01 / Promise
Name the outcome
The expected report, export, digest, or file reaches the right destination inside its scheduled window.
Report delivery monitoring
Operating spine
A rendered report is not enough. The proof path should end where the customer would look for the report.
01 / Promise
The expected report, export, digest, or file reaches the right destination inside its scheduled window.
02 / Signal
Open the run before query/render and close success only after provider accepted id, storage object, or delivery receipt exists.
03 / Drill
Make the provider or upload step fail while generation succeeds, so the alert proves delivery rather than cron execution.
04 / Incident
The incident should show report id, account id, provider response, output pointer, schedule window, and retry safety.
05 / Handoff
Route the owner to resend, regenerate, or notify customers with retained evidence of what did and did not happen.
The cron ran and the report rendered, but the email provider rejected the message or the delivery step never executed. Customers notice the missing report before your dashboard does.
Open a run before query/render work starts. Close success only after delivery evidence exists: provider accepted id, storage object, webhook receipt, or another customer-visible proof.
Attach account id, report id, schedule window, provider message id, output object URL, duration, and a short summary. Avoid full report contents unless the data is intentionally safe to store.
Send a generated-but-not-delivered test by forcing the provider step to fail. Confirm the incident says delivery failed, not simply that a cron job missed a ping.
Need a concrete migration or monitoring pattern? Start with the docs, then adapt the payload to the evidence your operator needs.
Open integration docs