DEV Community

Pavan
Pavan

Posted on

The Part of CMS Migration Nobody Warns You About: Post-Launch Metadata Chaos

You just finished migrating 400 pages from WordPress to HubSpot. The redirects are mapped. The DNS is updated. The team is celebrating.

Then you open Google Search Console three days later and realize half your pages have duplicate meta descriptions, 87 blog posts lost their featured image alt text, and someone forgot to update the Open Graph tags on your entire /resources section.

Welcome to the part of CMS migration that nobody puts in the project plan.

I have been through four CMS migrations in the past six years. Two were WordPress to HubSpot, one was Drupal to WordPress, and one was a Squarespace to HubSpot move that still gives me stress dreams. Every single time, the migration itself went fine. The data transferred. The pages loaded. The 301 redirects worked.

The disaster was always what came after.

Why Post-Migration Cleanup is the Real Project

Most migration guides focus on the move itself. Map your URLs. Set up redirects. Test your staging environment. Back up everything. That part is well documented and relatively straightforward if you follow the checklist.

What those guides skip is the metadata entropy that happens during any platform switch. Every CMS stores and renders metadata differently. WordPress might auto-generate your og:description from the first 160 characters of your post body. HubSpot expects you to set it manually in the page settings. Drupal handles canonical URLs one way. Webflow handles them another.

The result is that your content technically migrated, but the metadata layer that search engines and social platforms rely on is either missing, duplicated, or just wrong.

Here is what I typically find when I audit a site one week after migration:
Title tags that are either blank or populated with template defaults. Meta descriptions that were pulled from the wrong field during import. Featured images that transferred but lost their alt text entirely. URL slugs that gained extra characters or lost hyphens during the conversion. Internal links pointing to old URL structures that somehow bypassed the redirect map.

None of these are catastrophic on their own. But across hundreds of pages, they compound into a measurable traffic drop that teams blame on "Google catching up" when it is actually just sloppy data.

The Spreadsheet Phase Nobody Budgets For

After every migration, there is a phase I call "the spreadsheet phase." This is where someone (usually me) exports every page and blog post into a spreadsheet and manually audits the metadata, field by field, row by row.

For a 200-page site, this takes about three full days of focused work. For a 500-page site with an active blog, you are looking at a week or more. And this is just the audit. Fixing everything takes even longer because most CMS platforms force you to edit page settings one at a time.

HubSpot is particularly frustrating here. The editor is great for working on a single page. But if you need to update the meta description on 150 blog posts, you are clicking into each post individually, scrolling to the settings panel, making your edit, and publishing. Multiply that by every field that needs attention and you understand why teams either skip this step entirely or burn out halfway through.

This is where a bulk editing tool changes the equation. I recently started using Smuves for exactly this kind of post-migration cleanup. Instead of clicking through HubSpot page by page, you can export your entire CMS content to a Google Sheet, fix the metadata in bulk, and push it back. The time savings are not incremental. They are the difference between a one-week cleanup sprint and a month of half-finished tickets sitting in your backlog.

A Practical Post-Migration Audit Workflow

Here is the workflow I now follow after every migration. It is not glamorous, but it catches problems before Google notices them.

Step one: Export all pages, posts, and landing pages with their metadata fields. You want titles, meta descriptions, URLs, canonical tags, featured image URLs, alt text, and publication status all in one view.

Step two: Sort by content type and check for blanks. Any empty title tag or meta description is a priority fix. These are the pages most likely to show up poorly in search results.

Step three: Run a find-and-replace pass for old domain references. This catches hardcoded links in meta descriptions or Open Graph tags that still reference your previous domain.

Step four: Check for duplicate metadata. CMS imports sometimes apply the same default description to entire page groups. Sort your spreadsheet by the meta description column and look for repeats.

Step five: Validate alt text on featured images. This is the field most likely to get dropped during migration because different platforms store image metadata in completely different ways. The Smuves blog has a solid breakdown of what good page metadata actually looks like and why each element matters.

Step six: Push your fixes back to the CMS in bulk and verify a random sample of pages in the live environment.

The Redirect Audit You Probably Skipped

One more thing. Most teams set up their 301 redirects before launch and then never look at them again. But redirects break. Pages get renamed during post-migration edits. New content gets published at URLs that conflict with redirect rules.

Run a crawl of your site two weeks after migration and check for redirect chains (A redirects to B, which redirects to C), redirect loops, and any 404s that slipped through. This is not optional. It is the single highest-impact technical SEO task you can do in the first month after a platform switch.

The Takeaway for Your Next Migration

Budget for post-migration cleanup at the same level you budget for the migration itself. If your migration project plan has five phases and none of them explicitly cover "metadata audit and bulk correction," you are going to lose organic traffic you did not need to lose.

The move is the easy part. Cleaning up after the move is where the real work happens. And the teams that do it well are the ones using tools that let them work at scale instead of clicking through a CMS one page at a time.

Top comments (0)