Build a useful investigation brief
Build the ranking fixture
Create categories with clear top members, a boundary tie, a small member, a null or unknown value, and an empty category when supported. Record filtered rows, aggregates, expected order, N, tie rule, Other members, and total reconciliation.
Trace ranking and grouping
Capture filter population, grouped values, sort order, tie resolution, Other assignment, total, color or legend mapping, and serialized export. Compare each stage with the independent fixture. Mark the first point where a member leaves the top set or totals diverge.
Test one ranking assumption
Change only N, direction, tie handling, or null grouping and rerun the fixture. Compare membership, labels, totals, chart, table, and export. Treat an Other-bucket policy as a hypothesis until the report owner confirms whether hidden members may be aggregated.
What to carry forward
The investigation is ready when the brief names the category fixture, first membership divergence, tie or null rule, affected output, and tested alternative. Recommend a scoped ranking or explanation change. Keep the result open when bucket membership is not observable.
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