Baseline
The list of URLs that bring traffic, have external links, or convert – from Search Console, analytics, a crawl, and link data. Without it, there’s nothing to compare after launch.
A new platform, a new domain, or a new URL structure is when organic traffic is easiest to lose – and hardest to win back. Before you switch, I record what brings traffic today, build the redirect map, and check the new version URL by URL. After launch, I measure whether everything carried over.
In short
An SEO migration is the planning and control of any change that affects how search engines see a site: a new domain, platform, URL structure, templates, or language versions. The goal is to carry traffic, links, and rankings over to the new version – through a redirect map, pre-launch checks, and post-launch measurement.
When it’s not enough
If the old site has serious technical problems or hundreds of pages it doesn’t need, a migration is a good moment to fix them – pair it with an SEO audit or a content audit instead of moving the problems to a new platform.
Scope
Six elements that decide whether traffic carries over. Each ends with a test you can run before and after launch.
The list of URLs that bring traffic, have external links, or convert – from Search Console, analytics, a crawl, and link data. Without it, there’s nothing to compare after launch.
A 301 from every old URL to its closest equivalent, with no chains and no blanket redirects to the homepage.
A crawl of staging compared with production URL by URL: titles, headings, canonicals, hreflang, structured data, noindex, internal links.
Whether content is visible without JavaScript, whether new templates hurt Core Web Vitals, and whether AI bots get the same content as Google.
A checklist: redirects work, robots.txt doesn’t block the site, the new sitemap is submitted in Search Console, the Change of Address tool is used for a domain move, analytics is collecting data, and the launch is annotated.
Traffic, indexing, and errors compared with the baseline at 2 and 4 weeks, with the URLs that lost the most and a cause for each. For large migrations, monitoring runs for 60–90 days.
How I work
We agree what’s changing, when it launches, and who implements. I point out the biggest traffic risks before the team starts building.
I record the pre-migration state: URLs with traffic, links, and conversions; rankings; indexing; performance.
The redirect map and SEO requirements for the new templates – as developer tasks, each with an acceptance criterion.
A crawl of staging, a comparison with production, and a redirect test on the full list. Launch only after blockers are closed.
A launch-day checklist, measurement at 2 and 4 weeks, fixes where traffic didn’t carry over, and monitoring for 60–90 days.
Post-launch review
Weekly clicks from Search Console against the pre-migration baseline. A short dip after the switch is normal – Google has to process the redirects.
The checks at 2 and 4 weeks tell a swing from a loss. If the curve doesn’t come back, I look for the cause URL by URL.
Redirects from the map working
100%
404s with traffic
−92%
Illustrative example – a shape typical of a well-run migration, not a specific client’s data.
What you get
The redirect map is a sheet where every row has the old URL, the new URL, the reason for the choice, and the test result. Plus a developer spec and a post-launch report.
Illustrative example – it shows the format, not a specific client’s data.
Honestly
Redirects fully testable
Every URL in the map can be tested before and after launch. There’s no room for “probably works.”
Traffic swings normal
For a few weeks after a big migration, rankings can fluctuate while Google processes the changes. The post-launch review tells a swing from a loss.
Launch decisions out of my control
If the launch date doesn’t leave time to close blockers, I say plainly how much traffic is at risk. Whether to launch stays your call.
The cheapest moment to bring SEO into a migration is the start of the project. The most expensive is a week after launch.
Pricing
Every migration is different, so there’s no fixed price. Four things affect the quote:
A company site with a hundred pages is a different scale from a store with tens of thousands of products and filters.
A CMS change with the same URLs is simpler than a domain and structure change at once.
Each language is its own map, its own hreflang, and its own post-launch review.
From the start of the project, just before launch, or after it – in the last case, I start by finding the losses.
After a short call, I send a proposal with scope, timeline, and price – before any work starts.
Technical health, content, links, and AI readiness – findings grouped by root cause, prioritized by business value, each with an acceptance criterion.
MoreGEO · AEOSo ChatGPT, Perplexity, Gemini, and AI Overviews can find, understand, and cite you – measured, not promised.
MoreContentA decision for every URL: keep, update, merge, or remove – with cannibalization, decay, and a redirect map.
MoreNot always, but the risk is real, especially when URLs change. Short ranking swings after a big change are normal. Lasting losses usually come from missing redirects, changed content, or blocked indexing – and those can be caught before launch.
At the start, when the URL structure and templates are being designed. Then SEO requirements are ordinary backlog tasks. Just before launch, all that’s left is checking and a list of blockers.
Every URL that has traffic, external links, or internal links – to its closest equivalent. URLs with no value and no equivalent can return 404 or 410. Google usually treats blanket redirects to the homepage like a 404.
Google says it can take a few weeks or more for new URLs to replace old ones in results, with ranking fluctuations along the way. That’s why I review at 2 and 4 weeks and, for large migrations, monitor traffic for 60–90 days. Redirects stay in place for at least a year, as Google recommends.
It depends on scale. The baseline and redirect map take from a few days for a site with hundreds of URLs to several weeks for tens of thousands. The staging check usually takes a few days, plus time to fix blockers. After launch comes monitoring for 60–90 days. We set dates at the start, in line with the development team’s schedule.
Developers are responsible for making the new site work. A redirect map based on traffic and links, a URL-by-URL comparison of staging and production, and post-launch checks are separate work that build projects rarely have time for. I work with your team, not instead of it.
Both codes say the page is gone. A 410 says it was removed on purpose, so search engines can stop visiting it sooner. I use 410 for URLs with no value and no equivalent, and leave 404 for accidental errors. Any URL with traffic or links always gets a 301.
I compare the pre-migration state – from Search Console and archived data – with today and find the URLs that lost the most. The usual causes are missing or wrong redirects, lost content, and indexing problems.
Implementation stays with your developers. I prepare the map and spec, test staging, and check production after launch.
Tell me what’s changing and when you plan to launch. I reply within one business day.
Ask about a migration