Start with the critical attributes
Choose what a meaningful change would look like for this product. A cracked dried piece and a minor change in label appearance may call for different responses.
When a threshold is appropriate, ask the reviewer how it was justified and measured. For qualitative attributes, keep representative examples and the review vocabulary aligned.
Write comparable decision rules
Link each criterion to a sample state, test or observation type. A result on a sealed sachet and a result from an opened one may serve different criteria.
Decide how you will handle conflicting results, missing data and findings close to a limit. Clarify who can decide whether a repeat or further review is justified.
Keep approval independent of the result
Have the relevant owner approve a version of the plan and criteria before results are reviewed. Record any later revision and its reason separately.
In the final report, show which criteria were assessed and which were outside scope. Reviewers then know what conclusion the study can support.
Frequently asked questions
Should every observed change fail a study?
Interpret the change against its agreed significance, the study context and the accepted criteria. Document uncertainty and additional questions explicitly.