Observability vs Monitoring
Monitoring watches predefined metrics and alerts on known failure modes you anticipated. Observability is the property of being able to ask new questions about a system's internal state from its outputs, letting you debug unknown failures you never predicted…
WHY IT EXISTS Traditional monitoring assumes you can predict how a system will fail and instrument for those cases ahead of time. That assumption breaks down in modern distributed systems, where microservices, autoscaling, and complex dependencies produce novel failure modes nobody anticipated. Observability emerged to handle these unknown unknowns.
THE MENTAL MODEL Monitoring is asking known questions: is CPU above 90 percent, is the error rate above 1 percent. You decide the questions in advance and build dashboards and alerts for them. Observability is a property of the system: how well you can understand its internal state purely from its external outputs. A highly observable system lets you pose and answer new questions during an investigation without shipping new instrumentation.
HOW IT WORKS Observability is typically built on metrics, logs, and traces, often with rich high-cardinality attributes attached, such as user id, request id, and version. Metrics are cheap aggregates good for alerting; logs are detailed events; traces follow a request across services. Together with the ability to slice by arbitrary dimensions, they let you drill from a symptom to a root cause. Monitoring sits on top, consuming a slice of these signals to fire alerts on conditions you already know matter.
WHEN IT MATTERS For a simple, well-understood service, monitoring known thresholds may be enough. For a sprawling distributed system where failures are emergent and hard to predict, observability is what lets an on-call engineer investigate a brand-new symptom at 3am without deploying code first. The distinction matters most when the failure was never imagined.
ONE CONCRETE EXAMPLE Monitoring tells you that checkout latency crossed its alert threshold. Observability lets you then ask, unprompted, which downstream service, which customer tier, and which app version are responsible, by slicing traces and high-cardinality metrics until you find that one payment provider in one region is slow for users on the newest build.
Get five bites like this every day.
Tezvyn delivers a daily feed of 60-second tech bites with quizzes to lock in what you learn.