All modules
MODULE 12 · KNOW IT BROKE BEFORE USERS DO
Observability
The 2am question: what broke, and where?
Error tracking tells you something broke; structured logs tell you why; an uptime check tells you it's down before a customer does. This card is boring right up until it saves your launch.
WHEN YOU NEED IT
Sentry from Tier 2. Structured logs and uptime checks at Tier 3.
THE CHOICE, PER TIER
- T1Vercel analytics + function logs
- T2Sentry on frontend and backend
- T3+ structured JSON logs and an uptime monitor
SAY THIS TO YOUR AGENT
"Add Sentry to frontend and backend with release tagging. Log every request as structured JSON with a request id. Point an uptime check at the health endpoint."
DONE WHEN
- A thrown error shows up in Sentry within a minute
- You can trace one user's failing request in the logs
- Downtime pings you before a user emails you