\n\n\n\u003Cp>To address this, marketers can explore alternative methodologies, such as A\u002FB testing, to validate findings and ensure robustness. Additionally, prior to testing, analysis of goodness-of-fit statistics and hold out testing can help you to generate a sufficient synthetic control. \u003C\u002Fp>\n\n\n\n\u003Cp>Moreover, disparities between treatment and control groups may introduce bias into the analysis, necessitating the utilisation of rigorous control selection processes to mitigate individual differences. Additionally, the generalizability of findings from treatment geographies to the broader population may be uncertain, prompting the execution of multiple tests across diverse geographies to capture geographical variations. Furthermore, lift tests often provide short-term insights, limiting their applicability to long-term strategic planning.\u003C\u002Fp>\n\n\n\n\u003Cp> To extend the utility of lift test results, marketers can integrate them with Marketing Mix Modeling (MMM) to extrapolate findings and glean insights into medium- to long-term impacts, thereby enhancing the depth and breadth of their strategic decision-making processes. By acknowledging and overcoming these limitations, you can harness the full potential of incrementality testing to optimise your marketing strategies and drive sustainable growth.\u003C\u002Fp>\n\n\n\n\u003Cdiv class=\"wp-block-columns is-layout-flex wp-container-core-columns-is-layout-9d6595d7 wp-block-columns-is-layout-flex\">\n\u003Cdiv class=\"wp-block-column is-layout-flow wp-block-column-is-layout-flow\">\n\u003Ch2 class=\"wp-block-heading\" id=\"hold-out-testing-for-the-synthetic-control\">Hold out testing for the synthetic control\u003C\u002Fh2>\n\n\n\n\u003Cp>There are multiple measures you can use to validate the predictive accuracy of your synthetic control. A few of these measures include leave-one-out-cross-validation (LOO-CV), the Bayesian Information Criterion (BIC), R-squared and hold out testing.\u003C\u002Fp>\n\u003C\u002Fdiv>\n\n\n\n\u003Cdiv class=\"wp-block-column is-vertically-aligned-center is-layout-flow wp-block-column-is-layout-flow\">\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"956\" height=\"478\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fdataset-testing-model-2.png?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-28352\"\u002F>\u003C\u002Ffigure>\u003C\u002Fdiv>\u003C\u002Fdiv>\n\u003C\u002Fdiv>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Cstrong>LOO-CV\u003C\u002Fstrong> is a method for estimating the predictive performance of a statistical model, such as the synthetic control group in incrementality testing. LOO-CV helps to estimate how well a model will generalise to new, unseen data, hence providing insights into the predictive accuracy of your synthetic control group. \u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Cstrong>BIC\u003C\u002Fstrong> on the other hand is a criterion used for model selection. It balances the goodness of fit of the model with the complexity of the model, penalising models that are more complex. Lower values of BIC indicate a better balance between model fit and complexity, with the best model being the one with the lowest BIC value. This is particularly useful when using weighted averages to derive your synthetic control as it aims to quantify the point in which including fewer control regions in your model outweighs model accuracy.\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Alternatively, R-squared is used to assess how well your synthetic control mimics the treatment group during the pre-treatment phase. R-squared takes values between 0 and 1, 0 indicating poor fit and 1 indicating perfect fit. An R-squared of 0.7 indicates that 70% of the variation in the treatment group can be explained by the synthetic control group. Care must be taken when relying solely on R-squared to assess model fit as it doesn’t penalise for model complexity. This means that you may encounter overfitting, particularly when you observe an R-squared close to 1.\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Lastly, as opposed to using numerical goodness-of-fit statistics to validate the predictive accuracy of the synthetic control, you may wish to conduct a \u003Cstrong>hold out test\u003C\u002Fstrong>. Hold out testing involves splitting your pre-treatment data into two parts: train data and test data. The train data should be between 70% and 80% of the total pre-treatment data. The remaining 20-30% of total data is to be the testing data. You would then fit a synthetic control group to the train data and use it to forecast the test data. As the test data is unseen by the synthetic control, comparing these forecasts against what actually happened can provide insights into the predictive accuracy of the actual synthetic control. If the forecasts are close to the test data, you’ll have confidence that the synthetic control will perform well. This is also sometimes referred to as back-testing when the hold out is over a period of observations pre-test.\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"how-does-incrementality-testing-complement-marketing-mix-modelling\">How does incrementality testing complement Marketing Mix Modelling?\u003C\u002Fh2>\n\n\n\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full is-resized\">\u003Cimg loading=\"lazy\" decoding=\"async\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F11\u002Fmedia-mix-modelling-graphic.jpg?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-25577\" width=\"488\" height=\"270\"\u002F>\u003C\u002Ffigure>\u003C\u002Fdiv>\n\n\n\u003Cp>As mentioned earlier, incrementality testing serves as a valuable complement to Marketing Mix Modeling (MMM) by \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fincrementality-testing-planning\u002F\">enhancing the calibration process\u003C\u002Fa> and reducing uncertainty in the model’s predictions. By isolating the incremental impact of specific marketing interventions, incrementality testing provides valuable insights that inform and refine the MMM model. The results of incrementality tests can help identify channels or tactics with the largest uncertainty in their estimated impact, prompting the model to recommend further testing or adjustments to improve accuracy. \u003C\u002Fp>\n\n\n\n\u003Cp>Ultimately, the integration of incrementality testing with \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fmedia-solutions\u002Fmeasurement\u002F\">marketing mix models\u003C\u002Fa> enhances the businesses’ effectiveness in understanding the impact of marketing activities and optimising resource allocation.\u003C\u002Fp>\n\n\n\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Ca rel=\"noreferrer noopener\" href=\"\u002Fget-in-touch\u002F\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"3840\" height=\"540\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002FGeneric_Banner_01.jpg?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-27527\"\u002F>\u003C\u002Fa>\u003C\u002Ffigure>\u003C\u002Fdiv>",{"rendered":138,"protected":213},28330,{"_yoast_wpseo_focuskw":5,"_yoast_wpseo_title":728,"_yoast_wpseo_metadesc":729,"inline_featured_image":213,"footnotes":5},"The Ultimate Guide to Incrementality Testing | Impression","Incrementality testing aims to measure the additional sales generated exclusively as a result of the advertising efforts. Read the ultimate guide.",[222,223],[],[226],[734,113,229,230,231,232,233,234,235],"post-28273",[],[],[738,739,740,741],{"path":187,"name":188},{"path":241,"name":242},{"path":156,"name":157},{"path":196,"name":137},{"thumbnail":743,"medium":744,"medium_large":746,"large":748,"1536x1536":750,"2048x2048":752,"post-thumbnail":754,"12_col":756,"default_landscape":758,"default_portrait":760,"6_col_and_margin_short":762,"11_col_and_margin":764,"9_col_rectangle":766,"11_col":768,"staff_image_696_600":770,"rectangular_thumbnail":772},{"url":142,"width":247,"height":247},{"url":745,"width":70,"height":70},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpietro-jeng-n6B49lTx7NM-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=133&q=90&w=200",{"url":747,"width":73,"height":23},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpietro-jeng-n6B49lTx7NM-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=512&q=90&w=768",{"url":749,"width":254,"height":255},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpietro-jeng-n6B49lTx7NM-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=1000&q=90&w=1892",{"url":751,"width":79,"height":79},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpietro-jeng-n6B49lTx7NM-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=1024&q=90&w=1536",{"url":753,"width":82,"height":82},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpietro-jeng-n6B49lTx7NM-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=1365&q=90&w=2048",{"url":755,"width":85,"height":86},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpietro-jeng-n6B49lTx7NM-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=1180&q=90&w=1860",{"url":757,"width":88,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpietro-jeng-n6B49lTx7NM-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=2880",{"url":759,"width":91,"height":65},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpietro-jeng-n6B49lTx7NM-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=600&q=90&w=800",{"url":761,"width":65,"height":91},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpietro-jeng-n6B49lTx7NM-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=800&q=90&w=600",{"url":763,"width":85,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpietro-jeng-n6B49lTx7NM-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=1860",{"url":765,"width":95,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpietro-jeng-n6B49lTx7NM-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=3080",{"url":767,"width":97,"height":98},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpietro-jeng-n6B49lTx7NM-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=1400&q=90&w=2150",{"url":769,"width":100,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpietro-jeng-n6B49lTx7NM-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=2640",{"url":771,"width":64,"height":65},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpietro-jeng-n6B49lTx7NM-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=600&q=90&w=696",{"url":773,"width":103,"height":104},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpietro-jeng-n6B49lTx7NM-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=300&q=90&w=750",[775],{"id":215,"path":282,"display_name":4,"job_title":11,"description":6,"avatar_url":122,"acf":776},{"full_bio":10,"job_title":11,"specialist_service_links":777,"photo":784},[778,780,782],{"page":779,"link_text":29},{"ID":15,"post_author":16,"post_date":17,"post_date_gmt":17,"post_content":5,"post_title":18,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":21,"to_ping":5,"pinged":5,"post_modified":22,"post_modified_gmt":22,"post_content_filtered":5,"post_parent":23,"guid":24,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":28},{"page":781,"link_text":34},{"ID":32,"post_author":16,"post_date":33,"post_date_gmt":33,"post_content":5,"post_title":34,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":35,"to_ping":5,"pinged":5,"post_modified":36,"post_modified_gmt":36,"post_content_filtered":5,"post_parent":37,"guid":38,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":39},{"page":783,"link_text":44},{"ID":42,"post_author":16,"post_date":43,"post_date_gmt":43,"post_content":5,"post_title":44,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":45,"to_ping":5,"pinged":5,"post_modified":46,"post_modified_gmt":46,"post_content_filtered":5,"post_parent":37,"guid":47,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":48},{"ID":50,"id":50,"title":51,"filename":52,"filesize":53,"url":54,"link":55,"alt":5,"author":56,"description":5,"caption":5,"name":57,"status":58,"uploaded_to":23,"date":59,"modified":59,"menu_order":23,"mime_type":60,"type":61,"subtype":62,"icon":63,"width":64,"height":65,"sizes":785},{"thumbnail":67,"thumbnail-width":68,"thumbnail-height":68,"medium":69,"medium-width":70,"medium-height":71,"medium_large":72,"medium_large-width":73,"medium_large-height":74,"large":75,"large-width":76,"large-height":77,"1536x1536":78,"1536x1536-width":79,"1536x1536-height":80,"2048