This is a follow-up to our article on redesigns. There we covered how to refresh a site's design without losing search traffic. Here we go one level of risk higher: changing the platform itself. Moving from WordPress or a website builder like Wix to Next.js, replacing the engine of a custom-built site, a new URL structure or a new domain: search engines treat all of these as major changes, and a migration is far easier to get wrong.
Below is the order of work that lets you move a site to a new CMS without losing rankings, or with only a small temporary dip: what to collect before you start, how to build a 301 redirect map, what to check on switchover day and after it, how to handle a domain change as a separate step, and what to do if rankings drop anyway.
How a platform migration differs from a redesign
A redesign mostly changes what people see. Changing the site's engine changes what the search crawler sees: the page HTML, the URLs, how text is rendered, the sitemap, speed, error handling. Even if the site looks almost the same, to a search engine it can be a completely different site.
There are three kinds of migration, and they are often combined:
| What changes | Main risk | |
|---|---|---|
| New CMS | Code, templates, content rendering | Lost copy, meta tags, structured data |
| New URL structure | Page addresses and hierarchy | Pages without redirects, 404s |
| New domain | The address of the whole site | Slow consolidation, temporary dip |
The more you change at once, the higher the risk and the harder it is to find the cause if something goes wrong. So the first planning question is what you can leave unchanged this time.
Inventory: what the old site has
The key migration document is a spreadsheet of every page on the old site. The redirect map is built from it, and after launch it is what you check against to make sure nothing was lost. You need several sources to build it, because none of them shows everything.
- Google Search Console. Indexed pages, the queries the site appears for, and the pages that get clicks. Export reports over a long period to account for seasonality.
- Google Analytics. Organic landing pages, goal completions, leads by page. This shows which URLs bring in not just traffic but revenue.
- A crawler. A tool that walks the site like a search bot and collects every reachable URL, status code, title, description, H1 and canonical tag. This is how you find pages everyone forgot about.
- The sitemap and the CMS. An export of sections, products, articles and files, including pages that are not linked from the rest of the site.
- Backlinks. URLs that other sites, directories and partners link to. Their redirects matter most.
The result is a table where every old URL has its traffic, rankings for key queries and a value flag. Save it together with a snapshot of rankings and traffic before the move: without a baseline you will have nothing to compare against.
The 301 redirect map
A 301 redirect tells the search engine a page has moved permanently and helps pass the old page's accumulated signals to the new one. The redirect map is the same inventory table with an extra "new URL" column.
Rules for building it
- If a page stays the same in substance, keep its old URL whenever you can. When moving to Next.js you can define any URLs you like, so there is no need to force them into a new "clean" pattern.
- If the URL changes, the redirect points to the closest matching page: product to product, category to category, article to article.
- If there is no equivalent and the page is no longer needed, it deliberately returns a 404 or 410 instead of being hidden behind a redirect.
- Redirects are permanent, with a 301 status, with no A → B → C chains and no loops.
- Technical URL variants count too: with and without a trailing slash, with and without www, http and https, old URLs with parameters.
Technically, redirects can live on the web server or in the application itself. Where matters less than completeness and automated checks: run the full list of old URLs through a crawler and confirm each one returns the expected status and goes where the map says.
Content, meta tags and data migration
When a site changes CMS, what gets lost most often is not the design but the content. New templates render text differently, build headings differently, and some fields from the old system simply have nowhere to go.
- the title, description and H1 of well-ranking pages are carried over unchanged, or changed deliberately;
- category, service and article copy moves over in full and is present in the HTML right away, not after a click or a script load;
- structured data (breadcrumbs, products, organization, FAQ) exists on the new site too;
- images move over with captions and alt text, and old image and file URLs get redirects if they had backlinks or image search traffic;
- internal links in the copy are rewritten to the new URLs instead of going through redirects.
Migrating the catalog and data
For an online store or a B2B catalog, data migration is a project of its own. From WordPress or another CMS, products, attributes, categories and users are pulled via export or direct database access, and if the data lives in an ERP, the new site is connected to that sync directly. Website builders can be harder: they hold little structured data, and some content has to be moved by hand.
The key is preserving relationships: which attributes belong to which product, which category it sits in, what its URL used to be. If catalog filters used to generate indexable pages with traffic, the new site needs equivalents, or the site will simply vanish for those queries.
The staging site and switchover day
The new version is built on a staging domain that is blocked from indexing, ideally with a password rather than just a robots.txt rule. That way search engines will not find a duplicate of the site too early, and the team can test everything on real data.
A forgotten noindex on the live site can wipe out search traffic within days.
