Analytics Pulse

Pulse is the Analytics landing page at /analytics. It summarizes activity and review signals for the active account's selected repositories and timeframe.

Opening Pulse

Open Analytics and select Pulse. Check account/team/repository scope and the timeframe before comparing values. The platform's main Overview page is a different destination.

Metric Cards

The current Pulse includes PRs merged, Median cycle, First review, Review coverage, Open PRs and Critical findings, plus summaries such as Merges per day, Security findings, Review health, a Brief and Developer usage where data is available.

Health Status

Review status and risk indicators summarize underlying data. Time-based thresholds can be configured in Analytics Settings, so the old universal good/warning/critical tables are not fixed rules for every account.

Risky Pull Requests

Open Flow (/analytics/delivery) for delivery detail, stale/risky PRs, review bottlenecks and review-quality signals. Follow the listed PR to inspect its actual change and reviews before deciding what to do.

Charts and Verification

Use timeframe controls consistently when comparing charts. Missing timestamps, unavailable provider data or insufficient history can leave a card empty. Check the source PRs and selected scope rather than treating a missing metric as zero.

First Investigation from Pulse

  1. Confirm the active account, team, repositories and period.
  2. Read the merged/open PR counts alongside timing and review-coverage signals. Counts describe activity; timing and coverage describe different aspects of the available history.
  3. If a timing or review-health indicator looks unusual, open Flow and inspect the underlying PRs or bottlenecks.
  4. If a security indicator needs investigation, open Analytics Security and inspect the recorded scan and finding evidence.
  5. If a contributor signal needs context, open Developers and review the work behind it.

For example, a lower merge count with a longer median cycle does not establish that the team worked less. The selected period may contain fewer completed PRs, a change in repository scope or unusually long changes. Check those factors and the source changes before deciding on an action.

Compare like periods with the same repository selection. If a card has no value, investigate missing history or unavailable data instead of treating the absence as a healthy or unhealthy result.