How it works
Four steps from signal to inbox.
01 · Pick a signal
Any signal errorgap sees
Errors, uptime, performance or security.
02 · Set a rule
Threshold and window
A condition, a duration and who gets told.
03 · Backtest
Check last week
See every time the rule would have fired before it can page anyone.
04 · Deliver
Where people look
Email, Slack, webhooks and the in-app inbox.
What’s different
Fewer surprises, in both directions.
Backtest
How often would it have fired?
Every rule runs against the last seven days before you save it. Too noisy, and you adjust the threshold first.
- Fires shown per day
- Adjust and re-run in place
Alert rules · acme-co
RuleWould fire
Error rate > 2% on POST /checkout/* for 5 min3×
p95 > 1.5s on GET /products for 10 min41×noisy
Uptime check fails twice1×
High security finding0×
Would have fired 3× last week
Mon
Tue
Wed
Thu
Fri
Sat
Sun
Recovery
Recovery, not just failure.
Uptime monitors send a recovery alert when a check passes again, with the downtime and the incident attached. Alert emails carry the errorgap brand and link straight to the incident.
- Uptime recovery alerts
- Branded alert emails
Inbox · ana@acme.example
errrgapRECOVERED
api.acme.example is back up
Down for 4m 12s, from 14:32 to 14:36 UTC. 3 checks failed in a row. The incident stays open until you resolve it.
Open INC-241
Per-person preferences
Each person picks their channels.
Notification preferences per user and event, an in-app inbox for everything, and a daily digest email for what doesn’t need to interrupt.
- Per-user, per-event preferences
- In-app inbox
- Daily digest email
Notifications · ana
EventEmailInbox
Incident opened
Uptime recovered
New SSH key
Security finding · high
Assigned to me
Daily digest at 08:00 Europe/Lisbon
Related capabilities