Connect the evidence
Check that requests, actions and outcomes can be followed through stable references. Identify missing files, broken links and ambiguous identities. Export success is not the same as a reconstructable history.
Measure against a declared scope
Specify the records eligible for each check and explain how samples were selected. Report denominators, exceptions and unmeasured areas. Do not turn a favorable selected example into a claim about the entire company archive.
Keep quality and permission separate
A clean, complete dataset may still be restricted. A cleared dataset may still lack usable outcomes. Evaluate both dimensions, version the findings and revisit them when the underlying records or proposed use changes.
Practical guidance, not legal advice or a buyer specification. Examples are illustrative. Reviewed October 20, 2026.