Impression’s Head of SEO Performance, Charlie Norledge, and Head of Behavioural Science, Mike Weir, discuss why a business might want to replatform its website, through site migrations, new designs and new domains.
Charlie talks about the risk to organic visibility and how to avoid losing out on search engine traffic. Mike explores how redesigning a site by democracy may hinder a website’s ability to perform if the customer isn’t considered at every stage of the journey.
- 00:06 – Navigating website replatforms
- 02:00 – The risks associated with replatforming and how to mitigate them
- 04:11 – Impression’s approach to website replatforming
As referenced in the video, get your free copy of our site migration checklist here.
Mike Weir: Website re-platforms are big-ticket items for lots of businesses, so they have considerable investment and lots of different stakeholders want lots of different things from them. Can you talk a little bit about how re-platforms can mean different things to different people?
Charlie Norledge: Yeah, re-platforms mean a lot of different things to different businesses. We have people come to us when they’re trying to re-platform. It can mean a new design, a migration, a new domain, changes to content. What we need to do is make sure we understand what they’re actually asking for on those re-platforms and work with them to make sure they understand the risks of what’s going on.
Mike Weir: Yeah, that’s really interesting. We find often as a DX team that sometimes they start off with like a website migration from one position to another and very quickly they turn into like a redesign project by stealth. So someone somewhere doesn’t like a web page or doesn’t like a blog, so the next thing you know, they’re trying to edge that into the designs as well. Obviously that can bake in a lot of risk if people haven’t considered their audience and haven’t considered how Google might treat that content afterwards.
Charlie Norledge: And what we have to be sure to do is make sure that those businesses understand the risks. Often during a replatforming process, they might think, we’ll just change these designs, we’ll bring in this new content, but that carries the inherent risk to any organic visibility.mAnd something we need to make sure that they’re aware of early on in the project.
Mike Weir: Yes, there’s numerous hazards, isn’t there, I suppose, around things that can go wrong. I think one of the things we often see is design by democracy. So basically lots of different people taking the web designs and giving feedback, there can be a lack of involvement from the actual customer. Ultimately, if you’re redesigning a web page, the person who that web page is for is the customer so you want to get their feedback rather than someone in the Marketing Department or someone in the Sales Department or someone in an Accounts Department because ultimately they’re not the people who this is actually designed for.
So trying to make sure you’ve kind of qualitative insight, trying to make sure that you’ve done AB Testing to some of these bigger changes that you’re thinking about is definitely crucial if you’re going to launch a new website. What are the pivotal risks involved in something
like a re-platform?
Charlie Norledge: As an SEO, any time that you’re changing a URL or a design or content, there are always risks to traffic. When you migrate one URL to another, it takes time for search engines to understand what those new URLs mean. It has to pass all those ranking signals over and whilst that might seem fairly straightforward, actually Google and other search engines say that can take up to four months for that to be done.
A really good example of where this is something that hasn’t gone too well is Woo.com, they rebranded from WooCommerce to Woo.com, left it for a month, and they actually made a decision to revert back to WooCommerce.com because that migration didn’t work.
So they went through a re-platforming process and part of that was changing the domain. But that change in domain caused so many issues they actually reverted on it. So rather than sticking the course and going ahead with all that investment they made, they went back.
Mike Weir: What could they have done to avoid that challenge?
Charlie Norledge: So my favourite thing when it comes to any re-platforming, is to do it in stages. So rather than doing a massive thing like changing your design and move into new domain in one go, do it in stages.
So the first thing would be if you have a new design, put that on the same website. That way you’re going to make sure that you can understand what changes happen when you implement that new design. Then if you have any new content, update that content into that new design, that means that yet again, it’s another way you can start to measure what those changes do. And then finally, if you’re happy with that and you are going to a new domain, then do that as the very last thing. By chunking that up into those different stages, you’ll be able to spot any massive performance decreases.
This is something on a lot of the migrations I’ve worked on, when that hasn’t been done, that’s when things go really wrong during a re-platform process.
