Aglet

Investigate report cache freshness with timestamp evidence

A useful freshness investigation follows one value through its source event, data availability boundary, query result, and rendered report. The goal is to identify the earliest boundary that explains the discrepancy, without treating a timestamp correlation as proof of cache behavior.

Build a useful investigation brief

  1. Choose a changing fixture

    Select a record or aggregate with a documented update and capture its source event time, expected period, and visible report value. Use a controlled example or an observed case with enough context to tell a missing event from a value that should not change.

  2. Build the timeline

    Create a timeline of source update, ingestion completion, query execution, cache write or read evidence when available, and page observation. Mark unknown times explicitly. Compare the report's stated as-of boundary with the event the reader expected to see.

  3. Test competing explanations

    Check the selected period, filters, joins, and source availability before attributing the result to caching. Reproduce the report at a different observation time if possible, and document which change would support or weaken each explanation.

What to carry forward

The investigation is complete when a reviewer can trace the value across known time boundaries and see which explanation remains supported. If cache read or write evidence is unavailable, report the observed freshness gap and keep the mechanism as an unconfirmed hypothesis.

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