Skip to content

benarch/Github-Copilot-Usage-Insights

BranchesTags

Repository files navigation

GitHub Copilot Usage Extended Insights Dashboard

A full-stack analytics dashboard that visualizes GitHub Copilot usage data with extended organizational visibility.

What it does:

  • Displays Copilot usage metrics: active users, chat requests, code suggestions, acceptance rates
  • Shows data by IDE, chat mode, AI model, and timeframe (7/14/28 days)
  • Provides team-level analytics by importing team/org data from GitHub
  • Aggregates individual user metrics to team and organization levels
  • Includes an AI chatbot for natural language queries about the data

Key features:

  • Familiar design - styled to match the native GitHub Copilot usage dashboard for easy adoption
  • Dashboard with interactive charts (Recharts)
  • Table views: Summary, Detailed Report, Query Builder, Teams View
  • Acceptance rate sorting - adjust user license/tier by their usage performance
  • Nested team/org views - drill into users within teams and organizations with hierarchical visibility
  • License management insights - identify usage patterns for:
    • License removal candidates (low/no usage)
    • License tier adjustments (Copilot Business 300 → Enterprise 1000 premium requests)
    • Cost center allocation for specific users
  • Search & filter - searchbox support across all table views
  • Teams & Organizations import (CSV/JSON) with nested membership support
  • People page linking users to their teams
  • Dark/light theme toggle

Related Tools

GitHub Users, Teams & Organizations Exporter

github-users-teams-export

A companion tool that exports GitHub users, teams, and organizations data. Use this to:

  • Import users into the Teams tab for team-level analytics
  • Add extended visibility on users, team membership, and nested team/organization membership
  • Map Copilot usage data to organizational structure

GitHub Copilot Usage Anonymizer

github-copilot-usage-anonymizer

A data anonymization tool used to generate the demo data in this project. It:

  • Anonymizes Copilot usage data by replacing real user names with generated names
  • Preserves data structure and relationships while protecting privacy
  • Enables safe sharing and demonstration of Copilot analytics without exposing actual user identities

Screenshots

Detailed Report - Table View

Detailed Report Search, filter, and sort users by interactions, code generation, acceptances, and lines of code metrics

Code Generation Insights

Code Generation Total suggestions, acceptance rates, code completions trends, and lines of code added/deleted

Model Usage Analytics

Model Usage Model usage per day, chat model distribution, and model usage per chat mode breakdown

📸 Want to see all dashboard views? Check out the complete screenshots gallery in app-screenshots/README.md

Tech Stack

Layer Technologies
Frontend React 18+, TypeScript, Tailwind CSS, Vite, Recharts, TanStack Query
Backend Express.js, TypeScript, SQLite (better-sqlite3), OpenAPI/Swagger
DevOps Docker & Docker Compose

Getting Started

Local Development

# Clone the repository
git clone https://github.com/benarch/Github-Copilot-Usage-Insights.git
cd Github-Copilot-Usage-Insights

# Install dependencies
npm run install:all

# Seed database with sample data
npm run db:seed

# Start development servers
npm run dev

Docker Deployment

# Build and start containers
docker compose up -d --build

# Or run in foreground to see logs
docker compose up --build

Stopping the Application

# Stop Docker containers
docker compose down

# Stop local development (press Ctrl+C in terminal)

Importing Data

Copilot Usage Data (Insights Tab)

Import your GitHub Copilot usage data from the native GitHub Copilot usage dashboard:

  1. Export from GitHub: Go to your organization's GitHub Copilot usage page and use the Export option to download usage data
  2. Navigate to Insights tab in this dashboard
  3. Upload file using the upload button
  4. Supported formats:
    • JSON - Standard JSON array format
    • NDJSON - Newline-delimited JSON (one record per line)

The uploaded data will populate all dashboard views with your organization's Copilot usage metrics.

Users, Teams & Organizations (Teams Tab)

Import organizational structure from GitHub Enterprise to enable team-level analytics:

  1. Export from GitHub Enterprise: Use the github-users-teams-export tool to export users, teams, and organizations
  2. Navigate to Teams tab in this dashboard
  3. Upload file using the import button
  4. Supported formats:
    • JSON - Standard JSON array format
    • NDJSON - Newline-delimited JSON (one record per line)
    • CSV - Comma-separated values with header row

Once imported, you can:

  • View aggregated Copilot usage by team
  • Drill into nested team hierarchies
  • See organization-level metrics
  • Link individual users to their team membership

