Core Web Vitals: what each metric measures and how to fix it
LCP, INP and CLS measure different failures with different causes. Knowing which metric is failing tells you almost exactly where to look.
ZooBieDoo Infotech Engineering TeamRead
Key takeaways
- Crawl plus analytics plus Search Console — a crawl alone misses valuable pages.
- Redirect to the closest equivalent, never blanket-redirect to the homepage.
- Use 301 for permanent moves and avoid chains.
- Verify the whole map by script before cutover and watch 404s after.
Start with a full inventory
Before designing the new structure, crawl the existing site and export every URL that returns a 200. Then supplement it with sources a crawler misses: Search Console's indexed pages, analytics landing pages for the past year, and any URLs used in printed material or campaigns.
Pages with no internal links but real inbound traffic are exactly the ones a crawl-only inventory misses, and exactly the ones whose loss is noticed.
Map old to new, one row at a time
Build a spreadsheet with the old URL, the new URL and a decision. Most rows map one-to-one. Some map many-to-one, where several thin pages are consolidated. A few have no destination.
For that last category, redirect to the closest relevant page rather than the homepage. A homepage redirect tells search engines and users that the content is gone, and generally loses whatever value the page had.
Implement with the correct status codes
- 301 for permanent moves — this is what a replatform needs.
- 302 only for genuinely temporary situations.
- 410 for content deliberately removed with no equivalent.
Avoid redirect chains. Each hop loses a little and slows the response; consolidate so every old URL points directly at its final destination in one step.
Verify before and after cutover
Test the redirect map against a staging environment by requesting every old URL and confirming the status code and destination. This is a script, not a manual task, and it will find errors.
After launch, watch Search Console coverage and 404 logs daily for the first fortnight. Real traffic finds URLs your inventory missed, and the log is the cheapest source of that list.
Handling the awkward cases
A few categories reliably cause trouble. Paginated archives generate large numbers of URLs that rarely map cleanly; redirecting them to the first page of the new archive is usually acceptable.
Parameterised URLs from old campaigns, print pages and session identifiers can multiply the inventory enormously. Handle them with pattern-based rules rather than individual entries, and confirm the patterns do not catch anything you meant to keep.
Finally, decide about content you are deliberately removing. A 410 tells search engines the removal is intentional and stops them retrying, which is cleaner than redirecting unrelated content to an irrelevant page.
Frequently asked questions
Indefinitely where practical. Search engines eventually consolidate signals, but external links and bookmarks persist for years. Redirects are cheap to keep.
Some fluctuation for a few weeks is normal and usually recovers. Persistent loss almost always traces to missing redirects, changed metadata or a substantial reduction in content depth.
A 301 passes the large majority of ranking signals. Chains and irrelevant destinations lose more, which is why direct, relevant mapping matters.
