Redesigns, replatforms, rebrands, and domain changes are where organic traffic goes to die — when they're unmanaged. I've led migrations that held every ranking through the switch, including full domain changes.
Get a Quote View Case StudiesSix years of compounding SEO for an Atlanta-area plumber — including a full domain migration that didn't drop a single ranking.
A full rebrand and domain change for an Indianapolis exteriors contractor, carried out without losing organic traffic, alongside growth from one office to three.
Website migration SEO is the practice of carrying search equity intact through any change that alters your URLs, your platform, your domain, or your site architecture. The goal is not to make the new site prettier than the old one on launch day. The goal is to make sure search engines recognize the new site as the same trusted entity, so the organic traffic, search engine rankings, and link equity you already paid for survive the switch.
Most teams treat a redesign or a replatform as a design and development project with an SEO review bolted on at the end. That sequencing is where traffic drops come from. By the time a staging site is ready for review, the url structure is already decided, the content migration is already scoped, and the redirect budget is already spent. What is left is damage control.
An seo website migration run properly starts before the first URL changes. It treats every ranking page as an asset with a measurable value, maps that asset to a destination, and verifies the handoff. Handled that way, a site migration is a routine engineering exercise. Handled reactively, it is the fastest way to erase years of search engine optimization work in a single deploy.
I have led migrations on both sides of that line. The two documented below held every ranking. I have also been called in after launch to diagnose why organic visibility fell off a cliff, and the answer is almost always the same: nobody mapped the old URLs before they stopped existing.
Not every project needs a migration plan. These six do, and they routinely arrive disguised as something smaller.
Moving between content management system platforms, for example WordPress to Shopify, or a custom build to headless. A new CMS almost always changes url structure, template markup, and internal links at the same time, which is three migrations wearing one ticket.
A move to a new domain, including the shift from a long legacy name to a cleaner one. This is the highest risk category, because every backlink you have ever earned now points at an address that has to redirect correctly and permanently.
A new company name usually means a new domain, new messaging, new landing pages, and a new Google Business Profile. Rebranding stacks a domain migration on top of a content migration and an entity change.
A visual refresh sounds safe until the new templates turn out to have dropped the H1s, halved the body copy, and moved the internal links into a component that search engine crawlers never execute.
Reorganizing site architecture, consolidating thin pages, or changing a category hierarchy. Even when the domain stays put, a site structure change redistributes link equity and can manufacture duplicate content.
Going http to https, or folding a subdomain into a subfolder. These look like infrastructure tickets. Search engines treat them as a different site.
You cannot prove a migration succeeded if you never measured what you had. Every engagement starts with a benchmark that becomes the scoreboard for the following six months.
I pull a full crawl with Screaming Frog to capture every indexable URL, its title tags, meta descriptions, canonical tags, and status codes exactly as they exist today. That crawl is the master inventory. Alongside it I export current keyword rankings and the backlinks profile from Semrush and Ahrefs, then pull the last thirteen months of sessions, conversions, and revenue by landing page out of Google Analytics.
Google Search Console supplies the piece a crawl cannot: which of your indexed pages actually earn impressions and clicks, and for which queries. That distinction matters, because most sites have far more pages than earning pages. Knowing which URLs carry the organic traffic tells you where to spend redirect and QA attention, and where a consolidation is the better call.
The output is a prioritized inventory. Every URL gets tagged by the traffic it earns, the backlinks pointing at it, the conversion rates it produces, and whether it is a stakeholder priority. That one sheet drives every decision that follows, and it is the artifact that lets you answer "did we lose anything" with data instead of a feeling. It is the same discipline behind a standing technical SEO audit, applied to a target that is about to move.
URL mapping is the heart of the work. Every URL in the old inventory gets exactly one destination on the new site, chosen deliberately.
Most map one to one. Some consolidate, where three thin pages become one strong page. A few are retired on purpose, and those should return a 410 rather than a soft redirect to the homepage, which search engines read as a soft 404 and discount anyway.
The rules I hold to:
Once the map is built I test it against the old crawl file, so every URL that existed is confirmed to resolve in one hop to a 200. That test runs again on staging, again within an hour of launch, and again a week later to catch the broken links and 404 errors that only surface under real traffic.
The staging environment is the last place a mistake is cheap. I run the same crawl against the staging site that I ran against production, then diff the two.
