Aglet

Investigate a Report Ratio Denominator

A ratio investigation needs a controlled population with numerator events, eligible non-events, excluded cases, and an empty-eligibility boundary when supported. Reconstruct the formula independently, compare the report surfaces, and retain the difference between a coding defect and an unresolved definition.

Build a useful investigation brief

  1. Construct the population fixture

    Create rows that cover numerator success, eligible non-success, explicit exclusions, repeated membership, and a zero-eligible case where the report allows it. Record the intended numerator, denominator, ratio, and displayed wording for each scope.

  2. Trace eligibility and arithmetic

    Capture rows after filters and joins, eligibility flags, numerator contributions, denominator inputs, division, rounding, and final labels. Compare the independent calculation with card, chart, table, and export output. Mark the first stage that includes or removes the wrong population.

  3. Test a single assumption

    Change one eligibility rule, filter, join, or zero policy at a time and rerun the fixture. Compare rate movement and denominator explanation across surfaces. Treat a disagreement about who belongs in the population as a policy hypothesis until the metric owner confirms it.

What to carry forward

The investigation is ready when the brief identifies the population fixture, first denominator divergence, affected outputs, strongest evidence, and competing eligibility rule. Recommend a scoped metric or label correction. Leave the case open when the denominator is a policy question without an accountable owner.

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