Hi there,
I wanted to share a conversation I had a few weeks ago.
A website owner, Mr. S., reached out to me in need of SEO support.
Mr. S. owns a website that had recently migrated. He had support from an SEO partner but felt there were gaps in how it was executed based on declining performance.
Although I didn’t have the bandwidth to support a migration cleanup, I did offer him advice; advice that I find is quite common when working backwards from a messy migration.
A migration that hadn’t adequately prepared for SEO ahead of launch through a migration checklist or other means.
“Messy Migration”
Let’s dig in…
In my email, I suggested he focus on 2 major areas to assess where the bleeding was coming from – Redirect Behavior and KPI review.
In my experience, understanding both redirect behavior and KPIs do a really good job at pinpointing gaps in performance, from there a plan of action can be taken.
Review Redirect Behavior
Obtain legacy URLs
I advised Mr. S. to ask his SEO team for a crawl of the legacy site – that is, a list of URLs from the previous site. A legacy crawl should’ve informed a redirect map but looking at a bare crawl will remove any biases built into a map.
Side note: crawl a website prior to a migration and save the data. At the time it’ll help you understand the makeup of the legacy site but, looking outward, it’ll also act as historical context if you ever need crawl data.
If Mr. S.’s SEO team does not have a crawl, they should attempt to access the URLs from a data aggregator such as Google Search Console, Google Analytics, or SEMrush / Ahrefs – something that gives everyone a snapshot of top URLs prior to the migration.
Examine the targets
Once they have their hands on a list of top URLs pre-launch that fed the redirect strategy, they should crawl the old URLs (Screaming Frog as my migration tool of choice.) and re-examine if they’re appropriately redirecting to their new locations – that is, they have a single 301 Moved Permanently redirect hop. I suggested this for a couple of reasons
- Multiple redirects layered on one another dilute search engine authority but also
- The targets of the redirects may be broken or resolving in a 4XX status error or a soft 404, wherein link authority isn’t leveraged properly
If the redirect was just a single 301 hop, I advised Mr. S. and his team to look at the resolving experience. If the redirect targets are a totally different location than the original (that is, if a product page is redirecting to the homepage, when the product is still live on the new site), they should be adjusted to be as relevant as possible.
If all is redirecting just fine, let’s examine KPIs…
Review KPI’s
I advised Mr. S. and his SEO team to look at their KPIs – reason being, looking at the whole can be overwhelming but pinpointing leaks can help you understand where to patch up.
What’s declining?
Is it rankings, traffic, revenue?
If rankings are in decline
If rankings were primarily affected, look at what has declined – which keywords or keyword groups?
Grouping / categorizing keywords is so significant in assessing issues. That is because it could be that certain keyword groups are in decline while others not so much. You can then pin point the decline to areas such as content, meta data or overall experience that had been altered or completely removed.
Historical ranking data also provides context around which URLs had been ranking and if they’re resolving appropriately post-migration (again, signaling that additional redirects may be needed).
What’s more, if all keywords and keyword groups were in decline, are search engines still able to see the site’s content? Is it possible the site is using JS and impairing visibility, or are (meta) robots allowing bots through, or was the sitemap submitted?
If traffic is in decline
If traffic was primarily affected, was it just organic traffic that was in decline or traffic from other channels too?
In the past, brands have come to me stating their traffic was in free-fall but as I dug in, it wasn’t necessarily the whole site. Rather it was only a portion of their site or isolated to another channel.
Which leads me to my next point…
Did redirects account for parameters – e.g. “?utm=123”? These parameter-based URLs are used across other channels but they have to be redirected as well – read here. Brands may neglect to account for the parameters in their redirects and because they’re seen as unique URLs, they error out.
If revenue is in decline
If revenue was primarily affected, was there an aspect of the new site that changed how users converted?
Where is the revenue pinch coming from? That is, is revenue across the site in decline or are they occurring in specific locations. Although I’m not an analytics expert, often, brands have tracking in place to understand where conversions fall off – I recommend looking at the site’s conversion path.
What’s more, have any of the below been altered substantially?
- Lead forms
- Checkout
- Overall UX
If any of the above were altered greatly, they have the potential to severely affect conversions.
Summary
As migrations can be pretty complex, there may not be a single nail in the coffin.
In worst case scenarios, declining performance can be attributed to one or several of the areas I mentioned above.
Best case scenarios, it’s a single facet that is driving declines.
More often than not, I have seen redirect mismanagement be the culprit, which is why I advised Mr. S. begin with redirects.
Don’t be too hard on yourself; SEO within website migrations can be very complex. I’ve had my share of rough website migrations as well but if you ever find yourself with a messy migration on your hands, I hope the above comes in hand.
All the best!
