Site migration without losing rankings

Map every URL before anything moves

Crawl the current site and export every URL that exists, including ones nobody remembers. Then decide, per URL, whether it moves, merges or goes.

Redirect old to new one to one wherever a match exists. Sending everything to the homepage is the single most common cause of a migration collapse.

Keep the content, not a summary of it

Rewriting during a migration mixes two variables. When rankings fall you cannot tell whether it was the move or the words.

Move first, confirm stability, rewrite afterwards. It is slower and it keeps the diagnosis clean.

Baseline before, measure after

Record rankings, indexed page counts, traffic by page and profile position the week before. Without a baseline you cannot tell recovery from decline, and you will argue about it for a month.

Then watch weekly for six weeks. Some fluctuation is normal, and the shape of the recovery curve tells you whether anything is broken.

Check the boring things on launch day

Robots.txt not carrying a staging block. Analytics still firing. Structured data still present. Search Console updated. Internal links pointing at new URLs rather than through redirect chains.

A staging robots.txt shipped to production is the classic, and it can cost weeks before anyone notices.

The honest expectation

Even a well-run migration usually dips before it recovers. Anyone promising no movement at all has either not done many or is not telling you about the ones that went badly.

“Sending everything to the homepage is the single most common cause of a migration collapse.”