Establish what is happening
Record the intended time sequence
Choose a controlled set spanning several periods, including a year boundary where relevant. Record raw dates or period keys and their expected order independently. Compare those with displayed labels so an alphabetic month sequence or a text-sorted numeric label can be distinguished from incorrect underlying dates.
Trace the active sort field
Inspect the query order, grouped period key, selected sort, and presentation ordering. Find whether the report sorts raw time, a formatted string, or a measure. Check any saved-view state, since a previously selected value ranking may persist even after readers switch back to what appears to be a chronological chart.
Compare chart and table order
Inspect the same controlled result in a supported table or export. Determine whether ordering differs only during rendering or already in the returned data. Keep missing periods separate from misordered periods so the next investigation does not attribute a gap in the timeline to a sort rule that never received those records.
What to carry forward
End with the intended sequence, raw time keys, active sort field, and first stage that reorders them. State whether the reader expected chronology or ranking. The next owner should be able to distinguish text sorting, saved sort state, and missing periods without changing date values merely to make their labels appear in order.
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