If evaluation criteria drift between projects, start by preserving one representative example. It gives the investigation a concrete reference and makes the eventual correction easier to judge.
Find the likely cause
Review rules change without being recorded.
Treat this as an explanation to verify against the actual work. Look at the input, the relevant decision, and the final result together. If the evidence does not support this diagnosis, investigate the mismatch before applying a convenient but unrelated fix.
Make the targeted correction
Version the rubric and note task-specific exceptions.
A rubric makes quality review more consistent by turning vague preferences into observable requirements.
Check that the repair worked
Compare drafts only when their evaluation conditions are understood.
Repeat the check on the final version that the reader or customer will encounter. An approved draft, a preview, and a published result can differ; the acceptance decision should concern the version people actually use.
Prevent the next related failure
A separate issue to watch for is this: aI quality scores look precise but are not actionable. Require evidence and a correction for each failed criterion.
Monitor the useful outcome
Track agreement between reviewers and recurring defects missed by the rubric.
Keep a short record of the original symptom, the evidence behind the diagnosis, and the result of the acceptance check. That record makes the solution reusable when the same condition appears again, without assuming that every superficially similar problem has the same cause.