Most engineering teams don’t have a monitoring problem. They have a signal-to-noise problem — and the tools they’re buying to fix it are making it worse.
More dashboards don’t create more clarity. More alerts don’t create more safety. At some point, the observability stack stops being a tool your team uses and starts being a system your team maintains. That shift is costing more than anyone wants to admit.
I was recently featured in a piece on Legit Net Worth that digs into exactly why modern observability platforms are drowning development teams — and what it actually looks like to fix that.
Here’s what stood out from the conversation:
The monitoring market has been optimized for comprehensiveness, not usefulness. Vendors compete on how many metrics they can ingest, how many integrations they support, how many data points fit on a single screen. Teams buy in expecting control. What they get instead is a second job — tuning thresholds, muting alerts, and building informal Slack channels because nobody trusts the official dashboards anymore.
The core principle I kept coming back to: “The goal isn’t to track everything; it’s to know what matters and why it’s happening.” That distinction rarely gets made during procurement. It shows up later, after the rollout, when alert fatigue has already set in and engineers are ignoring the signals they’re supposed to act on.
Alert fatigue isn’t a discipline problem. It’s an architecture problem. When every alert arrives in the same format — whether it’s critical or background noise — the system trains engineers to stop caring. The one that actually needs attention lands the same way as the hundred that didn’t. That’s not a people failure. That’s a design failure.
What the best teams do differently is simple, not sophisticated: they define what a degraded state looks like before they instrument it. Alerts map to runbooks. Runbooks map to decisions. Anything that doesn’t lead to an action gets removed. Observability configuration lives in the same pull request as the code it covers.
If you’ve ever wondered why your team’s monitoring feels like a liability instead of an asset — or why oncall rotations create dread instead of confidence — this piece is worth your time.
Read the full article on Legit Net Worth →
The measure of a good observability tool isn’t what it can capture — it’s how cleanly it disappears when things are working and how clearly it speaks when they’re not.
