Check whether the outcome improved
Exercise contrasting aggregation cases
Check an additive measure and the affected non-additive measure using small known datasets. Include unequal group sizes or overlapping identities where relevant. Compare both cell values and intended totals, keeping a naive visible-cell calculation as a contrast rather than using it as the expected answer for every pivot.
Check filters and hidden populations
Apply supported row and column filters, category visibility changes, and relevant limits. Record whether the total should cover displayed cells or the underlying filtered population. Confirm the result follows that declared contract and that its label or explanation remains accurate when the number no longer matches a casual sum of what is visible.
Compare detail and exported totals
Inspect any underlying-record view and supported export for the same controlled pivot. Confirm they communicate the same measure and population. Check a neighboring pivot that reuses the total expression so the correction does not change an already correct additive report while repairing the non-additive case that exposed the defect.
What to carry forward
Accept when cells, totals, labels, and supported detail or export paths agree with independent measure expectations. Record untested filter or limit combinations separately. Keep verification open if only visual-cell arithmetic was checked; that comparison cannot establish the correctness of a total whose intended measure is explicitly non-additive.
Technical background: Google Cloud documentation.
Keep the decision with the work.
Use a Work Item in Aglet to record the problem, the evidence you have, and the next decision. Add an owner and priority, then keep updates in the discussion so the next person can follow the reasoning.
Create an account See the product workflow