Glossary · Observability

What Are Percentiles (p95, p99)?

p95 and p99 tell you what your slowest users experience; the average tells you almost nothing useful about latency.

A percentile is the value below which a given share of observations fall. p95 latency is the value 95% of requests are faster than; p99 is the value 99% are faster than. Percentiles capture the tail of the distribution that averages hide.

Why averages mislead

An average latency can look healthy while a meaningful fraction of users suffer, because a few very slow requests barely move the mean. p95 and p99 expose that tail: p99 latency is the experience of your unluckiest one in a hundred requests, which is often where churn and complaints come from.

A common pitfall

You cannot average percentiles: the p99 of two services is not the average of their p99s. Percentiles must be computed from the underlying distribution (which is why latency is recorded as a histogram). Aggregating pre-computed percentiles gives wrong answers, a frequent and costly mistake.

How it fits Ops Singularity

The how-to dashboards on this site graph p95 and p99, not averages, and TelemetryOps preserves the distributions Sentinel AI needs to reason about tail latency.

Frequently asked questions

What does p99 latency mean?

The latency value that 99% of requests are faster than; only 1% of requests are slower, representing the tail of the experience.

Can I average p95 values?

No. Percentiles cannot be averaged; they must be computed from the underlying distribution, which is why latency is stored as a histogram.

One governed intelligence layer for every operation.

Ops Singularity turns open telemetry into autonomous, governed resolution. See it on your own stack.

Request a Demo → See TelemetryOps