x2048":81,"2048x2048-width":82,"2048x2048-height":83,"post-thumbnail":84,"post-thumbnail-width":85,"post-thumbnail-height":86,"12_col":87,"12_col-width":88,"12_col-height":89,"default_landscape":90,"default_landscape-width":91,"default_landscape-height":65,"default_portrait":92,"default_portrait-width":65,"default_portrait-height":91,"6_col_and_margin_short":93,"6_col_and_margin_short-width":85,"6_col_and_margin_short-height":89,"11_col_and_margin":94,"11_col_and_margin-width":95,"11_col_and_margin-height":89,"9_col_rectangle":96,"9_col_rectangle-width":97,"9_col_rectangle-height":98,"11_col":99,"11_col-width":100,"11_col-height":89,"staff_image_696_600":101,"staff_image_696_600-width":64,"staff_image_696_600-height":65,"rectangular_thumbnail":102,"rectangular_thumbnail-width":103,"rectangular_thumbnail-height":104},[787,788],{"id":222,"name":125,"path":124},{"id":223,"name":157,"path":156},[790,791,792,793,795,796,797,798,799,800,801,802,804,806,808,809,810,811,812,813,814,815],{"property":299,"content":728},{"property":301,"content":729},{"name":303,"content":304},{"property":201,"content":794},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fincrementality-testing\u002F",{"name":308,"content":309},{"property":311,"content":312},{"property":314,"content":728},{"property":316,"content":729},{"property":318,"content":794},{"property":320,"content":321},{"property":323,"content":324},{"property":326,"content":803},"2024-05-01T15:30:40+00:00",{"property":329,"content":805},"2024-10-14T13:51:39+00:00",{"property":332,"content":807},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fopen-graph-ultimate-guide-to-incrementality-testing.png?auto=compress%2Cformat&q=90",{"property":335,"content":336},{"property":338,"content":339},{"property":341,"content":60},{"property":182,"content":4},{"property":344,"content":345},{"property":347,"content":807},{"property":349,"content":350},{"property":352,"content":350},{"@context":354,"@graph":817},[818,829,835,837,843,847,865],{"@type":293,"@id":819,"isPartOf":820,"author":821,"headline":137,"datePublished":803,"dateModified":805,"mainEntityOfPage":822,"wordCount":823,"publisher":824,"image":825,"thumbnailUrl":827,"articleSection":828,"inLanguage":368},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fincrementality-testing\u002F#article",{"@id":794},{"name":4,"@id":360},{"@id":794},6934,{"@id":363},{"@id":826},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fincrementality-testing\u002F#primaryimage","https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F05\u002Fpietro-jeng-n6B49lTx7NM-unsplash.jpg?auto=compress%2Cformat&q=90",[125,157],{"@type":370,"@id":794,"url":794,"name":728,"isPartOf":830,"primaryImageOfPage":831,"image":832,"thumbnailUrl":827,"datePublished":803,"dateModified":805,"description":729,"breadcrumb":833,"inLanguage":377},{"@id":372},{"@id":826},{"@id":826},{"@id":834},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fincrementality-testing\u002F#breadcrumb",{"@type":379,"inLanguage":368,"@id":826,"url":827,"contentUrl":827,"width":380,"height":836},1600,{"@type":383,"@id":834,"itemListElement":838},[839,840,841,842],{"@type":386,"position":387,"name":188,"item":388},{"@type":386,"position":151,"name":242,"item":390},{"@type":386,"position":163,"name":157,"item":392},{"@type":386,"position":394,"name":137},{"@type":396,"@id":372,"url":388,"name":321,"description":5,"publisher":844,"inLanguage":368,"potentialAction":845},{"@id":398},{"@type":400,"target":846,"query-input":404},{"@type":402,"urlTemplate":403},{"@type":406,"@id":398,"name":321,"url":388,"logo":848,"image":412,"sameAs":849,"@context":417,"email":418,"foundingDate":419,"legalName":420,"vatID":421,"description":422,"telephone":423,"address":850,"location":851,"areaServed":861},{"@type":379,"inLanguage":368,"@id":408,"url":409,"contentUrl":409,"width":410,"height":411,"caption":321},[414,415,416],{"@type":425,"streetAddress":426,"addressLocality":427,"addressRegion":428,"postalCode":429,"addressCountry":430},[852,855,858],{"@type":433,"name":434,"geo":853,"address":854,"photo":412,"telephone":440},{"@type":436,"latitude":437,"longitude":438},{"@type":425,"streetAddress":426,"addressLocality":427,"addressRegion":428,"postalCode":429,"addressCountry":430},{"@type":433,"name":442,"geo":856,"address":857,"photo":451,"telephone":452},{"@type":436,"latitude":444,"longitude":445},{"@type":425,"streetAddress":447,"addressLocality":448,"addressRegion":449,"postalCode":450,"addressCountry":430},{"@type":433,"name":454,"geo":859,"address":860},{"@type":436,"latitude":456,"longitude":457},{"@type":425,"streetAddress":459,"addressLocality":460,"addressRegion":461,"postalCode":462,"addressCountry":463},[862,863,864],{"@type":466,"name":467},{"@type":466,"name":469},{"@type":466,"name":463},{"@type":472,"@id":473,"name":4,"image":866,"description":6,"sameAs":867,"url":202},{"@type":379,"inLanguage":368,"@id":475,"url":8,"contentUrl":8,"caption":4},[198],{"self":869,"collection":874,"about":876,"author":878,"replies":880,"version-history":883,"predecessor-version":887,"wp:featuredmedia":891,"wp:attachment":894,"wp:term":897,"curies":904},[870],{"href":871,"targetHints":872},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F28273",{"allow":873},[483,484,485,486,487],[875],{"href":490},[877],{"href":493},[879],{"embeddable":108,"href":496},[881],{"embeddable":108,"href":882},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcomments?post=28273",[884],{"count":885,"href":886},62,"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F28273\u002Frevisions",[888],{"id":889,"href":890},30693,"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F28273\u002Frevisions\u002F30693",[892],{"embeddable":108,"href":893},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fmedia\u002F28330",[895],{"href":896},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fmedia?parent=28273",[898,900,902],{"taxonomy":516,"embeddable":108,"href":899},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcategories?post=28273",{"taxonomy":519,"embeddable":108,"href":901},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcollections?post=28273",{"taxonomy":182,"embeddable":108,"href":903},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcoauthors?post=28273",[905],{"name":525,"href":526,"templated":108},{"id":907,"date":908,"date_gmt":908,"guid":909,"modified":150,"modified_gmt":150,"slug":911,"status":19,"type":113,"link":149,"title":912,"content":913,"excerpt":915,"author":215,"featured_media":916,"comment_status":20,"ping_status":20,"sticky":213,"template":5,"format":217,"meta":917,"categories":918,"collections":919,"coauthors":920,"class_list":921,"acf":923,"head_links":924,"breadcrumbs":925,"url_path":149,"search_description":930,"featured_image_urls":931,"authors":963,"type_nicename":293,"reading_minutes":151,"category":975,"yoast_title":977,"yoast_meta":978,"yoast_json_ld":1005,"_links":1056},27823,"2024-04-08T14:08:24",{"rendered":910},"\u002F?p=27823","navigating-mmm",{"rendered":147},{"rendered":914,"protected":213},"\n\u003Cp>Before diving headfirst into marketing mix modelling implementation, it’s essential to consider several crucial questions. From understanding the intricacies of variable selection to aligning modelling objectives with your business goals, thoughtful consideration of the following points can pave the way for more accurate and actionable insights. \u003C\u002Fp>\n\n\n\n\u003Cp>Below, you’ll find a list of key questions to consider before embarking on your MMM journey.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"boundaries\">\u003Cstrong>Boundaries\u003C\u002Fstrong>\u003C\u002Fh2>\n\n\n\n\u003Cp>What model boundaries are needed for analysis? Should the model be built at a national, regional, or individual level? How do model boundaries affect the accuracy and interpretability of the results?\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"data-granularity\">\u003Cstrong>Data granularity\u003C\u002Fstrong>\u003C\u002Fh2>\n\n\n\n\u003Cp>What level of granularity is needed for modelling? Shall we use daily, weekly or monthly data? Daily data can be hard to model, and monthly data isn’t granular enough.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"variable-selection\">\u003Cstrong>Variable selection\u003C\u002Fstrong>\u003C\u002Fh2>\n\n\n\n\u003Cp>Which marketing and non-marketing variables should be included in the model? How do different marketing channels interact with each other and with external factors? Are there any significant drivers of sales that need to be accounted for? Do these variables explain the dependent variable enough?\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"parameter-estimation\">\u003Cstrong>Parameter estimation\u003C\u002Fstrong>\u003C\u002Fh2>\n\n\n\n\u003Cp>Parameter estimation is the essence of any statistical model. How are the parameters estimated in your model? Which parameters are estimated and which are chosen by the MMM practitioner? Can the practitioner justify their choice of parameters?\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"seasonality-and-trends\">\u003Cstrong>Seasonality and trends\u003C\u002Fstrong>\u003C\u002Fh2>\n\n\n\n\u003Cp>How does seasonality affect sales? Are there any long-term trends that need to be captured in the model? How can seasonality and trends be separated from the effects of marketing activities?\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"model-assumptions\">\u003Cstrong>Model assumptions\u003C\u002Fstrong>\u003C\u002Fh2>\n\n\n\n\u003Cp>What assumptions are made about the relationship between marketing inputs and sales outputs? How robust are these assumptions to changes in market conditions?\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"model-validation\">\u003Cstrong>Model validation\u003C\u002Fstrong>\u003C\u002Fh2>\n\n\n\n\u003Cp>How will the model be validated and tested? If our model recommends increasing the Meta budget by 10%, could we do it for validation purposes?\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"stakeholder-buy-in\">\u003Cstrong>Stakeholder buy-in\u003C\u002Fstrong>\u003C\u002Fh2>\n\n\n\n\u003Cp>How will the results of the market mix model be communicated to stakeholders? What level of understanding and involvement do they have in the model-building process?\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"continuous-improvement\">\u003Cstrong>Continuous improvement\u003C\u002Fstrong>\u003C\u002Fh2>\n\n\n\n\u003Cp>How will the MMM be maintained and updated over time? What mechanisms are in place to incorporate new data and insights into the model? Frequent tests help to validate the model.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"cost-benefit-analysis\">\u003Cstrong>Cost-benefit analysis\u003C\u002Fstrong>\u003C\u002Fh2>\n\n\n\n\u003Cp>What are the costs of building and maintaining the \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fmedia-solutions\u002Fmedia-effectiveness\u002F\">MMM\u003C\u002Fa>? What are the expected benefits in terms of improved decision-making and return on investment?\u003C\u002Fp>\n\n\n\n\u003Cp>Statistical modelling as a whole (not just MMM) is difficult. A good marketing mix model requires in-depth planning and careful consideration. As you delve into the realm of marketing measurement, it’s essential to lay a solid foundation by addressing key considerations before implementing your marketing mix model strategy. \u003C\u002Fp>\n\n\n\n\u003Cp>By carefully considering these thoughts and questions, you’ll be better equipped to construct a robust measurement framework that accurately captures the impact of your marketing initiatives.