Launch is the start of the risky part, not the end. Monitoring is how you find out whether the migration did what the plan said it would.
Quick answer
On launch day, check robots.txt, noindex, canonicals, redirects and tracking. In the first week, re-crawl the live site against your baseline and watch Search Console for errors. Over the following weeks, track indexation, impressions, clicks and priority pages. The aim is to separate normal fluctuation from fixable faults, quickly.
You need a baseline
Monitoring only works if you know what the site looked like before. Before my own migration I saved a crawl of every live URL with its status, title, description, H1, canonical and schema. Every post-launch crawl is compared against it. See the pre-launch checklist.
Day 1: launch day checks
- The live robots.txt allows crawling and has no staging rules
- Key pages return 200, have no noindex and canonicalise to themselves
- Old priority URLs redirect in one hop to the right page
- Analytics and conversion tracking are recording
- The new sitemap is reachable, and lists only final URLs
Week 1: re-crawl and compare
Crawl the live site and compare it with the baseline. Look for:
- URLs that have disappeared, or now return an error
- Changed titles, headings, canonicals or noindex tags
- Redirect chains and loops
- Internal links that point to redirected or broken URLs
Check Search Console for new 404s, server errors and "page with redirect" entries, and review the Pages report for unexpected exclusions.
What I found on my own site
Even with a careful plan, a post-launch review found things. Links across the site pointed at `/category/` archive URLs and an old `/migration-support/` URL that no longer existed. They had been broken on the old site as well, but the migration was the point at which I checked every link, so they surfaced. They were fixed with permanent redirects to the relevant pages, and the internal links were updated to point at the final URLs.
The Search Console coverage report for 15 October showed 55 indexed URLs, 47 not indexed and 4 pages marked as redirects. The redirect entries were expected. The unindexed ones needed investigating, and some turned out to be pages with weak internal links. See how to check if Google can crawl and index your site.
Weeks 2 to 4: indexation and visibility
- Are priority pages indexed, and is the right URL showing in Google?
- Are impressions and clicks for key pages broadly in line with before?
- Have any pages lost their rankings, and is there a reason?
- Are rich results and structured data still being picked up?
- Is the sitemap being processed without errors?
Some movement is normal while Google recrawls and reprocesses the site. A sharp drop on an important section, or one that persists for weeks, is not.
Month 2 onwards: confirm and clean up
- Keep the redirect map in place. Do not remove redirects for at least a year.
- Review the pages that still have impressions on old URLs
- Update remaining internal links and external profiles that point at old URLs
- Compare traffic against the baseline, by section, not just in total
Use real URLs, not only dashboards
Dashboards help, but they can hide page-level problems. Keep a list of priority URLs, including high-traffic pages, pages with backlinks, key templates and commercial landing pages, and test them by hand after launch.
Common mistakes
- Only checking the homepage
- Waiting weeks before reviewing errors
- Ignoring redirect chains
- Missing changes to tracking
- Assuming all traffic movement is normal
- Removing redirects too early
When to get support
SEO migration support includes post-launch monitoring, not just the pre-launch plan, so problems are caught in the first days. For the mistakes behind most of these problems, see common SEO migration mistakes.
FAQ
How long should migration monitoring continue?
Intensively for the first two weeks, then regularly for at least three months. Keep redirects for a year or more.
Is a traffic dip always bad?
Some fluctuation is normal. Sharp or lasting drops on important pages should be investigated quickly.
What should I do first if traffic drops?
Check that the site is crawlable and indexable, then check redirects and canonicals on the pages that lost traffic, and compare them with the baseline.