Skip to main content
This guide is for whoever is setting Bilanc up. It lists every integration Bilanc supports, who needs to do it, what access they need, and the exact steps. You don’t need to connect everything. Connect the version control system you use, plus whichever ticketing, AI, and CI tools your teams rely on. You can always add more later.

How connecting works

Almost everything is done from one screen in the Bilanc dashboard:
  1. Sign in to app.bilanc.co as an admin.
  2. Go to Admin → Connections.
  3. Find the tool you want, click the + (Connect) button on its card, and follow the prompt.
Connections fall into two types:
  • OAuth (GitHub, GitLab, Bitbucket, Jira, Slack) — you click Connect and authorize Bilanc in a popup. The person clicking must be signed in to that tool with the right level of access (see each section).
  • API key / token (Azure DevOps, Cursor, Codex, Augment, Devin, Incident.io, CircleCI) — you generate a key or token in the other tool first, then paste it into Bilanc when prompted.
Claude Code and GitHub Copilot have a couple of extra options depending on how you run them — see AI Coding Tools.
Once connected, Bilanc backfills roughly the last 2 months of history automatically and then keeps syncing. There’s nothing to schedule.

What you’ll need at a glance

A note on permissions

Bilanc only ever requests read access. We pull metadata (pull requests, commits metadata, issues, usage counts, etc.) to compute metrics — we don’t write to your repos, tickets, or channels, and we don’t read source code beyond what’s needed for diff-size metrics.

Deployment & DORA data

Bilanc reports Deployment Frequency, Lead Time for Changes, Change Failure Rate and Time to Restore for every account. Which connections feed them:
  • Version control (GitHub, GitLab, Bitbucket, Azure DevOps) – required. Provides pull requests, published releases and Bitbucket deployment records.
  • CI/CD (GitHub Actions via GitHub, CircleCI, GitLab CI, Azure Pipelines) – production deploy workflow runs, detected by workflow name, plus failed deployments.
  • Incident.io – turns Change Failure Rate and Time to Restore from revert-based proxies into incident-based measurements.
Without CI/CD or incident data the metrics still show, using labelled proxies. See DORA Metrics for exactly how each one is calculated and how to improve it.

Need a hand?

If anything below is unclear or a connection won’t complete, email sam@bilanc.co or reach out on your shared Slack channel and we’ll jump on a call.