\u003C\u002Fp>\n",{"rendered":148,"protected":213},27838,{"_yoast_wpseo_focuskw":5,"_yoast_wpseo_title":5,"_yoast_wpseo_metadesc":5,"inline_featured_image":213,"footnotes":5},[223],[],[226],[922,113,229,230,231,232,233,235],"post-27823",[],[],[926,927,928,929],{"path":187,"name":188},{"path":241,"name":242},{"path":156,"name":157},{"path":196,"name":147},"Before diving headfirst into marketing mix modelling implementation, it’s essential to consider several crucial questions. From understanding the intricacies of variable selection to aligning modelling objectives with your business goals, thoughtful consideration of the following points can pave the way for more accurate and actionable insights. Below, you’ll find a list of key questions to […]",{"thumbnail":932,"medium":933,"medium_large":935,"large":937,"1536x1536":939,"2048x2048":941,"post-thumbnail":943,"12_col":945,"default_landscape":947,"default_portrait":949,"6_col_and_margin_short":951,"11_col_and_margin":953,"9_col_rectangle":955,"11_col":957,"staff_image_696_600":959,"rectangular_thumbnail":961},{"url":152,"width":247,"height":247},{"url":934,"width":70,"height":70},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Fjon-tyson-hhq1Lxtuwd8-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=200&q=90&w=200",{"url":936,"width":73,"height":23},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Fjon-tyson-hhq1Lxtuwd8-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=768&q=90&w=768",{"url":938,"width":254,"height":255},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Fjon-tyson-hhq1Lxtuwd8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=1000&q=90&w=1892",{"url":940,"width":79,"height":79},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Fjon-tyson-hhq1Lxtuwd8-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=1536&q=90&w=1536",{"url":942,"width":82,"height":82},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Fjon-tyson-hhq1Lxtuwd8-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=2048&q=90&w=2048",{"url":944,"width":85,"height":86},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Fjon-tyson-hhq1Lxtuwd8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=1180&q=90&w=1860",{"url":946,"width":88,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Fjon-tyson-hhq1Lxtuwd8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=2880",{"url":948,"width":91,"height":65},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Fjon-tyson-hhq1Lxtuwd8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=600&q=90&w=800",{"url":950,"width":65,"height":91},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Fjon-tyson-hhq1Lxtuwd8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=800&q=90&w=600",{"url":952,"width":85,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Fjon-tyson-hhq1Lxtuwd8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=1860",{"url":954,"width":95,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Fjon-tyson-hhq1Lxtuwd8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=3080",{"url":956,"width":97,"height":98},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Fjon-tyson-hhq1Lxtuwd8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=1400&q=90&w=2150",{"url":958,"width":100,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Fjon-tyson-hhq1Lxtuwd8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=2640",{"url":960,"width":64,"height":65},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Fjon-tyson-hhq1Lxtuwd8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=600&q=90&w=696",{"url":962,"width":103,"height":104},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Fjon-tyson-hhq1Lxtuwd8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=300&q=90&w=750",[964],{"id":215,"path":282,"display_name":4,"job_title":11,"description":6,"avatar_url":122,"acf":965},{"full_bio":10,"job_title":11,"specialist_service_links":966,"photo":973},[967,969,971],{"page":968,"link_text":29},{"ID":15,"post_author":16,"post_date":17,"post_date_gmt":17,"post_content":5,"post_title":18,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":21,"to_ping":5,"pinged":5,"post_modified":22,"post_modified_gmt":22,"post_content_filtered":5,"post_parent":23,"guid":24,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":28},{"page":970,"link_text":34},{"ID":32,"post_author":16,"post_date":33,"post_date_gmt":33,"post_content":5,"post_title":34,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":35,"to_ping":5,"pinged":5,"post_modified":36,"post_modified_gmt":36,"post_content_filtered":5,"post_parent":37,"guid":38,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":39},{"page":972,"link_text":44},{"ID":42,"post_author":16,"post_date":43,"post_date_gmt":43,"post_content":5,"post_title":44,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":45,"to_ping":5,"pinged":5,"post_modified":46,"post_modified_gmt":46,"post_content_filtered":5,"post_parent":37,"guid":47,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":48},{"ID":50,"id":50,"title":51,"filename":52,"filesize":53,"url":54,"link":55,"alt":5,"author":56,"description":5,"caption":5,"name":57,"status":58,"uploaded_to":23,"date":59,"modified":59,"menu_order":23,"mime_type":60,"type":61,"subtype":62,"icon":63,"width":64,"height":65,"sizes":974},{"thumbnail":67,"thumbnail-width":68,"thumbnail-height":68,"medium":69,"medium-width":70,"medium-height":71,"medium_large":72,"medium_large-width":73,"medium_large-height":74,"large":75,"large-width":76,"large-height":77,"1536x1536":78,"1536x1536-width":79,"1536x1536-height":80,"2048x2048":81,"2048x2048-width":82,"2048x2048-height":83,"post-thumbnail":84,"post-thumbnail-width":85,"post-thumbnail-height":86,"12_col":87,"12_col-width":88,"12_col-height":89,"default_landscape":90,"default_landscape-width":91,"default_landscape-height":65,"default_portrait":92,"default_portrait-width":65,"default_portrait-height":91,"6_col_and_margin_short":93,"6_col_and_margin_short-width":85,"6_col_and_margin_short-height":89,"11_col_and_margin":94,"11_col_and_margin-width":95,"11_col_and_margin-height":89,"9_col_rectangle":96,"9_col_rectangle-width":97,"9_col_rectangle-height":98,"11_col":99,"11_col-width":100,"11_col-height":89,"staff_image_696_600":101,"staff_image_696_600-width":64,"staff_image_696_600-height":65,"rectangular_thumbnail":102,"rectangular_thumbnail-width":103,"rectangular_thumbnail-height":104},[976],{"id":223,"name":157,"path":156},"Key questions to consider when implementing marketing mix modelling - Impression",[979,980,981,983,984,985,986,987,988,989,990,992,994,996,997,998,1000,1001,1002,1003,1004],{"property":299,"content":977},{"name":303,"content":304},{"property":201,"content":982},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fnavigating-mmm\u002F",{"name":308,"content":309},{"property":311,"content":312},{"property":314,"content":977},{"property":316,"content":930},{"property":318,"content":982},{"property":320,"content":321},{"property":323,"content":324},{"property":326,"content":991},"2024-04-08T14:08:24+00:00",{"property":329,"content":993},"2024-04-18T13:51:27+00:00",{"property":332,"content":995},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002FKey-questions-to-consider-when-implementing-marketing-mix-modelling-2.jpg?auto=compress%2Cformat&q=90",{"property":335,"content":336},{"property":338,"content":339},{"property":341,"content":999},"image\u002Fjpeg",{"property":182,"content":4},{"property":344,"content":345},{"property":347,"content":995},{"property":349,"content":350},{"property":352,"content":350},{"@context":354,"@graph":1006},[1007,1018,1024,1025,1031,1035,1053],{"@type":293,"@id":1008,"isPartOf":1009,"author":1010,"headline":147,"datePublished":991,"dateModified":993,"mainEntityOfPage":1011,"wordCount":1012,"publisher":1013,"image":1014,"thumbnailUrl":1016,"articleSection":1017,"inLanguage":368},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fnavigating-mmm\u002F#article",{"@id":982},{"name":4,"@id":360},{"@id":982},471,{"@id":363},{"@id":1015},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fnavigating-mmm\u002F#primaryimage","https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F04\u002Fjon-tyson-hhq1Lxtuwd8-unsplash.jpg?auto=compress%2Cformat&q=90",[157],{"@type":370,"@id":982,"url":982,"name":977,"isPartOf":1019,"primaryImageOfPage":1020,"image":1021,"thumbnailUrl":1016,"datePublished":991,"dateModified":993,"breadcrumb":1022,"inLanguage":377},{"@id":372},{"@id":1015},{"@id":1015},{"@id":1023},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fnavigating-mmm\u002F#breadcrumb",{"@type":379,"inLanguage":368,"@id":1015,"url":1016,"contentUrl":1016,"width":380,"height":380},{"@type":383,"@id":1023,"itemListElement":1026},[1027,1028,1029,1030],{"@type":386,"position":387,"name":188,"item":388},{"@type":386,"position":151,"name":242,"item":390},{"@type":386,"position":163,"name":157,"item":392},{"@type":386,"position":394,"name":147},{"@type":396,"@id":372,"url":388,"name":321,"description":5,"publisher":1032,"inLanguage":368,"potentialAction":1033},{"@id":398},{"@type":400,"target":1034,"query-input":404},{"@type":402,"urlTemplate":403},{"@type":406,"@id":398,"name":321,"url":388,"logo":1036,"image":412,"sameAs":1037,"@context":417,"email":418,"foundingDate":419,"legalName":420,"vatID":421,"description":422,"telephone":423,"address":1038,"location":1039,"areaServed":1049},{"@type":379,"inLanguage":368,"@id":408,"url":409,"contentUrl":409,"width":410,"height":411,"caption":321},[414,415,416],{"@type":425,"streetAddress":426,"addressLocality":427,"addressRegion":428,"postalCode":429,"addressCountry":430},[1040,1043,1046],{"@type":433,"name":434,"geo":1041,"address":1042,"photo":412,"telephone":440},{"@type":436,"latitude":437,"longitude":438},{"@type":425,"streetAddress":426,"addressLocality":427,"addressRegion":428,"postalCode":429,"addressCountry":430},{"@type":433,"name":442,"geo":1044,"address":1045,"photo":451,"telephone":452},{"@type":436,"latitude":444,"longitude":445},{"@type":425,"streetAddress":447,"addressLocality":448,"addressRegion":449,"postalCode":450,"addressCountry":430},{"@type":433,"name":454,"geo":1047,"address":1048},{"@type":436,"latitude":456,"longitude":457},{"@type":425,"streetAddress":459,"addressLocality":460,"addressRegion":461,"postalCode":462,"addressCountry":463},[1050,1051,1052],{"@type":466,"name":467},{"@type":466,"name":469},{"@type":466,"name":463},{"@type":472,"@id":473,"name":4,"image":1054,"description":6,"sameAs":1055,"url":202},{"@type":379,"inLanguage":368,"@id":475,"url":8,"contentUrl":8,"caption":4},[198],{"self":1057,"collection":1062,"about":1064,"author":1066,"replies":1068,"version-history":1071,"predecessor-version":1074,"wp:featuredmedia":1078,"wp:attachment":1081,"wp:term":1084,"curies":1091},[1058],{"href":1059,"targetHints":1060},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F27823",{"allow":1061},[483,484,485,486,487],[1063],{"href":490},[1065],{"href":493},[1067],{"embeddable":108,"href":496},[1069],{"embeddable":108,"href":1070},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcomments?post=27823",[1072],{"count":173,"href":1073},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F27823\u002Frevisions",[1075],{"id":1076,"href":1077},28065,"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F27823\u002Frevisions\u002F28065",[1079],{"embeddable":108,"href":1080},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fmedia\u002F27838",[1082],{"href":1083},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fmedia?parent=27823",[1085,1087,1089],{"taxonomy":516,"embeddable":108,"href":1086},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcategories?post=27823",{"taxonomy":519,"embeddable":108,"href":1088},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcollections?post=27823",{"taxonomy":182,"embeddable":108,"href":1090},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcoauthors?post=27823",[1092],{"name":525,"href":526,"templated":108},{"id":1094,"date":1095,"date_gmt":1095,"guid":1096,"modified":162,"modified_gmt":162,"slug":1098,"status":19,"type":113,"link":161,"title":1099,"content":1100,"excerpt":1102,"author":215,"featured_media":1103,"comment_status":20,"ping_status":20,"sticky":213,"template":5,"format":217,"meta":1104,"categories":1106,"collections":1107,"coauthors":1108,"class_list":1109,"acf":1111,"head_links":1112,"breadcrumbs":1113,"url_path":161,"search_description":1118,"featured_image_urls":1119,"authors":1151,"type_nicename":293,"reading_minutes":163,"category":1163,"yoast_title":1105,"yoast_meta":1165,"yoast_json_ld":1191,"_links":1243},27611,"2024-03-15T11:20:51",{"rendered":1097},"\u002F?p=27611","bayesian-statistics-in-media-mix-modelling-a-real-world-analogy",{"rendered":159},{"rendered":1101,"protected":213},"\n\u003Cp>\u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fguide-to-media-effectiveness-measurement\u002F\">Bayesian statistics\u003C\u002Fa> is a powerful tool for tackling the complexities inherent in media effectiveness analysis, making it an attractive approach for practitioners in the field. Bayesian statistics particularly shines in \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fmedia-mix-modelling\u002F\">media mix modelling\u003C\u002Fa> (MMM) for its ability to incorporate prior knowledge and enhance model robustness with limited or missing data. \u003C\u002Fp>\n\n\n\n\u003Cp>Additionally, Bayesian methods facilitate the handling of uncertainty, enabling more reliable inference and decision-making processes.