What Good Observability Actually Looks Like

By Marta Ellison · September 3, 2026 · Operations

Dashboards multiply like rabbits and answer questions nobody asked this month. Good observability starts from the opposite direction: the alert fired, a human is awake, and the tooling should answer 'what changed' within a minute.

In practice that means a small set of high-cardinality traces on the request path, deployment markers on every graph that matters, and logs structured enough that you can pivot from a customer ID to a timeline without five grep commands.

If your incident retros keep ending with 'add a dashboard', your problem is not dashboards. It is that the signal you needed was never emitted in the first place.