How we handle incidents

When something goes wrong, we follow a consistent process.

Detection

We track application health, error rates, and payment-webhook delivery. The on-call engineer is paged on threshold breach.

Triage and communication

1. Confirm the incident scope (which customers, which surfaces).
2. Post an internal status update.
3. If customer-impacting, send notification to affected customers within 60 minutes via email.
4. Begin remediation.

Post-mortem

For any incident lasting longer than 30 minutes or affecting more than one customer, we publish a post-mortem within 5 business days. The post-mortem covers root cause, timeline, and what we are changing to prevent recurrence.

Get notified

Email [email protected] with subject "Subscribe to incident notifications" and the domains or accounts you want notifications for. We add you to the proactive alert list.