\u003C\u002Fp>\n\n\n\n\u003Cp>The Bayesian framework consists of three main components: the prior, the likelihood and the posterior.\u003C\u002Fp>\n\n\n\n\u003Cp>In the context of MMM, the prior reflects any initial beliefs or assumptions on the model parameters before observing any data. The likelihood quantifies the probability of observing the data given the model’s prior. The posterior integrates the prior and likelihood, yielding “updated” beliefs after considering the observed data.\u003C\u002Fp>\n\n\n\n\u003Cp>It is the posterior that contains all relevant insights for an informed decision-making process.\u003C\u002Fp>\n\n\n\n\u003Cp>As an example, let us consider the task of predicting whether or not to wear a raincoat based on weather conditions. This can be modelled in a similar way to the digital marketer’s crucial challenge of allocating budgets for a campaign across various media channels.\u003C\u002Fp>\n\n\n\n\u003Cfigure class=\"wp-block-table\">\u003Ctable>\u003Ctbody>\u003Ctr>\u003Ctd>\u003Cstrong>Initial Uncertainty (Objective)\u003C\u002Fstrong>\u003C\u002Ftd>\u003Ctd>\u003Cstrong>Raincoat Scenario:\u003C\u002Fstrong> You wake up and your curtains are closed, you can’t hear any rainfall initially. This tells you a little about the chance of rainfall, but not enough to decide whether or not to wear a raincoat.\u003C\u002Ftd>\u003Ctd>\u003Cstrong>Marketing Campaign:\u003C\u002Fstrong> You wish to launch a campaign but don’t have any insights into channel performance to help you decide which channels to advertise on.\u003Cbr>\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Cstrong>Historical Beliefs (Prior)\u003C\u002Fstrong>\u003C\u002Ftd>\u003Ctd>\u003Cstrong>Raincoat Scenario:\u003C\u002Fstrong> Based on previous years you think there is about a 40% chance of rainfall around this time of year.\u003C\u002Ftd>\u003Ctd>\u003Cstrong>Marketing Campaign:\u003C\u002Fstrong> You recall your knowledge of channel performance for previous campaigns and how budgets were allocated then.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Cstrong>Seeing the data (Likelihood)\u003C\u002Fstrong>\u003C\u002Ftd>\u003Ctd>\u003Cstrong>Raincoat Scenario: \u003C\u002Fstrong>Upon looking outside you see the sky is overcast with dark, heavy clouds. The likelihood of needing a raincoat increases considerably, but not to certainty.\u003C\u002Ftd>\u003Ctd>\u003Cstrong>Marketing Campaign:\u003C\u002Fstrong> You update your campaign’s expected success using real-time data from various media channels.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Cstrong>Accounting for externalities (Externalities)\u003C\u002Fstrong>\u003C\u002Ftd>\u003Ctd>\u003Cstrong>Raincoat Scenario: \u003C\u002Fstrong>You check a weather app on your phone, which factors in wind speeds and humidity, and see a 60% chance of rainfall in your area.\u003C\u002Ftd>\u003Ctd>\u003Cstrong>Marketing Campaign:\u003C\u002Fstrong> You decide to consider external factors that may impact the performance of your campaign, such as seasonality and consumer behaviour.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Cstrong>Decision-making insights (Posterior)\u003C\u002Fstrong>\u003C\u002Ftd>\u003Ctd>\u003Cstrong>Raincoat Scenario:\u003C\u002Fstrong> Combining your historical beliefs with today’s observations and the weather forecasts, you estimate there is a 75% chance of needing a raincoat and deciding to wear one.\u003C\u002Ftd>\u003Ctd>\u003Cstrong>Marketing Campaign:\u003C\u002Fstrong> Through incorporating prior knowledge, historical data and externalities, the Bayesian MMM has produced an estimate of your optimal media mix and its associated probabilities. You choose to allocate budgets for the campaign this way.\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Ffigure>\n\n\n\n\u003Cp>To conclude, the increasing popularity of Bayesian statistics within media effectiveness measurement reflects a paradigm shift towards more robust and flexible frameworks. By leveraging Bayesian methods, marketers can better account for uncertainty, incorporate prior knowledge, and refine their insights, leading to more accurate and actionable marketing strategies.\u003C\u002Fp>\n\n\n\n\u003Cfigure class=\"wp-block-image size-full\">\u003Ca rel=\"noreferrer noopener\" href=\"\u002Fmedia-solutions\u002Fmedia-effectiveness\u002F\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"3840\" height=\"540\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002FMeasurement_Banner_01.jpg?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-27690\"\u002F>\u003C\u002Fa>\u003C\u002Ffigure>\n",{"rendered":160,"protected":213},27624,{"_yoast_wpseo_focuskw":5,"_yoast_wpseo_title":1105,"_yoast_wpseo_metadesc":5,"inline_featured_image":213,"footnotes":5},"Bayesian Statistics in Media Mix Modelling: A real-world analogy | Impression",[223],[],[226],[1110,113,229,230,231,232,233,235],"post-27611",[],[],[1114,1115,1116,1117],{"path":187,"name":188},{"path":241,"name":242},{"path":156,"name":157},{"path":196,"name":159},"Bayesian statistics is a powerful tool for tackling the complexities inherent in media effectiveness analysis, making it an attractive approach for practitioners in the field. Bayesian statistics particularly shines in media mix modelling (MMM) for its ability to incorporate prior knowledge and enhance model robustness with limited or missing data. Additionally, Bayesian methods facilitate the […]",{"thumbnail":1120,"medium":1121,"medium_large":1123,"large":1125,"1536x1536":1127,"2048x2048":1129,"post-thumbnail":1131,"12_col":1133,"default_landscape":1135,"default_portrait":1137,"6_col_and_margin_short":1139,"11_col_and_margin":1141,"9_col_rectangle":1143,"11_col":1145,"staff_image_696_600":1147,"rectangular_thumbnail":1149},{"url":164,"width":247,"height":247},{"url":1122,"width":70,"height":70},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002Fjakub-zerdzicki-ip7GFn5JqX8-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=133&q=90&w=200",{"url":1124,"width":73,"height":23},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002Fjakub-zerdzicki-ip7GFn5JqX8-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=512&q=90&w=768",{"url":1126,"width":254,"height":255},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002Fjakub-zerdzicki-ip7GFn5JqX8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=1000&q=90&w=1892",{"url":1128,"width":79,"height":79},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002Fjakub-zerdzicki-ip7GFn5JqX8-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=1025&q=90&w=1536",{"url":1130,"width":82,"height":82},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002Fjakub-zerdzicki-ip7GFn5JqX8-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=1366&q=90&w=2048",{"url":1132,"width":85,"height":86},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002Fjakub-zerdzicki-ip7GFn5JqX8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=1180&q=90&rect=0%2C78%2C2400%2C1522&w=1860",{"url":1134,"width":88,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002Fjakub-zerdzicki-ip7GFn5JqX8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=2880",{"url":1136,"width":91,"height":65},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002Fjakub-zerdzicki-ip7GFn5JqX8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=600&q=90&w=800",{"url":1138,"width":65,"height":91},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002Fjakub-zerdzicki-ip7GFn5JqX8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=800&q=90&w=600",{"url":1140,"width":85,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002Fjakub-zerdzicki-ip7GFn5JqX8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=1860",{"url":1142,"width":95,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002Fjakub-zerdzicki-ip7GFn5JqX8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=3080",{"url":1144,"width":97,"height":98},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002Fjakub-zerdzicki-ip7GFn5JqX8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=1400&q=90&w=2150",{"url":1146,"width":100,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002Fjakub-zerdzicki-ip7GFn5JqX8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=2640",{"url":1148,"width":64,"height":65},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002Fjakub-zerdzicki-ip7GFn5JqX8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=600&q=90&w=696",{"url":1150,"width":103,"height":104},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002Fjakub-zerdzicki-ip7GFn5JqX8-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=300&q=90&rect=0%2C640%2C2400%2C960&w=750",[1152],{"id":215,"path":282,"display_name":4,"job_title":11,"description":6,"avatar_url":122,"acf":1153},{"full_bio":10,"job_title":11,"specialist_service_links":1154,"photo":1161},[1155,1157,1159],{"page":1156,"link_text":29},{"ID":15,"post_author":16,"post_date":17,"post_date_gmt":17,"post_content":5,"post_title":18,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":21,"to_ping":5,"pinged":5,"post_modified":22,"post_modified_gmt":22,"post_content_filtered":5,"post_parent":23,"guid":24,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":28},{"page":1158,"link_text":34},{"ID":32,"post_author":16,"post_date":33,"post_date_gmt":33,"post_content":5,"post_title":34,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":35,"to_ping":5,"pinged":5,"post_modified":36,"post_modified_gmt":36,"post_content_filtered":5,"post_parent":37,"guid":38,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":39},{"page":1160,"link_text":44},{"ID":42,"post_author":16,"post_date":43,"post_date_gmt":43,"post_content":5,"post_title":44,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":45,"to_ping":5,"pinged":5,"post_modified":46,"post_modified_gmt":46,"post_content_filtered":5,"post_parent":37,"guid":47,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":48},{"ID":50,"id":50,"title":51,"filename":52,"filesize":53,"url":54,"link":55,"alt":5,"author":56,"description":5,"caption":5,"name":57,"status":58,"uploaded_to":23,"date":59,"modified":59,"menu_order":23,"mime_type":60,"type":61,"subtype":62,"icon":63,"width":64,"height":65,"sizes":1162},{"thumbnail":67,"thumbnail-width":68,"thumbnail-height":68,"medium":69,"medium-width":70,"medium-height":71,"medium_large":72,"medium_large-width":73,"medium_large-height":74,"large":75,"large-width":76,"large-height":77,"1536x1536":78,"1536x1536-width":79,"1536x1536-height":80,"2048x2048":81,"2048x2048-width":82,"2048x2048-height":83,"post-thumbnail":84,"post-thumbnail-width":85,"post-thumbnail-height":86,"12_col":87,"12_col-width":88,"12_col-height":89,"default_landscape":90,"default_landscape-width":91,"default_landscape-height":65,"default_portrait":92,"default_portrait-width":65,"default_portrait-height":91,"6_col_and_margin_short":93,"6_col_and_margin_short-width":85,"6_col_and_margin_short-height":89,"11_col_and_margin":94,"11_col_and_margin-width":95,"11_col_and_margin-height":89,"9_col_rectangle":96,"9_col_rectangle-width":97,"9_col_rectangle-height":98,"11_col":99,"11_col-width":100,"11_col-height":89,"staff_image_696_600":101,"staff_image_696_600-width":64,"staff_image_696_600-height":65,"rectangular_thumbnail":102,"rectangular_thumbnail-width":103,"rectangular_thumbnail-height":104},[1164],{"id":223,"name":157,"path":156},[1166,1167,1168,1170,1171,1172,1173,1174,1175,1176,1177,1179,1181,1183,1184,1185,1186,1187,1188,1189,1190],{"property":299,"content":1105},{"name":303,"content":304},{"property":201,"content":1169},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fbayesian-statistics-in-media-mix-modelling-a-real-world-analogy\u002F",{"name":308,"content":309},{"property":311,"content":312},{"property":314,"content":1105},{"property":316,"content":1118},{"property":318,"content":1169},{"property":320,"content":321},{"property":323,"content":324},{"property":326,"content":1178},"2024-03-15T11:20:51+00:00",{"property":329,"content":1180},"2024-03-22T10:21:02+00:00",{"property":332,"content":1182},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002Fopen-graph-statistics-anaology.png?auto=compress%2Cformat&q=90",{"property":335,"content":336},{"property":338,"content":339},{"property