Build a useful investigation brief
Create distinguishable schedule variants
Define two controlled variants of the same report with different known filters or periods. Record expected values and labels independently. Keep generation and data timestamps distinct, and avoid relying on actual recipient delivery for the content comparison when the artifact can be inspected through an authorized preview or captured generation result.
Trace parameters into generation
Inspect the saved schedule definition, resolved defaults, applied filters, and data or cache freshness used for the run. Record the first mismatch with the intended variant. Check whether a mutable report default is read at generation time and changes the result without the scheduled definition or artifact explaining that context.
Compare artifact and destination meaning
Inspect titles, filter labels, units, timestamps, and any link to the interactive report. Determine whether the link preserves parameters or opens a current default view. Evaluate one implicated context change against both variants so the resulting snapshots become distinguishable without implying that a generated timestamp establishes fresh underlying data.
What to carry forward
Deliver the schedule-to-artifact trace, applied parameters, freshness evidence, and first lost or incorrect context. Keep delivery acceptance separate from content correctness. A supported correction makes the snapshot and its destination interpretable as the intended report instance rather than a generic image whose numbers can only be explained by its original author.
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