Small teams often make one of two mistakes: no monitoring at all, or a giant monitoring stack nobody has time to operate.
Start with four questions:
- Is the service reachable?
- Is it behaving normally?
- Are errors increasing?
- Will the right person know before customers start reporting it?
A useful first monitoring layer can be surprisingly small: external uptime checks, structured application logs, a few service-level metrics, and alerts that have an owner and an action.
The test for an alert is simple: when it fires, does somebody know what to do? If not, it is probably noise rather than monitoring.