":341,"content":60},{"property":182,"content":4},{"property":344,"content":345},{"property":347,"content":1182},{"property":349,"content":350},{"property":352,"content":350},{"@context":354,"@graph":1192},[1193,1204,1210,1212,1218,1222,1240],{"@type":293,"@id":1194,"isPartOf":1195,"author":1196,"headline":159,"datePublished":1178,"dateModified":1180,"mainEntityOfPage":1197,"wordCount":1198,"publisher":1199,"image":1200,"thumbnailUrl":1202,"articleSection":1203,"inLanguage":368},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fbayesian-statistics-in-media-mix-modelling-a-real-world-analogy\u002F#article",{"@id":1169},{"name":4,"@id":360},{"@id":1169},522,{"@id":363},{"@id":1201},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fbayesian-statistics-in-media-mix-modelling-a-real-world-analogy\u002F#primaryimage","https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F03\u002Fjakub-zerdzicki-ip7GFn5JqX8-unsplash.jpg?auto=compress%2Cformat&q=90",[157],{"@type":370,"@id":1169,"url":1169,"name":1105,"isPartOf":1205,"primaryImageOfPage":1206,"image":1207,"thumbnailUrl":1202,"datePublished":1178,"dateModified":1180,"breadcrumb":1208,"inLanguage":377},{"@id":372},{"@id":1201},{"@id":1201},{"@id":1209},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fbayesian-statistics-in-media-mix-modelling-a-real-world-analogy\u002F#breadcrumb",{"@type":379,"inLanguage":368,"@id":1201,"url":1202,"contentUrl":1202,"width":380,"height":1211},1601,{"@type":383,"@id":1209,"itemListElement":1213},[1214,1215,1216,1217],{"@type":386,"position":387,"name":188,"item":388},{"@type":386,"position":151,"name":242,"item":390},{"@type":386,"position":163,"name":157,"item":392},{"@type":386,"position":394,"name":159},{"@type":396,"@id":372,"url":388,"name":321,"description":5,"publisher":1219,"inLanguage":368,"potentialAction":1220},{"@id":398},{"@type":400,"target":1221,"query-input":404},{"@type":402,"urlTemplate":403},{"@type":406,"@id":398,"name":321,"url":388,"logo":1223,"image":412,"sameAs":1224,"@context":417,"email":418,"foundingDate":419,"legalName":420,"vatID":421,"description":422,"telephone":423,"address":1225,"location":1226,"areaServed":1236},{"@type":379,"inLanguage":368,"@id":408,"url":409,"contentUrl":409,"width":410,"height":411,"caption":321},[414,415,416],{"@type":425,"streetAddress":426,"addressLocality":427,"addressRegion":428,"postalCode":429,"addressCountry":430},[1227,1230,1233],{"@type":433,"name":434,"geo":1228,"address":1229,"photo":412,"telephone":440},{"@type":436,"latitude":437,"longitude":438},{"@type":425,"streetAddress":426,"addressLocality":427,"addressRegion":428,"postalCode":429,"addressCountry":430},{"@type":433,"name":442,"geo":1231,"address":1232,"photo":451,"telephone":452},{"@type":436,"latitude":444,"longitude":445},{"@type":425,"streetAddress":447,"addressLocality":448,"addressRegion":449,"postalCode":450,"addressCountry":430},{"@type":433,"name":454,"geo":1234,"address":1235},{"@type":436,"latitude":456,"longitude":457},{"@type":425,"streetAddress":459,"addressLocality":460,"addressRegion":461,"postalCode":462,"addressCountry":463},[1237,1238,1239],{"@type":466,"name":467},{"@type":466,"name":469},{"@type":466,"name":463},{"@type":472,"@id":473,"name":4,"image":1241,"description":6,"sameAs":1242,"url":202},{"@type":379,"inLanguage":368,"@id":475,"url":8,"contentUrl":8,"caption":4},[198],{"self":1244,"collection":1249,"about":1251,"author":1253,"replies":1255,"version-history":1258,"predecessor-version":1262,"wp:featuredmedia":1266,"wp:attachment":1269,"wp:term":1272,"curies":1279},[1245],{"href":1246,"targetHints":1247},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F27611",{"allow":1248},[483,484,485,486,487],[1250],{"href":490},[1252],{"href":493},[1254],{"embeddable":108,"href":496},[1256],{"embeddable":108,"href":1257},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcomments?post=27611",[1259],{"count":1260,"href":1261},10,"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F27611\u002Frevisions",[1263],{"id":1264,"href":1265},27691,"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F27611\u002Frevisions\u002F27691",[1267],{"embeddable":108,"href":1268},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fmedia\u002F27624",[1270],{"href":1271},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fmedia?parent=27611",[1273,1275,1277],{"taxonomy":516,"embeddable":108,"href":1274},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcategories?post=27611",{"taxonomy":519,"embeddable":108,"href":1276},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcollections?post=27611",{"taxonomy":182,"embeddable":108,"href":1278},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcoauthors?post=27611",[1280],{"name":525,"href":526,"templated":108},{"id":1282,"date":1283,"date_gmt":1283,"guid":1284,"modified":172,"modified_gmt":172,"slug":1286,"status":19,"type":113,"link":171,"title":1287,"content":1288,"excerpt":1290,"author":215,"featured_media":1291,"comment_status":20,"ping_status":20,"sticky":213,"template":5,"format":217,"meta":1292,"categories":1294,"collections":1296,"coauthors":1297,"class_list":1298,"acf":1301,"head_links":1302,"breadcrumbs":1303,"url_path":171,"search_description":1308,"featured_image_urls":1309,"authors":1341,"type_nicename":293,"reading_minutes":173,"category":1353,"yoast_title":1293,"yoast_meta":1356,"yoast_json_ld":1383,"_links":1436},27087,"2024-02-05T15:00:02",{"rendered":1285},"\u002F?p=27087","guide-to-media-effectiveness-measurement",{"rendered":169},{"rendered":1289,"protected":213},"\n\u003Cp>Upon researching Media Effectiveness you might have stumbled across the following terms: media mix modelling, econometric modelling, incrementality testing, synthetic data, multiple linear regression, Bayesian, frequentist, R-squared, geo-experiments and multi-touch attribution.\u003C\u002Fp>\n\n\n\n\u003Cp>You’re not alone if you read all this and find the world of Media Effectiveness a little intimidating. For a discipline based on transparency and coherence, Media Effectiveness is encompassed with technical terms. If you’re eager to learn about Media Effectiveness measurement and all the business insights it can provide, then you’ve come to the right place!\u003C\u002Fp>\n\n\n\n\u003Cp>This blog will outline how to approach Media Effectiveness from first principles, the crucial data-driven insights you’ll get from it and list some benefits & limitations of doing so.\u003C\u002Fp>\n\n\n\n\u003Cdiv class=\"table-of-contents\">\u003Cul>\u003Cli class=\"heading-h2\">\u003Ca href=\"#why-media-effectiveness\">Why Media Effectiveness?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#media-mix-modelling\">Media mix modelling\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#incrementality-testing\">Incrementality testing\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#bayesian-vs-frequentist-statistics\">Bayesian vs Frequentist statistics\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#strengths-and-limitations-the-bayesian-media-mix-model\">Strengths and limitations: The Bayesian media mix model\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"heading-h2\">\u003Ca href=\"#strengths-and-limitations-the-frequentist-media-mix-model\">Strengths and limitations: The Frequentist media mix model\u003C\u002Fa>\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdiv>\n\n\n\n\u003Chr class=\"wp-block-separator has-alpha-channel-opacity is-style-wide\"\u002F>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"why-media-effectiveness\">\u003Cstrong>Why Media Effectiveness?\u003C\u002Fstrong>\u003C\u002Fh2>\n\n\n\n\u003Cp>For decades digital marketers have used digital \u003Cstrong>multi-touch attribution\u003C\u002Fstrong> \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fanalytics\u002F\">web analytics\u003C\u002Fa> to track and assign credit to multiple media channels based on “touchpoints” within the customer journey. In recent years this type of measurement has become increasingly flawed as a result of the rise in usage of multiple digital devices, the decaying lifetime of third-party cookies, and device and app-based privacy controls. To address this challenge, many marketers are turning to Media Effectiveness measurement techniques in order to enhance decision-making processes. \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fmedia-solutions\u002Fmedia-effectiveness\u002F\">Media Effectiveness measurement\u003C\u002Fa> utilises both modelling and testing to ensure that it remains the most reliable and trustworthy method of marketing measurement.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"media-mix-modelling\">Media mix modelling\u003C\u002Fh2>\n\n\n\n\u003Cdiv class=\"wp-block-columns is-layout-flex wp-container-core-columns-is-layout-9d6595d7 wp-block-columns-is-layout-flex\">\n\u003Cdiv class=\"wp-block-column is-layout-flow wp-block-column-is-layout-flow\" style=\"flex-basis:66.66%\">\n\u003Cp>Digital marketing has become an integral part of modern business strategies, enabling companies to reach and engage their target audiences online. However, making informed decisions in digital marketing requires more than just intuition; it demands a data-driven approach. \u003Cstrong>Statistical modelling\u003C\u002Fstrong> is a powerful tool that can be used to help marketers extract valuable insights, predict future trends, and optimise the impact of their campaigns. \u003C\u002Fp>\n\u003C\u002Fdiv>\n\n\n\n\u003Cdiv class=\"wp-block-column is-layout-flow wp-block-column-is-layout-flow\" style=\"flex-basis:33.33%\">\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-thumbnail is-resized\">\u003Cimg loading=\"lazy\" decoding=\"async\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Fmodelling-types.png?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-27124\" width=\"160\" height=\"160\"\u002F>\u003C\u002Ffigure>\u003C\u002Fdiv>\u003C\u002Fdiv>\n\u003C\u002Fdiv>\n\n\n\n\u003Cp>One type of statistical modelling is \u003Cstrong>econometric modelling\u003C\u002Fstrong>\u003Cem>. \u003C\u002Fem>Econometric modelling is a broad term that encompasses the modelling of a wide range of economic factors and relationships, not limited to just marketing. \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fmedia-mix-modelling\u002F\">\u003Cstrong>Media mix modelling\u003C\u002Fstrong>\u003C\u002Fa> and marketing mix modelling, often abbreviated to “MMM” are both subsets of econometric modelling. In the context of digital marketing, both are considered interchangeable. \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fmedia-solutions\u002Fmedia-effectiveness\u002F\">Media mix modelling\u003C\u002Fa> is a type of econometric model specifically tailored to digital marketing.\u003Cstrong> \u003C\u002Fstrong>It is the media mix model that is most commonly used among Media Effectiveness specialists. \u003C\u002Fp>\n\n\n\n\u003Cp>If you’re seeking data-driven insights into the effectiveness of various media channels, the delayed effects of advertising, the diminishing returns of media spend, and media contribution to KPIs and budget forecasts then media mix modelling is the way to go about it.\u003C\u002Fp>\n\n\n\n\u003Cp>Before you attempt media mix modelling, there are some requirements that are absolutely essential:\u003C\u002Fp>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Cstrong>Good quality data\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Sufficient and clean historical data on marketing activities, sales, and other relevant metrics are essential. Typically a minimum of two to three years of granular data, such as daily or weekly, is recommended. This ensures the model can capture variations and trends adequately.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Cstrong>Clearly defined business objectives\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Articulate the objectives that the model aims to address. These objectives could include optimising marketing budget allocation, understanding channel contributions, insights into the delayed effects or predicting sales impact.