“Tell us about a successful site migration you have been through from start to finish”. This is a question I was asked almost a year ago. After a short moment of reflection, I had to admit that it is very unusual to be involved as an SEO consultant throughout the entire process. It is much more common to be involved in certain aspects of it. I have been involved in the planning of upcoming site migrations, done page mapping and followed up afterwards – but the most common thing happening is that I, as an SEO expert, come in when it is a little too late. The majority of the projects I have been involved in have been when changes to the website have already been done – and the management panics when something has gone wrong and the traffic has started to decline.
In short, site migration is when changes to the website are made. This results in the search engines having to find the new pages and understand that they have replaced the pages that previously existed, as well as replace them in their index so that they can appear in the search results. From an SEO perspective, it is not just about changing a technical platform, it can also be about a new site structure, URL structure or moving to a completely new domain.
A couple of years ago, I wrote a debate article in Dagens Media where I said that it is entirely possible to do a migration without losing traffic. Unfortunately, this is information that still has not reached everyone. A year later, I sat again in a meeting where a web agency claimed that “it is completely normal to lose traffic right after a migration”. I had to bite my tongue. The problem was not a small dip in traffic during the time that Google found and added the new pages to the search results. The problem was that about half of all pages that previously drove organic traffic now redirected the visitor several times before finally landing on a URL that said the page did not exist. A lot of content that had previously driven visitors had been deleted, and features that Google could not read had been added. There was thus no chance that the new pages would quickly appear and regain the rankings in the search results. At the time, we listed a number of measures to resolve the issue – but I’m not sure if they have recovered yet.
In the article from 2018, I described problems that arise when SEO is taken in too late, there is no plan and there is a lack of understanding of the skills needed to be able to carry out the work. However, I feel it is time to clarify this. It’s not just about having a plan and the right skills, it’s also about organizational structure and the ability to actually get the work prioritized. It’s about people.
“I did not know it took so long …”
Some of my colleagues have been involved in successful migrations, from start to finish. In one case, the organic traffic even increased almost immediately afterwards. In that particular case, the two SEO experts that were on-site (one full-time, one part-time), had direct access to the board – and it took two years. Afterwards, a comment from a leader in the organization was “I did not know that it took so much time…”.
“There are a number of recurring mistakes that are made during site migrations, several of which can be very expensive if not handled.”
There are a number of recurring mistakes that are made during site migrations, several of which can be very expensive if not handled. It is important to be aware of them – and of the increased competitiveness of the organization if they are handled properly.
1. Lack of understanding of time spent and workload
As a consultant, it is unfortunately still a bit too common to hear “We will move the site to a new platform” in two weeks – could you spend a couple of hours checking that we did not miss anything?”.
“Organic traffic often accounts for a majority of the total traffic – and a poorly planned migration threatens the organic traffic.”
The understanding of how much time a site migration requires and how much risk it poses to online sales is still low in many organizations. The truth is that organic traffic in many cases accounts for a majority of the total traffic – and a poorly planned migration threatens the organic traffic. Buying that same amount of traffic is expensive and will affect the company’s financial performance.
2. The SEO role is placed in the wrong department
Many IT organizations have understood the need to include UX in development projects, preferably one in each team. If we do not build sites that are adapted for users, we will not sell either. Unfortunately, the understanding of SEO has not come as far. Many companies have an SEO expert but lack the understanding of what the role entails and what responsibilities it should have.
The SEO role is often difficult to place and has a tendency to move around between different departments. During my time as a consultant, I have been part of the content team, web analysts, sat with performance marketing (paid traffic) and most recently with UX. I have even been part of an agile team where everyone had their own specialist area, which was fantastic in several ways, but like everything also had its disadvantages.
Personally, I think the UX department has been the best match, as the user experience is such a big part of the SEO work. In this particular case, the UXs are also part of various development teams, which gives me insight into what is going on and where my expertise will be needed.
As the understanding of SEO increases, I would also like to see the role evolve from working reactively – where the expertise is taken in too late and too much time is spent correcting already developed functions – to acting as an advisory. The SEO expert should be included earlier in the development work, to discuss the pros and cons of different technical solutions.
3. Google is not seen as a visitor
Releases are often made in several stages during migrations; a certain proportion of visitors are released to the new site to see that nothing unforeseen happens. Unfortunately, the same thing is not done for Google. When the human visitors have reached 100%, you wait a while and then release the entire site to Google in one go. If any step in the migration has been missed, it will not be noticed until the entire site has been moved and it is already too late. You forget to see Google as a visitor which means you cannot make sure that the content is available to search engines.
“If any step in the migration has been missed, it will not be noticed until the entire site has been moved and it is already too late.”
Instead of doing this type of “big bang” launch, search engines should instead be seen as users to continuously test and evaluate new functionalities toward, to ensure that everything works as intended.
4. It is not obvious that it is a site migration
There are several different types of migrations and sometimes it is not entirely obvious that this is what it is. Changing the technical platform is an obvious change where you renew and develop new functionality. In the same way, a change of brand and domain means extensive work with brand-building efforts and redirection of the old domain to the new one.
But there are other changes that are less obvious forms of site migrations. Google representative John Mueller recently published a Twitter post in which he clarified:
“The SEO effect of keywords in the URL is minimal once the content is indexed. Make URLs that work for your users, not for SEO. Also, changing URLs on an existing site is a site-migration & it will take time/fluctuations to be reprocessed, so I’d avoid that unless it’s critical.”
This means that if the content department decides to restructure information on the site and at the same time change the URL structure, it is also a site migration.
5. There is an understanding of the recommendations, but not their importance
No one would be happier than me if it were possible to make an estimate and say “this is how much you will benefit from implementing the changes”. It would make it easier for both the product owner and me to evaluate and prioritize the initiatives in the backlog. Unfortunately, this is almost impossible, as we do not work in a vacuum. Even if we do our best to make the site as good as possible, we do not know what the competitors or the search engines are doing at the same time.
Various types of recommendations will also come into play and it is important to understand the differences. If I recommend entering structured data on product pages to get price and ratings in the search results, then there are studies that show that CTR (click-through rate) increases by a certain number of per cent, depending on what ranking in the search results each page has. Based on that, it is possible to make an estimate of how much more traffic the site will get.
If I instead recommend removing three-quarters of the site because it is duplicated, it is to increase the quality of the site as a whole and remove an obstacle that negatively affects the site’s positions at present. This is not about an improvement based on the current value of the site – this is about ensuring that there is a solid foundation to stand on. This type of fundamental problem needs to be addressed, preferably as soon as possible, as they otherwise risk sabotaging the other work that is being done on the site. This means that it is almost impossible to estimate the effect of the recommendations. We have no baseline (unless the duplication is something that has been done recently and we can see a decrease in traffic since launch) and we do not know how the pages would rank without the duplication. What I can say is that it should be prioritized to give further changes the maximum effect.
“Understanding and acting on this provides greater flexibility and greater competitiveness in a highly competitive market.”
Organizations that understand and have a plan to deal with these challenges will have a greater chance of performing a site migration without losing organic sales. In addition, placing the SEO role where it is available to developers, seeing search engines as visitors and understanding the difference between improvements and issues will help the organization deliver faster and with higher quality. Having an understanding of and acting on this provides greater flexibility and higher competitiveness in a highly competitive market.
Would you like to read more articles?