The switchover itself should follow a plan written in advance, on a low-traffic day and not right before a weekend, so the team can react to errors immediately.
- Final data sync. Orders, leads, products and articles added to the old site during development are moved to the new one.
- Removing the indexing block. The password, the robots.txt disallow and noindex tags are removed, and canonical tags and the sitemap point to the production domain, not staging.
- Switchover and redirects. The domain is pointed at the new server, and the 301 redirect table goes live together with the site.
- Quick check. A crawler runs through the list of old URLs, and forms, cart, checkout and customer account are tested by hand.
- Signal to search engines. The new sitemap.xml is submitted to Google Search Console, and recrawling of key pages is requested.
Do not delete the old site right away: keep a copy so you can compare content or quickly restore something that went missing.
Changing the domain: a separate move
If the site's address changes as well as its engine, that is a separate procedure with its own rules. The search engine needs to understand that the new domain replaces the old one and transfer the accumulated signals to it.
- page-by-page 301 redirects are set up from the old domain to the new one, not a single redirect to the homepage;
- both domains are added and verified in Google Search Console;
- the move itself is announced with Search Console's Change of Address tool;
- the old domain is renewed and stays under your control, and its redirects are never removed.
Checks after the migration
The first weeks after launch are for monitoring. Migration errors show up in webmaster tools before they show up as fewer leads, so check them regularly.
- 404 errors. Search Console reports and server logs reveal URLs that were missed in the redirect map. Add them as you find them.
- Indexing. Is the number of new indexed pages growing, are old ones dropping out, are any pages excluded as duplicates or because of noindex?
- Rankings and traffic by page. Compare against the pre-migration snapshot by section and key page rather than sitewide, adjusting for seasonality.
- Analytics goals. The Google Analytics tag is on every template, goals and events fire, and form leads reach the CRM.
- Speed and mobile. The new platform must perform at least as well as the old one, or the migration was not worth it.
What to do if rankings drop
Small fluctuations in the first weeks after a site migration are expected: search engines recrawl pages and consolidate old URLs with new ones. The signal to act is a sharp or sustained drop on specific pages. Then look for the cause in order, from blunt errors to subtle ones.
- Indexing. Is there a leftover robots.txt disallow, a noindex in a template, or a canonical tag pointing to the staging domain?
- Redirects. Run the list of old URLs again: any 404s, chains, or redirects to the homepage instead of the right page?
- Content. Compare the pages that dropped with their old versions: did copy, headings or blocks that the page ranked for disappear?
- Technical. Load speed, content served to the crawler, a working mobile version, no duplicates.
- Internal linking. Have important pages lost links from the menu, the catalog or the copy?
Fix what you find page by page and request a recrawl of the fixed pages. Rolling the whole site back to the old platform is almost never necessary: the cause is usually local.
How we handle migrations
For us as a web development agency, migrating a site is not "copy the pages to a new engine." It is a project with the same stages as a new site: analysis and inventory, structure with a redirect map, design, development and testing on a closed staging site, launch and monitoring. We build sites on Next.js, traditional CMSs and other stacks, and we choose the platform to fit the business, not the other way around.
If you are choosing a web agency for a migration, ask about the redirect map and the post-launch check plan before you ask about design. You can see the kinds of problems clients bring us in our website development projects, and read about the Sphere digital agency's approach on the about page.
FAQ
Can you move a site to a new CMS without losing any rankings at all?
Nobody can guarantee rankings won't move: search engines need time to recrawl the site and consolidate old URLs with new ones. But if you keep URLs where possible, set up page-by-page 301 redirects, carry over meta tags and copy, and check indexing after launch, fluctuations usually stay temporary, and a sustained drop points to a specific error you can find.
How long should 301 redirects stay in place after a migration?
Ideally, forever. Backlinks, bookmarks and mentions in third-party directories keep pointing to the old URLs, and once the redirects are switched off, that traffic turns into 404s. Maintaining a redirect table costs almost nothing.
Can you change the CMS and the domain at the same time?
Technically yes, but each change adds uncertainty, and if rankings drop it will be hard to tell what caused it. When the timeline allows, it is safer to move to the new platform on the existing domain first, wait for things to stabilize, and then change the domain separately using the Change of Address tool in Google Search Console.
What should you do if pages drop out of the index after a migration?
Check that they are not blocked in robots.txt or by a noindex tag, that they return a 200, that they are in the new sitemap, and that their canonical tags do not point elsewhere. Then request a recrawl in Google Search Console and compare the page list with the inventory you made before the migration.