Mike Weir: That’s really interesting because from a DX perspective, again, we would want to see things tested. So, if someone’s going to test a design, we’d want to test a design before it went live. We’d want to test the content before it went live and try to understand what the pitfalls are. So, actually I suppose that speaks to the idea of like testing things from both angles,
from a Google angle and from a person angle to make sure you get the best possible approach for everyone.
We spoke a little bit about the risks of SEO and the risk of migrating from platform to platform. What would you say the approach here is at Impression?
Charlie Norledge: So the first thing we need to do is get all of those people into a room. So that could be the client, their marketing team, what we often see is how developers on as early as possible, because if they are making changes during re-platform, we need to make sure that they’re there and involved. Once we all sit down, we can work on a project plan, put timelines against it, work out what’s changing and the really critical part about having Impression involved at that point is that we can really give our own milestones and make sure we’re involved
at each step of the process. We frequently have clients that come to us and they’ve already gone through a migration process and we’re coming in after that’s happened.
So we’re almost coming in right at the end, rather than being there at the start. So being there at
the start is really important. And then alongside that, we want to get detailed documentation. So we actually have a checklist here at Impression that we’ll work on as part of the start of that meeting. We’ll share that with the client, they’re either in-house or external dev resource that will be owned by Impression, that gives us a single source of truth to be able to work through that migration So that will cover any technical changes, making sure the site has a good experience, is it fast? Content changes, making sure that’s all covered off, that we’re not going to be making changes to content that don’t need to happen.
And then finally, we do have some UX considerations in there, which I’m sure you want to cover.
Mike Weir: Yeah, of course. I think that it’s just bringing the customer on the journey and along for the ride. I think it’s really important to have clear objectives. Like, you don’t want to be moving for a project and then sort of changing the scope of it as you go. I think it speaks volumes that lots of people come to you once they’ve already started because actually they’ve kind of already got it wrong to begin with and that’s why a lot of the time they want that help and actually it would be so much more cheap, effective and the right thing to do, to pre-plan it and to test some of that stuff.
In terms of the checklist, I suppose, a lot of that stuff has to do with making sure that you’re getting the right level of insights. So, every time you make a design decision, you make an assumption and you make a certain amount of risk. And if you haven’t tested those assumptions beforehand, you bake more and more and more risk into your project. So actually it’s about doing the right amount of homework and the right amount of research to answer all the questions so that when you are making those design decisions, you’ve got research and data to back it up.
Charlie Norledge: And I can’t tell you how difficult it is that when you’re picking a migration up after it’s been done, they’re asking you, “we’ve seen a ranking decrease, can you fix this from the re-platforming?” Trying to go back in a project it’s already happened already been paid for, it’s almost a fool’s errand. It’s almost impossible to do. You just have to make the best of it
you can, but it is a very challenging approach.
Mike Weir: Yeah, and also emotions are running high because these things we have like we have commercial outputs. I see this a lot in ecommerce, where a website was changed at some point and then a couple of weeks later they see a drop-off in conversion rate or drop off in our average order value, but they don’t know exactly, because they change so much all the time, they don’t know exactly when that occurred, and that’s a real problem because it’s I kind of like trying to un-bake a cake. You’ve put your ingredients in and now you’ve got this challenge where you’ve got to try and go backwards. You’ve got to try and sort of dissect back to the problem, rather than diagnose what could go wrong in the first place. Prevention is better than cure in that case.
Charlie Norledge: A great example of that is actually Impression. We went from Impression.co.uk through to Impressiondigital.com and made some changes during that time. When that move was made, we did see a pretty huge decrease in traffic that lasted around three months. But as we have the knowledge and know migrations and re-platforming take time, we stayed the course and we’re starting to reap the benefits now and we’re starting to see that increase in visibility again. But where a business maybe didn’t have that involvement from the agency at that point, they’d probably be pretty concerned if after a month they’ve had a huge investment in a website, they’re not seeing any increases.
Mike Weir: And that happens all the time. I always think people think when they
put the website live they’re going to see this sudden take-off in all the key metrics. Actually, these changes sometimes do take a while to bed in, if you’ve got a lot of returning users on your website, who are very stuck in the way that how to use it and you make a change. The website sometimes even something that is net positive in the future can initially have a hiccup. So it’s that analysis and making sure you run the experiments and getting the data.
