Dashboards & Data Sources

GitHub Copilot provides multiple ways to measure adoption, engagement, and impact — from built-in dashboards with visual trends to raw data exports and REST API endpoints. This guide covers every native and API-based data source available for Copilot metrics, so you can choose the right one for your reporting needs.

Native Usage Metrics Dashboard

The Usage Metrics dashboard is the primary built-in view for understanding how Copilot is being adopted across your enterprise.

How to access

Navigate to your Enterprise account on GitHub.
Go to Insights → Copilot usage.

Policy prerequisite

The "Copilot usage metrics" policy must be enabled at the enterprise level before any data appears. An enterprise owner can enable this under Enterprise settings → Policies → Copilot.

What it shows

Metric Category	Details
Adoption trends	28-day rolling view of active users, seat utilization, and onboarding velocity
Engagement breakdown	Activity split by feature — completions, chat, agent mode, edit mode
Model trends	Which AI models are being used and how usage shifts over time
Language breakdown	Top programming languages by Copilot interaction volume
IDE distribution	Usage across VS Code, JetBrains, Visual Studio, Neovim, and other editors

Who can access

Role	Access Level
Enterprise owner	Full access
Organization admin	Org-scoped data
Billing manager	Full access
Custom role	Requires "View Enterprise Copilot Metrics" permission

What to do next

Verify the "Copilot usage metrics" policy is enabled in your enterprise settings.
Bookmark the Insights → Copilot usage page for regular check-ins.
Identify teams with low adoption and target them for enablement outreach.

Code Generation Dashboard

The Code Generation dashboard focuses specifically on how AI is influencing code output across your organization.

How to access

Navigate to your Enterprise account on GitHub.
Go to Insights → Code generation.

What it shows

Metric Category	Details
Lines of code changed with AI	Total lines added, modified, or removed where Copilot contributed
User-initiated activity	Code changes driven by developer-triggered completions and chat
Agent-initiated activity	Code changes driven by Copilot agent mode autonomously
Model breakdown	Which models generated the code (e.g., GPT-4o, Claude Sonnet)
Language breakdown	Lines changed per programming language

Permissions

The same permission requirements apply as the Usage Metrics dashboard — enterprise owner, org admin, billing manager, or a custom role with "View Enterprise Copilot Metrics" permission.

What to do next

Compare user-initiated vs. agent-initiated ratios to understand how teams are leveraging agent mode.
Track language-level trends to see where Copilot delivers the most code generation value.
Use this data alongside the Usage Metrics dashboard for a complete adoption picture.

NDJSON Export

The NDJSON (Newline-Delimited JSON) export provides raw, granular data from the dashboard for custom analysis in your own BI tools.

What it is

A downloadable file containing one JSON object per line, representing a single user's activity for a single day. This is the same data that powers the native dashboards, exported in a machine-readable format.

How to download

Open the Usage Metrics or Code Generation dashboard.
Click the Export button in the dashboard UI.
The browser downloads an .ndjson file.

What's in the file

Each line is a self-contained JSON object with fields including:

{
  "date": "2025-01-15",
  "user": "octocat",
  "feature": "completions",
  "interactions": 42,
  "acceptances": 28,
  "language": "python",
  "editor": "vscode",
  "model": "gpt-4o"
}

Field Category	Examples
Identity	User login, team, organization
Interaction counts	Suggestions shown, acceptances, dismissals
Acceptance activity	Lines accepted, acceptance rate per feature
Feature usage	Completions, chat, agent mode, edit mode
Context	Language, editor, model, date

Use cases

Tool	How to Use
Power BI	Import the NDJSON file as a JSON data source and build executive dashboards
Jupyter Notebooks	Load with `pandas.read_json('export.ndjson', lines=True)` for ad-hoc analysis
Data warehouses	Ingest into Snowflake, BigQuery, or Azure Synapse for long-term trend analysis
Excel	Use Power Query to transform NDJSON into tabular format

Use Copilot Chat for analysis

Upload your exported NDJSON file to Copilot Chat and ask questions like "Which teams have the highest acceptance rate?" or "Show me adoption trends over the last 4 weeks." Copilot can analyze the data and generate insights directly.

What to do next

Export your first NDJSON file and explore it in your preferred BI tool.
Set up a recurring workflow to pull exports weekly for trend tracking.
Combine NDJSON data with your internal systems (e.g., JIRA, Azure DevOps) for ROI correlation.

REST API Endpoints

For programmatic access, GitHub provides REST API endpoints that return Copilot usage metrics at various scopes.