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cp>Provided you have good quality data at your disposal and clearly defined business objectives you aim to answer, the next thing to do is to go ahead and commence the modelling process. \u003C\u002Fp>\n\n\n\n\u003Cp>\u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fwww.warc.com\u002Fnewsandopinion\u002Fopinion\u002Fan-open-letter-to-marketing-departments-from-experts-on-evaluating-advertising\u002Fen-gb\u002F6261\">Be cautious of automated MMMs\u003C\u002Fa> as they can’t always capture the complexities of your business. It is recommended that you opt for a bespoke model as you’ll be able to make the relevant changes to best fit your business. Take comfort from the fact that there are a large number of bespoke model-building packages out there such as LightweightMMM, Robyn and PyMC-Marketing.\u003C\u002Fp>\n\n\n\n\u003Cp>Regardless of whether you invest in automated MMMs or begin building your own media mix model, the nature of these models varies. The most common statistical technique for MMM is multiple linear regression. Multiple linear regression is a form of a statistical model that uses two or more independent variables (marketing inputs) to predict the dependent variable (marketing output). This technique relies heavily on identifying the linear correlations and associations between variables. Within the linear regression model, certain variable transformations can be made in order to account for delayed effects and the effects of diminishing returns.\u003C\u002Fp>\n\n\n\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"1517\" height=\"746\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Fdecomposition-chart.png?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-27122\"\u002F>\u003Cfigcaption class=\"wp-element-caption\">Theoretical stacked decomposition chart\u003C\u002Ffigcaption>\u003C\u002Ffigure>\u003C\u002Fdiv>\n\n\n\u003Cp>It is good practice when conducting an MMM to account for the “base” in your model. The base contains anything that might contribute to the outcome in the absence of any marketing activity. Often, for brands that are well known and have already developed their long-term brand building, this “base” will take up the majority and is generally increasing over time. Accounting for the base in your model provides a stronger understanding of the impact of your marketing efforts and avoids incorrect channel contributions.\u003C\u002Fp>\n\n\n\n\u003Cp>Whether the nature of your MMM is multiple linear regression or anything else, you might ask yourself: “Do I trust these results?”. When conducting media mix modelling it is important to validate the model’s accuracy and reliability using various techniques such as measuring \u003Cstrong>R-squared or \u003C\u002Fstrong>comparing model predictions with actual outcomes. \u003C\u002Fp>\n\n\n\n\u003Cp>R-squared is simply a measure of how well the model can account for the variance in the data, hence the closer R-squared is to 1 the better the model fit. There is no “acceptable” value, but as a rule of thumb, I would recommend aiming for an R-squared of at least 0.65 when conducting an MMM. Additionally, it’s important to note that a really good R-squared value doesn’t necessarily mean you have an excellent model. Depending on the nature of your model, it is important to note that there are other statistics, such as LOO-CV, that serve similar purposes. Which statistic you use to compare models should depend on the nature of your models.\u003C\u002Fp>\n\n\n\n\u003Cp>Once you have a well-constructed MMM which aligns with your business and that fits the data really well you can begin to assess what tangible impacts these results might have on your business. Some examples include:\u003C\u002Fp>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Cstrong>Optimal budget allocation:\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>MMM can be used to identify the most effective media channels and activities that drive sales or desired outcomes. As a result, you can allocate marketing budget more efficiently by investing in channels that deliver a higher ROAS.\u003Cbr>\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Enhance overall strategy:\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>MMM reveals the relative contributions of each media channel, allowing you to understand the holistic impact of your current strategy. As a result, you can refine your strategies based on data-driven insights, leading to more effective and efficient campaigns.\u003Cbr>\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Data-driven decision-making:\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>MMM provides a data-driven approach to decision-making, eliminating the need for MTA or guesswork. As a result, you can make more informed and strategic decisions based on concrete evidence from your marketing performance.\u003Cbr>\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Forecasting and planning:\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>MMM uses historical data and identifies trends, enabling you to forecast future performance under different scenarios. As a result, you can make informed decisions about resource allocation, product launches, and overall business strategy.\u003Cbr>\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Competitive advantages:\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>MMM can give your business a competitive edge by taking into account competitor activity when optimising marketing strategies and resource allocation. By leveraging MMM, you can be sure to outperform your competitors in terms of marketing efficiency and effectiveness.\u003Cbr>\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Testing insights:\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>MMM can be used to estimate ROAS. Often these models provide a measure of uncertainty with these estimates. It is usually implied that the channel with the largest uncertainty requires some testing in order to reduce this uncertainty in future models.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cp>By utilising the insights generated via MMM, you can align your marketing strategy with data-driven decision-making, leading to a tangible improvement in business performance.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"incrementality-testing\">Incrementality testing\u003C\u002Fh2>\n\n\n\n\u003Cp>Accurately measuring whether or not, and to what extent, a particular marketing channel or campaign has an impact on your business is a common challenge faced in digital marketing. \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fwww.warc.com\u002Fnewsandopinion\u002Fopinion\u002Fles-binet-on-why-long-term-marketing-matters-in-the-age-of-short-termism\u002Fen-gb\u002F3307#:~:text=Any%20advertising%20or%20marketing%20campaign,short%2Dterm%20metrics%20with%20caution.\">All forms of advertising should lead to incremental sales\u003C\u002Fa>, even long-term brand-building activity, but it’s quantifying the contribution that advertising has that marketers find particularly challenging. If you’re seeking ways to accurately measure and assign credit or impact to media channels to strengthen your understanding of what performs best then testing is definitely something you should add to the toolbox.\u003C\u002Fp>\n\n\n\n\u003Cp>In the world of Media Effectiveness, the most frequently used method of testing is incrementality testing. Incrementality testing is a statistical test used in digital marketing to assessing the true impact of a specific marketing intervention or campaign through the comparison of a treatment group with at least 1 other control group. The primary goal is to measure the incremental or additional value generated by a particular action, such as an advertising campaign or introducing a new media channel.\u003C\u002Fp>\n\n\n\n\u003Cp>In a similar sense to modelling, before even thinking about incrementality testing you must make sure you have good quality data at your disposal and clearly defined business objectives you aim to answer.\u003C\u002Fp>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Cstrong>Good quality data\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Sufficient and clean historical data on the KPI you wish to measure split up by region, product or however you choose to group your data. Typically a minimum of one to two years of granular data, such as daily or weekly, is recommended.\u003Cbr>\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Control of confounding variables\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>When conducting an incrementality test, it is essential that the only thing you’re measuring is the thing you intend to. Any other changes made during the testing period could jeopardise the validity of the test. \u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Cstrong>Clearly defined business objectives\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Clearly articulate the business objectives that the test aims to address.\u003Cbr>\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Risk assessments\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>In the case where you’re testing for something that might negatively impact business performance, knowing when to stop the test becomes crucial. For instance, if you’re testing if a decrease in channel spend will have an effect on revenue, then you should determine the point at which insights from the test no longer outweigh the revenue lost.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cp>Once you have these, you’re all set to conduct an incrementality test!\u003C\u002Fp>\n\n\n\n\u003Cp>As mentioned above, an incrementality test consists of the comparison of a treatment group with at least 1 other control group. In the context of marketing, the treatment group refers to those that experience the marketing intervention and the control group refers to those that do not. There are two methods by which an incrementality test can be conducted: through A\u002FB testing or using a synthetic control.\u003Cbr>\u003Cbr>A\u002FB testing, often referred to as “market matching”, is a subset of incrementality testing in which we compare one treatment group (A) with one control group (B) and assess the difference. This difference is known as “lift”.  Although A\u002FB testing is based on sound statistical principles, it doesn’t account for the effects of individual differences. For instance, if you group by location, the slight differences between the target audiences of each location may negatively impact the validity of the test.\u003C\u002Fp>\n\n\n\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"1545\" height=\"895\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Fcausal-impact-artificial-control.png?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-27121\"\u002F>\u003Cfigcaption class=\"wp-element-caption\">Testing the treatment group against an artificial control\u003C\u002Ffigcaption>\u003C\u002Ffigure>\u003C\u002Fdiv>\n\n\n\u003Cp class=\"has-text-align-left\">Testing with a synthetic control on the other hand eliminates this flaw. A \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fwhat-is-synthetic-data\u002F\">\u003Cstrong>synthetic \u003C\u002Fstrong>\u003C\u002Fa>\u003Cb>control-based\u003C\u002Fb> incrementality test compares a treatment group with an artificial control group and assesses the lift. The artificial control group is derived by modelling the treatment group at the pre-treatment stage and then forecasting what would’ve happened to the treatment group in the absence of the treatment. There are a large number of ways to do this such as taking weighted averages of other control groups, using linear regression or time series forecasting methods.\u003C\u002Fp>\n\n\n\n\u003Cp>Incrementality testing is generally quite quick to implement and relies on objective data, hence reducing biases that may influence decision-making. The results are derived from actual user behaviour and are generally easy to interpret. Additionally, there are lots of easy-to-use software out there to conduct incrementality testing of any nature, such as CausalImpact and CausalPy.\u003C\u002Fp>\n\n\n\n\u003Cp>As mentioned previously, incremental testing can be conducted using geographical location. We call these types of tests \u003Cstrong>geo-experiments\u003C\u002Fstrong>. While incrementality testing is a broader concept, geo-experiments represent a specific type of incrementality testing focused on geographic variations. In other words, geo-experiments are an element of incrementality testing, and the principles of incrementality testing can be applied to understand the causal impact of marketing interventions, whether they are geo-specific or not. \u003C\u002Fp>\n\n\n\n\u003Cp>Having all these different methods of testing is great, but how do you know which one to use and when?