Fill in these first
- Site and topic: [e.g. a personal finance blog with 600 articles]
- Update name and rollout dates: [as announced, e.g. a core update from 2026-03-12 to 2026-03-28]
- Clicks and impressions by page, before and after: [paste two equal periods of Search Console data, e.g. 28 days before the rollout and 28 days after it finished]
- Same periods last year, if available: [paste, or none]
- Query data for the biggest losers: [paste top queries per page, before and after]
- Changes made on the site in the same weeks: [e.g. new theme, URL changes, content pruned, none]
- Analytics or tracking changes: [e.g. consent banner added, none]
Task
Decide how much of the drop is caused by the update and where it hit, then plan a response.
Output format
- Verdict in three sentences: update-related, mixed, or other cause, with the numbers behind it.
- Other causes checked: seasonality, tracking, site changes, indexing, a single lost feature such as a rich result; each marked ruled out, possible or likely.
- | Page group | Clicks before | Clicks after | Change | Average position change |
- The ten pages that lost most, each with the queries lost and the type of page now ranking instead, if given.
- Recovery plan in order: what to improve on the losing pages, what to leave alone, and when to judge results.
Rules
- Compare equal-length periods and wait until the rollout has finished before calling a verdict.
- Never recommend mass deletion or rewriting every page; act on the evidence for specific pages.
- Say clearly that recovery often shows only after a later update.
- Use only the numbers pasted.






