# Performance metrics collected; response time p50, p95, error rate, and throughput visible in dashboard with baselines

- **Pattern:** `ab-000988` (`deployment-readiness.environment-configuration.performance-metrics`)
- **Severity:** info
- **Lifecycle:** active
- **Last modified:** 2026-04-18
- **Canonical URL:** https://auditbuffet.com/patterns/ab-000988
- **License:** CC-BY-4.0 — attribute to AuditBuffet Pattern Catalog (https://auditbuffet.com/patterns/ab-000988)

## Why it matters

Without response time percentiles and throughput baselines, every deployment is a blind release — you cannot distinguish a 20% p95 regression from normal variance, or confirm that a refactor didn't silently degrade performance. ISO 25010 performance-efficiency.time-behaviour and NIST AU-6 require that performance data be collected and reviewed. Teams without p50/p95 dashboards measure performance by user complaints, which means regressions are invisible until they're severe enough for users to notice and report.

## Severity rationale

Info because performance metrics are an observability gap rather than an active failure mode — the system works, but you cannot see whether it is degrading or whether deployments improve or worsen response times.

## Remediation

Instrument your app with Datadog RUM or a comparable APM to collect p50, p95, error rate, and throughput.

```bash
npm install @datadog/browser-rum
```

```ts
// src/lib/monitoring.ts
import { datadogRum } from '@datadog/browser-rum';

datadogRum.init({
  applicationId: process.env.NEXT_PUBLIC_DATADOG_APP_ID!,
  clientToken: process.env.NEXT_PUBLIC_DATADOG_CLIENT_TOKEN!,
  site: 'datadoghq.com',
  service: 'my-app',
  env: process.env.NODE_ENV,
  version: process.env.NEXT_PUBLIC_APP_VERSION,
  sessionSampleRate: 100,
  trackUserInteractions: true,
});
```

In the Datadog dashboard, create timeseries panels for: p50 and p95 response time, error rate (5xx%), and requests per minute. Set a 30-day baseline window. Enable deployment markers so each production deploy appears as an annotation on the chart.

## Detection

- **ID:** `performance-metrics`
- **Severity:** `info`
- **What to look for:** Enumerate every relevant item. Look for monitoring/APM integration (New Relic, Datadog, Grafana, AWS CloudWatch). Verify metrics are being collected: response time (p50, p95), error rate, throughput. Check if metrics are visible in a dashboard with baseline comparisons or trends.
- **Pass criteria:** At least 1 of the following conditions is met. Performance metrics are collected and visible in a monitoring dashboard. At minimum: response time percentiles (p50, p95), error rate, and throughput. Historical baselines are available for comparison.
- **Fail criteria:** No metrics collected, or metrics exist but no dashboard, or only basic metrics without percentiles or baselines.
- **Skip (N/A) when:** The project is not in production yet.
- **Detail on fail:** `"No monitoring dashboard found. Performance metrics are not being collected."` or `"Datadog collects response time but does not show percentiles (p50, p95)."`
- **Remediation:** Set up monitoring. Using Datadog or similar:

  ```bash
  npm install @datadog/browser-rum @datadog/browser-logs
  ```

  Initialize in your app:

  ```ts
  import { datadogRum } from '@datadog/browser-rum';

  datadogRum.init({
    applicationId: process.env.DATADOG_APP_ID,
    clientToken: process.env.DATADOG_CLIENT_TOKEN,
    site: 'datadoghq.com',
    service: 'your-app',
    env: process.env.NODE_ENV,
    version: process.env.APP_VERSION,
  });

  datadogRum.startSessionReplayRecording();
  ```

  Then in Datadog dashboard, create a dashboard with panels for:
  - Response time: p50, p95, p99
  - Error rate: % of requests with 5xx status
  - Throughput: requests per minute
  - Set time range to 7d or 30d to see trends

## External references

- iso-25010 performance-efficiency.time-behaviour
- nist AU-6

Taxons: observability

HTML version: https://auditbuffet.com/patterns/ab-000988