Architecture & Flow Diagrams

Data Import & Application Flow

flowchart TB
    subgraph DataSources["📥 Data Sources"]
        GH_API["GitHub Copilot<br/>Usage API"]
        GH_EXPORT["github-users-teams-export<br/>Tool"]
        ANON["github-copilot-usage-anonymizer<br/>(Demo Data)"]
    end

    subgraph Import["📤 Import Process"]
        JSON_UPLOAD["Upload JSON/CSV<br/>via UI or API"]
        TEAMS_IMPORT["Import Teams &<br/>Organizations"]
        USERS_IMPORT["Import Users with<br/>Team Membership"]
    end

    subgraph Backend["⚙️ Backend (Express + SQLite)"]
        API["REST API<br/>:3001"]
        DB[(SQLite<br/>Database)]
        SEED["Seed Script<br/>npm run db:seed"]
    end

    subgraph Frontend["🖥️ Frontend (React + Vite)"]
        UI["Dashboard UI<br/>:3000"]
        CHARTS["Charts<br/>(Recharts)"]
        TABLES["Table Views"]
        CHATBOT["AI Chatbot"]
    end

    subgraph Features["📊 Analytics Features"]
        OVERVIEW["Overview Dashboard"]
        CODE_GEN["Code Generation<br/>Insights"]
        COPILOT_USAGE["Copilot Usage<br/>Analytics"]
        TEAMS_VIEW["Teams View<br/>Aggregated Metrics"]
        PEOPLE["People Page<br/>User Details"]
        ORGS["Organizations<br/>Nested View"]
    end

    %% Data flow
    GH_API --> JSON_UPLOAD
    GH_EXPORT --> TEAMS_IMPORT
    GH_EXPORT --> USERS_IMPORT
    ANON --> JSON_UPLOAD

    JSON_UPLOAD --> API
    TEAMS_IMPORT --> API
    USERS_IMPORT --> API
    SEED --> DB
    API <--> DB

    API --> UI
    UI --> CHARTS
    UI --> TABLES
    UI --> CHATBOT

    CHARTS --> OVERVIEW
    CHARTS --> CODE_GEN
    CHARTS --> COPILOT_USAGE
    TABLES --> TEAMS_VIEW
    TABLES --> PEOPLE
    TABLES --> ORGS
Loading

User Workflow

flowchart LR
    subgraph Step1["1️⃣ Setup"]
        INSTALL["npm run install:all"]
        SEED["npm run db:seed"]
        START["npm run dev"]
    end

    subgraph Step2["2️⃣ Import Data"]
        UPLOAD_USAGE["Upload Copilot<br/>Usage JSON"]
        IMPORT_TEAMS["Import Teams<br/>CSV/JSON"]
        IMPORT_ORGS["Import Organizations"]
    end

    subgraph Step3["3️⃣ Analyze"]
        VIEW_DASHBOARD["View Dashboard"]
        FILTER_TIME["Filter by Timeframe"]
        SEARCH_USERS["Search Users"]
        SORT_METRICS["Sort by Metrics"]
    end

    subgraph Step4["4️⃣ Take Action"]
        LICENSE_REVIEW["Review License Usage"]
        TEAM_METRICS["Aggregate Team Metrics"]
        EXPORT_REPORT["Export Reports"]
    end

    INSTALL --> SEED --> START
    START --> UPLOAD_USAGE
    START --> IMPORT_TEAMS
    START --> IMPORT_ORGS
    
    UPLOAD_USAGE --> VIEW_DASHBOARD
    IMPORT_TEAMS --> VIEW_DASHBOARD
    IMPORT_ORGS --> VIEW_DASHBOARD
    
    VIEW_DASHBOARD --> FILTER_TIME
    VIEW_DASHBOARD --> SEARCH_USERS
    VIEW_DASHBOARD --> SORT_METRICS
    
    FILTER_TIME --> LICENSE_REVIEW
    SEARCH_USERS --> TEAM_METRICS
    SORT_METRICS --> EXPORT_REPORT
Loading

Synthetic Data for Functionality Tests

The synthetic_data_for_functionality_tests/ folder contains anonymized demo datasets to help you explore the dashboard's capabilities and test different data scenarios without needing to export real organizational data.

Dashboard & Features Testing

To review the look and feel of the dashboard and explore usage, trends, and view capabilities, use any of these files:

  • ben-test-it-anonymized-1836-records.ndjson - Comprehensive dataset with 1,836 user records in newline-delimited JSON format
  • copilot-usage-2026-02-01.csv - CSV format usage data for February 1, 2026
  • copilot-usage-2026-02-01.ndjson - Newline-delimited JSON format for the same date
  • demo-data-github-copilot-usage-anonymized-823-records.json - Smaller JSON dataset with 823 records (ideal for quick testing)
  • extensive-models-demo-data-copilot-usage-2026-02-01.csv - Comprehensive CSV with model usage variations
  • extensive-models-demo-data-copilot-usage-with_ide-variation-2026-02-01.json - JSON data showing IDE distribution across models and usage patterns

How to use: Upload any of these files to the Insights tab to populate the dashboard with usage metrics, charts, and detailed reports.

Team & Organizational Correlation Testing

To review data showing correlation between users, team membership, and aggregated team usage metrics, use the files in the correlated-data/ subfolder:

  • correlated--demo-data-copilot-usage-missing-some-views.json - Usage data with partial team coverage (useful for testing incomplete data scenarios)
  • correlated--demo-data-users.csv - User roster with team membership information

How to use:

  1. First, upload the correlated--demo-data-users.csv file to the Teams tab
  2. Then, upload the correlated--demo-data-copilot-usage-missing-some-views.json file to the Insights tab
  3. This enables the Teams View for aggregated team metrics and nested organizational hierarchy views

About

No description, website, or topics provided.

Resources

Readme
Activity

Stars

8 stars

Watchers

0 watching

Forks

1 fork
Report repository

Releases

No releases published

Packages

 
 
 

Contributors

Languages