A website redesign checklist before you move URLs

The most useful redesign work often starts with a crawl of the existing site. Before layouts are approved, you need to know which pages already rank, which URLs receive links, which forms convert, and which pages carry business context.

That audit gives the redesign a map. Without it, the team may accidentally remove pages that search engines and customers still rely on.

URL changes are not automatically bad. Unplanned URL changes are. A clear redirect map should connect every removed or changed URL to the strongest new equivalent.

• Export the current sitemap, crawl data, analytics landing pages, and Search Console pages.
• Decide which URLs stay, merge, redirect, or return a deliberate 404.
• Test redirects in staging before the domain is switched.

A redesign should make performance easier to understand, not reset the business dashboard. Analytics events, form tracking, consent behavior, Search Console, and conversion goals need to be checked before and after launch.

A redesign often changes two things at once: the message on the page and the technical location of the page. That is risky because it becomes harder to understand what caused traffic or conversion changes after launch.

When possible, keep important URLs stable while improving content and layout. If URLs must change, record the old intent, the new destination, and the reason for the change. That gives the team a way to review migration decisions instead of guessing later.

Migration QA should not be invented after development. The checklist should exist while design is still being approved, because page structure, CMS fields, and navigation decisions affect what must be tested.

A practical QA list includes redirects, canonical tags, metadata, structured data, forms, analytics events, sitemap output, robots rules, locale alternates, 404 handling, and page speed checks on the pages that matter most.

• Test top traffic pages, top conversion pages, and pages with backlinks.
• Check forms, booking links, email links, analytics events, and thank-you states.
• Crawl staging and production with the same URL list so differences are visible.

A migration plan should include what happens if something goes wrong. That does not always mean rolling the whole site back. It can mean restoring a redirect rule, reverting a template, republishing a missing page, or disabling a script that breaks conversion tracking.

Write the rollback owner and first-response steps before launch. When launch pressure is high, a clear escalation path is more useful than a perfect plan nobody can execute.