Available endpoints

Endpoint	Scope	What It Returns	Key Parameters
Enterprise 28-day report	Enterprise	Aggregated 28-day metrics across all orgs	`enterprise` slug
Enterprise 1-day report	Enterprise	Single-day aggregated metrics	`enterprise` slug, `day` (YYYY-MM-DD)
Users 28-day report	Enterprise (user-level)	Per-user 28-day metrics for every seat	`enterprise` slug
Users 1-day report	Enterprise (user-level)	Per-user metrics for a single day	`enterprise` slug, `day` (YYYY-MM-DD)
Org 28-day report	Organization	Org-level 28-day aggregated metrics	`org` slug
Org 1-day report	Organization	Org-level single-day metrics	`org` slug, `day` (YYYY-MM-DD)

Authentication

Auth Method	Required Scope
Personal Access Token (classic)	`manage_billing:copilot` or `read:enterprise`
Fine-grained PAT	"Enterprise Copilot metrics" — read permission

Example API call:

curl -H "Authorization: Bearer YOUR_PAT" \
     -H "Accept: application/json" \
     "https://api.github.com/enterprises/{enterprise}/copilot/usage"

Response format

The API returns a JSON response containing a download_links array. Each entry provides a signed URL pointing to an NDJSON file:

{
  "download_links": [
    {
      "url": "https://github-cloud.s3.amazonaws.com/...",
      "expires_at": "2025-01-16T12:00:00Z",
      "content_type": "application/x-ndjson"
    }
  ]
}

Download the signed URL to retrieve the raw NDJSON data for processing.

Full API documentation

See the complete endpoint reference, request/response schemas, and pagination details at: Copilot Usage Metrics API docs

What to do next

Generate a PAT with the appropriate scope and test a 28-day enterprise report call.
Build an automated pipeline that pulls data daily and loads it into your data warehouse.
Use the user-level endpoints to build per-team adoption dashboards.

Legacy APIs & Sunset Timeline

Legacy API Retirement

Two older Copilot metrics APIs are being retired. Migrate to the Copilot Usage Metrics API before the sunset dates.

User-Level Feature Engagement Metrics API: Sunset March 2, 2026
Copilot Metrics API (/copilot/metrics): Sunset April 2, 2026
Action required: Migrate all integrations to the current Copilot Usage Metrics API endpoints.
The legacy Copilot Metrics API does NOT include Agent mode or Edit mode telemetry.

API comparison

API Resource	Scope	Key Differences
Copilot Usage Metrics API (current)	Enterprise, Org, User	Unified telemetry across completions, chat, and agent. Actively maintained. Supports user-level granularity.
Copilot Metrics API (legacy, sunsetting)	Enterprise, Org, Team	Aggregated data only. No agent/edit mode telemetry. No user-level data. Sunset April 2, 2026.
Copilot User Management API	License / Seat	Seat assignments and license state. Source of truth for licensing. Not affected by sunset.

Migration checklist

Inventory all scripts and integrations that call /copilot/metrics endpoints.
Map legacy fields to their equivalents in the Copilot Usage Metrics API.
Update authentication — ensure tokens have manage_billing:copilot or read:enterprise scope.
Test in a non-production environment before switching over.
Decommission legacy API calls well before the sunset dates.

What to do next

Audit your existing integrations for any calls to the legacy /copilot/metrics endpoints.
Plan your migration timeline — aim to complete migration at least 30 days before sunset.
Review the Copilot Usage Metrics API docs for the current endpoint reference.

Data Freshness

Understanding when metrics data becomes available is important for setting expectations with stakeholders and scheduling automated reports.

Aspect	Details
Processing delay	Data is available within 3 full UTC days of the activity
Example	Monday's usage data becomes visible by end of day Thursday UTC
Weekend effect	Weekend activity may add up to 4 calendar days of delay due to processing schedules
Time zone	All timestamps and daily boundaries are based on UTC

Activity Day    Available By (latest)
──────────────  ─────────────────────
Monday          Thursday EOD UTC
Tuesday         Friday EOD UTC
Wednesday       Saturday EOD UTC
Thursday        Sunday EOD UTC
Friday          Monday EOD UTC
Saturday        Tuesday–Wednesday EOD UTC
Sunday          Wednesday–Thursday EOD UTC

What to do next

Schedule automated report generation to run no earlier than 4 days after the target date.
Communicate the 3-day lag to stakeholders so they don't expect real-time data.
Use the freshness window to batch daily pulls into a weekly cadence if real-time granularity isn't needed.

Permissions & Policy Prerequisites

A quick-reference summary of everything required to access Copilot metrics data.

Requirement	Details
Policy	"Copilot usage metrics" must be enabled at the enterprise level
Roles (enterprise)	Enterprise owner, org admin, billing manager, or custom role with "View Enterprise Copilot Metrics" permission
Roles (org-only)	Custom role with "View organization Copilot metrics" permission for org-scoped visibility
API auth (classic PAT)	`manage_billing:copilot` or `read:enterprise` scope
API auth (fine-grained PAT)	"Enterprise Copilot metrics" — read permission

Custom roles

If you need to grant metrics access without full admin or billing privileges, create a custom enterprise role with only the "View Enterprise Copilot Metrics" permission. For org-scoped access, use the "View organization Copilot metrics" permission on a custom org role.

What to do next

Confirm the "Copilot usage metrics" policy is enabled in Enterprise settings → Policies → Copilot.
Create a dedicated service account PAT with minimal scopes for automated API access.
Set up custom roles for team leads who need metrics visibility without admin privileges.
Review token expiration policies and set up rotation schedules for API credentials.