Findings and fixes
What happens after you mark a fix applied, and how Shruwd decides whether it worked.
A recheck is how Shruwd answers the only question that matters after you make a change: did it work?
Access findings do not need one. Their fix shows up in your own logs, robots.txt or page, and Shruwd resolves them when it does — see How findings work.
What happens after you mark a fix applied
- Your current numbers are captured as a baseline — the metric values, their intervals, the sample size and the timestamp. This is the "before" the recheck compares against.
- Shruwd waits 14 days. AI answer surfaces re-crawl and re-index on their own schedule. Rechecking on day two measures nothing and would report a false no sign of change.
- A recheck cycle runs on the prompts the finding affects, at nine repetitions per engine instead of the usual three. A small prompt subset needs the extra sampling for the result to be conclusive either way, and because the subset is small it is cheap. An engine that answered those prompts from memory — at least nine in ten of its answers citing no page — is left out, since no page change reaches it in weeks, and the recheck spends its repetitions on the engines that read pages: fourteen each on two, twenty-seven on one. An engine that was not answering those prompts through the whole 30-day baseline is left out too, since there is nothing before to compare it with: Google AI Mode, measured from 15 October 2026, for a fix marked applied before mid-November.
- The verdict appears in the finding's Closed tab.
- If it is rising but not proven yet, a second recheck runs two weeks later, and the two are judged together. There is no third.
Mark the fix applied only once the change is live. The baseline is captured at that moment, so marking it early compares your change against itself.
Reading the verdict
| Verdict | Meaning |
|---|---|
| Worked | The rate rose, and the rise passed both gates. The finding is resolved |
| Rising, not proven yet | The rate went up, but not by enough to rule out chance. Keep the fix: after a first recheck, a second one runs two weeks later |
| No sign of change | The answers after the fix look like the ones before. That is not proof the fix failed |
| The rate fell | The rate dropped, and the drop passed both gates. Check what the change removed |
| The model changed | The provider changed models between the two measurements, so the recheck cannot judge the fix |
Every verdict other than Worked shows as not proven in the findings list.
The prompts are compared together and one by one. Together is the fix's own claim — it was made for those prompts — and it is the comparison a recheck can usually decide: one prompt alone has few answers, so a modest real gain often comes back Rising, not proven yet. A fix aimed at one prompt still counts if that prompt moves on its own.
A finding the recheck could not credit re-opens after 30 days if the condition still holds.
Why a fix that looks like it worked is not called movement
The verdict uses the same rule as everything else in Shruwd: the change must be statistically significant, with its 95% range clear of zero, and at least five percentage points, after correcting for the fact that many prompts are being checked at once.
See When a number has moved for the full rule.
If the provider changed its underlying models between the baseline and the recheck, the comparison is marked model changed — the difference may be the new model rather than your fix.
Recheck limits
Rechecks are capped per month by plan.
| Plan | Rechecks per month |
|---|---|
| Crawl | None |
| Starter | 1 |
| Pro | 6 |
A second recheck of a rising result counts as one. If you are out, the finding waits until next month and says so on the fix panel. Spend them on findings where you actually changed something substantial.
Next steps
- How findings work — the full finding lifecycle
- When a number has moved — the movement rule
- Plans and limits — what your plan includes