Aglet

Prioritize distinct-count discrepancies by decision risk

A distinct count can drive capacity planning, cohort analysis, conversion ratios, or operational review. Prioritization should focus on whether the identity definition or population boundary could change a decision, while preserving ordinary differences caused by overlapping groups as documented behavior.

Decide where the work belongs

  1. Identify the decision measure

    Write the decision that uses the count and whether it needs unique entities globally, within a period, or within each group. Record any ratio or threshold that depends on the count, since a small definition change can alter its denominator or numerator.

  2. Assess identity and scope risk

    List unresolved identifiers, null handling, reused keys, filters, joins, and group overlaps present in the affected views. Separate a known semantic choice from evidence that the same entity is counted inconsistently under a single report contract.

  3. Choose the queue outcome

    Queue definition review when the entity or scope is disputed, metric repair when documented distinct logic is violated, or explanatory copy when the number is correct but grouped totals need context. Attach a named measure owner to the decision.

What to carry forward

Prioritize the issue when a documented unique-entity measure changes because identity or population scope is applied inconsistently. If the difference is the expected result of overlapping groups, queue a clarification with examples rather than treating arithmetic disagreement as implementation failure.

Technical background: Metabase 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