Build a useful investigation brief
Recreate the cohort conditions
Build a minimal test setup matching role, permissions, data shape, release, and workflow timing. Record the expected and observed outcomes and keep private customer data out of the reproduction notes. Mark any condition that could not be matched.
Probe adjacent exposure
Vary one condition at a time across nearby roles, account states, inputs, and versions. Record both failures and successful boundaries. A successful adjacent test narrows evidence but does not prove that every untested context is unaffected.
Separate cause and containment
Document immediate steps that prevent further harm separately from the mechanism hypothesis. Rank causes by evidence, state what would falsify each, and identify whether the small cohort reflects a real boundary or limited reporting.
What to carry forward
A useful brief explains the reproducible cohort conditions, severe outcome, tested exposure boundary, containment status, and remaining uncertainty. Stop when the team can act safely on the known scope; keep broader exposure as a named question rather than an implied conclusion.
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