\u003Cbr>Geo-experiments specifically concentrate on geographic regions, while customer testing may focus on a broader set of variables impacting the customer experience. If you aim to measure and optimise marketing strategies by region, geo-experiments are the ones to use, while customer testing is better suited when you aim to optimise the overall customer experience. A\u002FB testing is well-suited for comparing two groups with a randomised design, while synthetic control is useful when randomisation is not feasible, and researchers seek to estimate the causal impact on a single unit.\u003C\u002Fp>\n\n\n\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"1877\" height=\"1408\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Fgeo-location-testing.png?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-27123\"\u002F>\u003Cfigcaption class=\"wp-element-caption\">\u003Cem>Treatment and control locations were used to generate the treatment and artificial control groups.\u003C\u002Fem>\u003C\u002Ffigcaption>\u003C\u002Ffigure>\u003C\u002Fdiv>\n\n\n\u003Cp class=\"has-text-align-left\">Once you have correctly conducted an incrementality test which controls for confounding variables and has operated within the specified risk assessment, you can begin to assess what tangible impacts these results might have on your business. Some examples include:\u003C\u002Fp>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Optimise Spend\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Incrementality testing allows you to allocate marketing budget more efficiently, focusing on the channels that provide the highest incremental lift.\u003Cbr>\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\n\n\n\u003Cli>Decision making\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Incrementality testing promotes a data-driven culture within your business. Decisions about marketing strategies, product changes, or operational improvements are based on numerical evidence rather than assumptions.\u003Cbr>\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\n\n\n\u003Cli>Mitigation of unwanted consequences\n\u003Cul class=\"wp-block-list\">\n\u003Cli>By conducting an incrementality test on a small scale first, you can assess the unwanted consequences of certain strategies. This allows your businesses to identify and mitigate any negative impacts on other areas of the business.\u003Cbr>\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\n\n\n\u003Cli>Media mix model calibration\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Ca rel=\"noreferrer noopener\" href=\"\u002Fblog\u002Fleveraging-incrementality-testing-to-optimise-media-mix-modelling\u002F\">Impact or lift estimates can be used to fine-tune your media mix model\u003C\u002Fa> to give a better representation of your media mix, and hence to better inform your budget allocation forecasts.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cp>Overall, there are multiple tangible impacts of incrementality testing on your business, ranging from cost savings and revenue growth to improved customer relationships and strategic decision-making. By rigorously measuring the incremental impact of various initiatives, businesses can adapt and thrive in a dynamic market environment. Calibrating your media mix model through incrementality testing is considered the gold standard in \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fhbr.org\u002F2023\u002F03\u002Fa-new-gold-standard-for-digital-ad-measurement\">Media Effectiveness evaluation\u003C\u002Fa>.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"bayesian-vs-frequentist-statistics\">Bayesian vs Frequentist statistics\u003C\u002Fh2>\n\n\n\u003Cdiv class=\"wp-block-image\">\n\u003Cfigure class=\"aligncenter size-full is-resized\">\u003Cimg loading=\"lazy\" decoding=\"async\" src=\"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Fbayesian-statistics-example.png?auto=compress%2Cformat&q=90\" alt=\"\" class=\"wp-image-27120\" width=\"840\" height=\"302\"\u002F>\u003Cfigcaption class=\"wp-element-caption\">Illustration of how a combination of the prior and the data produces the posterior.\u003C\u002Ffigcaption>\u003C\u002Ffigure>\u003C\u002Fdiv>\n\n\n\u003Cp class=\"has-text-align-left\">Statistical models involve the estimation of parameters. The parameters in an MMM are unknown quantities associated with each marketing input, such as channel spend or the delayed effects of advertising. These parameters are then estimated and used to indicate the strength and direction of the influence each marketing input has on your business performance. \u003C\u002Fp>\n\n\n\n\u003Cp>Traditional statistical modelling utilises the \u003Cstrong>frequentist\u003C\u002Fstrong> framework. In frequentist statistics, parameters are treated as some unknown real value which we wish to estimate. Parameter estimation would then result in a singular point estimate and any uncertainty would be interpreted as the long-run frequency of events, hence the name “frequentist”. Uncertainty is quantified in the form of confidence intervals and is often expressed as a percentage level of confidence, such as 95%. If you were to repeatedly construct an MMM and estimate the parameters a large number of times, then the interval in which 95% of the estimates fall would be your 95% confidence interval.\u003C\u002Fp>\n\n\n\n\u003Cp>However, there exists an alternative framework known as \u003Cstrong>Bayesian\u003C\u002Fstrong> statistics, which approaches estimation and uncertainty from a different perspective. In Bayesian statistics, parameters are treated as a random variable as opposed to a fixed real value. Parameter estimation would then result in an entire distribution of point estimates, each with its probability attached to it. Under a Bayesian framework, uncertainty is quantified in the form of the highest posterior density intervals, which are simply an interval containing the point estimates of the highest probability, such as 95%. \u003C\u002Fp>\n\n\n\n\u003Cp>At the core of Bayesian statistics is Bayes’ theorem, a mathematical formula that describes how to update probabilities based on new evidence. In the context of MMM, this allows you to incorporate any business knowledge or beliefs such as marketing trends, pricing information or competitor activities into the modelling process in the form of a prior distribution. These prior distributions can be tailored towards the unique context of your business, hence ensuring recommendations derived from the MMM align with your unique goals and objectives. Once these prior distributions are fed into the model, the data is then introduced to update these distributions to produce what is called the posterior distribution.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"strengths-and-limitations-the-bayesian-media-mix-model\">Strengths and limitations: The Bayesian media mix model\u003C\u002Fh2>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Cstrong>Incorporation of prior knowledge:\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>This is particularly valuable when there is existing information or expert opinions available that will ensure your results align with your business.\u003Cbr>\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Quantifies uncertainty:\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Bayesian models naturally quantify uncertainty. This can be advantageous in the decision-making process, allowing you to understand the range of possible outcomes and associated uncertainties.\u003Cbr>\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Computational complexity:\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Bayesian MMM can be computationally demanding. The need for numerical methods can make the estimation process time-consuming, particularly for large datasets or intricate models.\u003Cbr>\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Subjectivity in the priors:\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>The choice of prior distributions can massively impact the results. Different analysts might have different opinions on appropriate priors. To counter this, elicit a larger variance in your prior distribution, this will allow your posterior to become more data-driven.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Ch2 class=\"wp-block-heading\" id=\"strengths-and-limitations-the-frequentist-media-mix-model\">Strengths and limitations: The Frequentist media mix model\u003C\u002Fh2>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Cstrong>Objectivity:\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Frequentist models are perceived as more objective, as they rely only on the observed data without incorporating subjective prior beliefs. This can be advantageous in situations where there is limited prior knowledge or when a more neutral approach is preferred.\u003Cbr>\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Simple to implement:\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Frequentist methods, such as maximum likelihood estimation (MLE), often involve simpler computational procedures compared to Bayesian methods.\u003Cbr>\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Fixed parameter interpretation:\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Frequentist models treat model parameters as fixed. This fixed interpretation might be limiting when dealing with complex, dynamic systems in which uncertainty is crucial to express.\u003Cbr>\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>Small data sets:\u003C\u002Fstrong>\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Frequentist methods may encounter challenges when dealing with small datasets, especially in situations where a Bayesian prior could provide valuable regularisation. Small sample sizes can lead to less stable parameter estimates.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Chr class=\"wp-block-separator has-alpha-channel-opacity is-style-wide\"\u002F>\n\n\n\n\u003Cp>Hopefully, you can see that once you strip away all the complicated jargon, the world of media effectiveness isn’t as intimidating as it’s perceived. This beginner’s guide has provided insights into the fundamental concepts, methodologies, and considerations that form the backbone of media effectiveness measurement. \u003C\u002Fp>\n\n\n\n\u003Cp>As you delve deeper into the realm of \u003Ca rel=\"noreferrer noopener\" href=\"\u002Fmedia-solutions\u002Fmedia-effectiveness\u002F\">Media Effectiveness\u003C\u002Fa>, remember that it is an ever-changing field, constantly evolving with technology advancements and market dynamics. Armed with this knowledge, you are better equipped to optimise your media strategies and drive success in an ever-changing digital landscape!