What the diff catches, in rough order of how often it bites:
A crawl diff will not catch everything, so the staging pass also includes a human review of the templates that earn the most revenue. Redesigns tend to improve user experience in ways that quietly cost rankings: an accordion that hides the body copy behind a click, a filter that generates thousands of crawlable parameter URLs, a hero image that pushes the answer below the fold. Usability and search visibility usually point the same direction, but where they conflict the tradeoff should be a decision somebody made on purpose, not a side effect discovered in the analytics three weeks later.
Nothing launches until that diff is clean and signed off. On launch day the sequence is fixed: remove the noindex, submit the new xml sitemap, file the Change of Address if the domain changed, and start monitoring. Google documents the Change of Address tool as the supported way to tell Search that a site has moved to a new domain, and skipping it measurably slows how fast the move is recognized.
A migration is not finished when the site goes live. The first ninety days are when you find out whether the plan worked.
Expect a short indexing lag. Search engines have to recrawl the new URLs, process the redirects, and transfer signals, and that takes days to weeks depending on site size and crawl budget. A modest dip in the first fortnight is normal. A sustained decline is a problem, and the only way to tell the two apart is to have benchmarked properly.
What I monitor daily for the first two weeks, then weekly:
Reporting runs through a dashboard the client owns, comparing post-launch seo performance against the pre-launch benchmark on identical segments. If organic visibility has not recovered to baseline inside a normal window, the inventory says exactly which URLs lost ground and why, which turns recovery into a specific fix list instead of guesswork. That measurement layer is standard on every engagement, and it is the same GA4 and Looker Studio build I set up for clients who are not migrating anything at all.
One more surface worth watching now: AI search. Migrations reset the citations and entity signals that generative systems rely on, so search engine visibility inside AI Overviews and assistants can lag conventional SERPs by a further few weeks. That recovery is work I handle as a GEO consultant alongside the conventional monitoring.
Plumb Smart, an Atlanta-area plumbing company, moved from beplumbsmart.com to plumbsmart.com in December 2021. Full URL mapping, 301 redirects in a single hop, Change of Address filed, monitoring in place before launch. Traffic and rankings held through the switch, and top-3 keyword rankings went on to grow 533 percent. That partnership is now in its sixth year.
Allegiant Exteriors was ClearCoat Roofing until late 2023. A rebrand is harder than a straight domain migration, because the entity itself changes: new name, new domain, new Google Business Profile, and new messaging across every landing page. I managed the full digital transition with zero traffic loss, and the business went from one office to three by the following roofing season.
Both are published on the case studies page with the numbers attached, and both were built with my team at Point Source Marketing.
If you take one thing from this page, take the sequence. This seo migration checklist is the condensed form of the website migration process above.
Keyword research and broader seo strategy work resumes once the site is stable. Chasing new rankings while a migration is still settling makes it impossible to tell which change caused what, and that ambiguity costs more than the delay does. If the migration is part of a larger program, it plugs into content strategy and full-stack SEO consulting from there.
SEO migration is the work of preserving search engine rankings and organic traffic through a change to your URLs, domain, platform, or site structure. It covers benchmarking, url mapping, 301 redirects, staging QA, and post-launch monitoring. It is a distinct discipline from the design and development work it accompanies.
Map every existing URL to one destination before launch, redirect with 301s in a single hop, keep the metadata and structured data intact in the new templates, and verify all of it on staging against a crawl of the live site. Then monitor against a benchmark you captured beforehand. Losses come from skipped mapping, not from migration itself.
No. A brief indexing lag is normal while search engines recrawl and reprocess redirects, but a sustained decline is a sign something was missed. Both migrations documented on this page held their rankings through the switch. A drop is a defect, not a cost of doing business.
Before a single URL changes. The highest-leverage work, meaning benchmarking, url mapping, and redirect logic, happens pre-launch. Calling me after launch is recovery, not prevention, and recovery costs more.
The SEO workstream typically runs four to eight weeks before launch for a mid-sized site, plus ninety days of monitoring afterward. Large e-commerce catalogs and multi-domain consolidations run longer. The pre-launch benchmark and url mapping are the phases worth protecting when a timeline compresses.
Start with a full crawl, a rankings export, a backlinks export, and thirteen months of analytics by landing page. Freeze the URL structure decision until that inventory exists. Getting a staging environment stood up early enough for a real crawl diff is the single best preparation step a dev team can take.
Yes, including Plumb Smart's move from beplumbsmart.com to plumbsmart.com and Allegiant Exteriors' complete rebrand from ClearCoat Roofing. Rankings held through both.
Free 30-minute discovery call. Bring your goals and pain points — I'll bring the plan.
Request a Consultation See the Numbers