Every migration checklist says to crawl, map redirects, and launch. The migrations that go wrong at capable companies rarely skip a checklist item. They fail on what a checklist doesn’t test: a redirect that returns a 200 instead of a 301, a 302 left in place for years, and internal links that quietly rewire when the new platform ships.
After 100+ replatformings, consolidations, and restructures, I’ve watched the same failures repeat. In 25 minutes, I’ll walk through the real cases behind them, including a homepage that took 40% of clicks away from the right URL, and a site-wide 302 that cost rankings until it became a 301. You’ll leave with three things you can use on your next project:
A request-level way to verify redirects, so you see what Googlebot sees instead of what your browser shows you
A method for mapping redirects on large sites with no clear URL patterns, using vector embeddings for a first pass and people and rules for the final call
A way to score internal link equity on staging before launch, so you can fix lost and orphaned pages while it is still cheap
I’ll also cover how to monitor the traffic shift after launch, and where to compromise with developers and designers and where to hold the line.