\u003C\u002Fp>\n",{"rendered":170,"protected":213},27132,{"_yoast_wpseo_focuskw":5,"_yoast_wpseo_title":1293,"_yoast_wpseo_metadesc":5,"inline_featured_image":213,"footnotes":5},"Beginners’ guide to Incrementality Testing and Media Mix Modelling | Impression",[1295,222],66,[],[226],[1299,113,229,230,231,232,233,1300,234],"post-27087","category-beginners-guides",[],[],[1304,1305,1306,1307],{"path":187,"name":188},{"path":241,"name":242},{"path":124,"name":125},{"path":196,"name":169},"Upon researching Media Effectiveness you might have stumbled across the following terms: media mix modelling, econometric modelling, incrementality testing, synthetic data, multiple linear regression, Bayesian, frequentist, R-squared, geo-experiments and multi-touch attribution. You’re not alone if you read all this and find the world of Media Effectiveness a little intimidating. For a discipline based on transparency […]",{"thumbnail":1310,"medium":1311,"medium_large":1313,"large":1315,"1536x1536":1317,"2048x2048":1319,"post-thumbnail":1321,"12_col":1323,"default_landscape":1325,"default_portrait":1327,"6_col_and_margin_short":1329,"11_col_and_margin":1331,"9_col_rectangle":1333,"11_col":1335,"staff_image_696_600":1337,"rectangular_thumbnail":1339},{"url":174,"width":247,"height":247},{"url":1312,"width":70,"height":70},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Furiel-sc-11KDtiUWRq4-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=133&q=90&w=200",{"url":1314,"width":73,"height":23},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Furiel-sc-11KDtiUWRq4-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=510&q=90&w=768",{"url":1316,"width":254,"height":255},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Furiel-sc-11KDtiUWRq4-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=1000&q=90&w=1892",{"url":1318,"width":79,"height":79},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Furiel-sc-11KDtiUWRq4-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=1020&q=90&w=1536",{"url":1320,"width":82,"height":82},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Furiel-sc-11KDtiUWRq4-unsplash.jpg?auto=compress%2Cformat&fit=scale&h=1360&q=90&w=2048",{"url":1322,"width":85,"height":86},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Furiel-sc-11KDtiUWRq4-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=1180&q=90&w=1860",{"url":1324,"width":88,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Furiel-sc-11KDtiUWRq4-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=2880",{"url":1326,"width":91,"height":65},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Furiel-sc-11KDtiUWRq4-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=600&q=90&w=800",{"url":1328,"width":65,"height":91},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Furiel-sc-11KDtiUWRq4-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=800&q=90&w=600",{"url":1330,"width":85,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Furiel-sc-11KDtiUWRq4-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=1860",{"url":1332,"width":95,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Furiel-sc-11KDtiUWRq4-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=3080",{"url":1334,"width":97,"height":98},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Furiel-sc-11KDtiUWRq4-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=1400&q=90&w=2150",{"url":1336,"width":100,"height":89},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Furiel-sc-11KDtiUWRq4-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=870&q=90&w=2640",{"url":1338,"width":64,"height":65},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Furiel-sc-11KDtiUWRq4-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=600&q=90&w=696",{"url":1340,"width":103,"height":104},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Furiel-sc-11KDtiUWRq4-unsplash.jpg?auto=compress%2Cformat&fit=crop&h=300&q=90&w=750",[1342],{"id":215,"path":282,"display_name":4,"job_title":11,"description":6,"avatar_url":122,"acf":1343},{"full_bio":10,"job_title":11,"specialist_service_links":1344,"photo":1351},[1345,1347,1349],{"page":1346,"link_text":29},{"ID":15,"post_author":16,"post_date":17,"post_date_gmt":17,"post_content":5,"post_title":18,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":21,"to_ping":5,"pinged":5,"post_modified":22,"post_modified_gmt":22,"post_content_filtered":5,"post_parent":23,"guid":24,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":28},{"page":1348,"link_text":34},{"ID":32,"post_author":16,"post_date":33,"post_date_gmt":33,"post_content":5,"post_title":34,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":35,"to_ping":5,"pinged":5,"post_modified":36,"post_modified_gmt":36,"post_content_filtered":5,"post_parent":37,"guid":38,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":39},{"page":1350,"link_text":44},{"ID":42,"post_author":16,"post_date":43,"post_date_gmt":43,"post_content":5,"post_title":44,"post_excerpt":5,"post_status":19,"comment_status":20,"ping_status":20,"post_password":5,"post_name":45,"to_ping":5,"pinged":5,"post_modified":46,"post_modified_gmt":46,"post_content_filtered":5,"post_parent":37,"guid":47,"menu_order":23,"post_type":25,"post_mime_type":5,"comment_count":26,"filter":27,"url_path":48},{"ID":50,"id":50,"title":51,"filename":52,"filesize":53,"url":54,"link":55,"alt":5,"author":56,"description":5,"caption":5,"name":57,"status":58,"uploaded_to":23,"date":59,"modified":59,"menu_order":23,"mime_type":60,"type":61,"subtype":62,"icon":63,"width":64,"height":65,"sizes":1352},{"thumbnail":67,"thumbnail-width":68,"thumbnail-height":68,"medium":69,"medium-width":70,"medium-height":71,"medium_large":72,"medium_large-width":73,"medium_large-height":74,"large":75,"large-width":76,"large-height":77,"1536x1536":78,"1536x1536-width":79,"1536x1536-height":80,"2048x2048":81,"2048x2048-width":82,"2048x2048-height":83,"post-thumbnail":84,"post-thumbnail-width":85,"post-thumbnail-height":86,"12_col":87,"12_col-width":88,"12_col-height":89,"default_landscape":90,"default_landscape-width":91,"default_landscape-height":65,"default_portrait":92,"default_portrait-width":65,"default_portrait-height":91,"6_col_and_margin_short":93,"6_col_and_margin_short-width":85,"6_col_and_margin_short-height":89,"11_col_and_margin":94,"11_col_and_margin-width":95,"11_col_and_margin-height":89,"9_col_rectangle":96,"9_col_rectangle-width":97,"9_col_rectangle-height":98,"11_col":99,"11_col-width":100,"11_col-height":89,"staff_image_696_600":101,"staff_image_696_600-width":64,"staff_image_696_600-height":65,"rectangular_thumbnail":102,"rectangular_thumbnail-width":103,"rectangular_thumbnail-height":104},[1354,1355],{"id":1295,"name":179,"path":178},{"id":222,"name":125,"path":124},[1357,1358,1359,1361,1362,1363,1364,1365,1366,1367,1368,1370,1372,1374,1375,1376,1377,1378,1379,1381,1382],{"property":299,"content":1293},{"name":303,"content":304},{"property":201,"content":1360},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fguide-to-media-effectiveness-measurement\u002F",{"name":308,"content":309},{"property":311,"content":312},{"property":314,"content":1293},{"property":316,"content":1308},{"property":318,"content":1360},{"property":320,"content":321},{"property":323,"content":324},{"property":326,"content":1369},"2024-02-05T15:00:02+00:00",{"property":329,"content":1371},"2024-02-06T13:57:02+00:00",{"property":332,"content":1373},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Fopen-graph-beginner-guide-incrementality-media-mix.png?auto=compress%2Cformat",{"property":335,"content":336},{"property":338,"content":339},{"property":341,"content":60},{"property":182,"content":4},{"property":344,"content":345},{"property":347,"content":1380},"https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Fopen-graph-beginner-guide-incrementality-media-mix.png?auto=compress%2Cformat&q=90",{"property":349,"content":350},{"property":352,"content":350},{"@context":354,"@graph":1384},[1385,1396,1402,1404,1411,1415,1433],{"@type":293,"@id":1386,"isPartOf":1387,"author":1388,"headline":169,"datePublished":1369,"dateModified":1371,"mainEntityOfPage":1389,"wordCount":1390,"publisher":1391,"image":1392,"thumbnailUrl":1394,"articleSection":1395,"inLanguage":368},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fguide-to-media-effectiveness-measurement\u002F#article",{"@id":1360},{"name":4,"@id":360},{"@id":1360},3107,{"@id":363},{"@id":1393},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fguide-to-media-effectiveness-measurement\u002F#primaryimage","https:\u002F\u002Fimages.impression.co.uk\u002F2024\u002F02\u002Furiel-sc-11KDtiUWRq4-unsplash.jpg?auto=compress%2Cformat&q=90",[179,125],{"@type":370,"@id":1360,"url":1360,"name":1293,"isPartOf":1397,"primaryImageOfPage":1398,"image":1399,"thumbnailUrl":1394,"datePublished":1369,"dateModified":1371,"breadcrumb":1400,"inLanguage":377},{"@id":372},{"@id":1393},{"@id":1393},{"@id":1401},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fguide-to-media-effectiveness-measurement\u002F#breadcrumb",{"@type":379,"inLanguage":368,"@id":1393,"url":1394,"contentUrl":1394,"width":380,"height":1403},1594,{"@type":383,"@id":1401,"itemListElement":1405},[1406,1407,1408,1410],{"@type":386,"position":387,"name":188,"item":388},{"@type":386,"position":151,"name":242,"item":390},{"@type":386,"position":163,"name":125,"item":1409},"https:\u002F\u002Fwww.impressiondigital.com\u002Fblog\u002Fcategory\u002Fdigital-marketing\u002F",{"@type":386,"position":394,"name":169},{"@type":396,"@id":372,"url":388,"name":321,"description":5,"publisher":1412,"inLanguage":368,"potentialAction":1413},{"@id":398},{"@type":400,"target":1414,"query-input":404},{"@type":402,"urlTemplate":403},{"@type":406,"@id":398,"name":321,"url":388,"logo":1416,"image":412,"sameAs":1417,"@context":417,"email":418,"foundingDate":419,"legalName":420,"vatID":421,"description":422,"telephone":423,"address":1418,"location":1419,"areaServed":1429},{"@type":379,"inLanguage":368,"@id":408,"url":409,"contentUrl":409,"width":410,"height":411,"caption":321},[414,415,416],{"@type":425,"streetAddress":426,"addressLocality":427,"addressRegion":428,"postalCode":429,"addressCountry":430},[1420,1423,1426],{"@type":433,"name":434,"geo":1421,"address":1422,"photo":412,"telephone":440},{"@type":436,"latitude":437,"longitude":438},{"@type":425,"streetAddress":426,"addressLocality":427,"addressRegion":428,"postalCode":429,"addressCountry":430},{"@type":433,"name":442,"geo":1424,"address":1425,"photo":451,"telephone":452},{"@type":436,"latitude":444,"longitude":445},{"@type":425,"streetAddress":447,"addressLocality":448,"addressRegion":449,"postalCode":450,"addressCountry":430},{"@type":433,"name":454,"geo":1427,"address":1428},{"@type":436,"latitude":456,"longitude":457},{"@type":425,"streetAddress":459,"addressLocality":460,"addressRegion":461,"postalCode":462,"addressCountry":463},[1430,1431,1432],{"@type":466,"name":467},{"@type":466,"name":469},{"@type":466,"name":463},{"@type":472,"@id":473,"name":4,"image":1434,"description":6,"sameAs":1435,"url":202},{"@type":379,"inLanguage":368,"@id":475,"url":8,"contentUrl":8,"caption":4},[198],{"self":1437,"collection":1442,"about":1444,"author":1446,"replies":1448,"version-history":1451,"predecessor-version":1455,"wp:featuredmedia":1459,"wp:attachment":1462,"wp:term":1465,"curies":1472},[1438],{"href":1439,"targetHints":1440},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F27087",{"allow":1441},[483,484,485,486,487],[1443],{"href":490},[1445],{"href":493},[1447],{"embeddable":108,"href":496},[1449],{"embeddable":108,"href":1450},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcomments?post=27087",[1452],{"count":1453,"href":1454},40,"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F27087\u002Frevisions",[1456],{"id":1457,"href":1458},27203,"\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F27087\u002Frevisions\u002F27203",[1460],{"embeddable":108,"href":1461},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fmedia\u002F27132",[1463],{"href":1464},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fmedia?parent=27087",[1466,1468,1470],{"taxonomy":516,"embeddable":108,"href":1467},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcategories?post=27087",{"taxonomy":519,"embeddable":108,"href":1469},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcollections?post=27087",{"taxonomy":182,"embeddable":108,"href":1471},"\u002Fwp-json\u002Fwp\u002Fv2\u002Fcoauthors?post=27087",[1473],{"name":525,"href":526,"templated":108},{"title":1475,"script":1476,"meta":1490,"link":1503},"Harry Brace | Director, Head of Solutions | Impression",[1477],{"type":1478,"key":1479,"innerHTML":1480},"application\u002Fld+json","yoast-json-ld",{"@context":354,"@graph":1481},[1482,1487],{"@id":473,"@type":472,"name":4,"url":-1,"sameAs":1483,"image":1484,"description":6,"jobTitle":11,"knowsAbout":1485,"hasCredential":5,"gender":5,"worksFor":1486},[198],{"@type":379,"inLanguage":368,"@id":475,"url":8,"contentUrl":8,"caption":4},"Web Analytics, Tag Management, Google Marketing Platform",{"@id":398},{"@type":406,"@id":398,"name":321,"url":388,"logo":1488,"image":412,"sameAs":1489,"@context":417,"email":418,"foundingDate":419,"legalName":420,"vatID":421,"description":422,"telephone":423},{"@type":379,"inLanguage":368,"@id":408,"url":409,"contentUrl":409,"width":410,"height":411,"caption":321},[414,415,416],[1491,1492,1493,1495,1496,1498,1499,1501,1502],{"name":303,"content":304},{"property":308,"content":368},{"property":311,"content":1494},"profile",{"property":314,"content":1475},{"property":318,"content":1497},"https:\u002F\u002Fadmin.impressiondigital.com\u002Fabout\u002Four-team\u002Fharry-brace\u002F",{"property":320,"content":321},{"property":332,"content":1500},"https:\u002F\u002Fimages.impression.co.uk\u002F2023\u002F10\u002FHarry-Brace2.png?auto=compress&fit=fit&fm=jpg&h=500&q=90&w=500",{"name":344,"content":345},{"name":352,"content":350},[]]