Unlocking Deeper Copilot Pro Insights: Tracking Your Software Developer Statistics

Understanding and optimizing resource consumption is a cornerstone of effective software planning and development. GitHub Copilot Pro subscribers, eager to gain deeper insights into their AI-powered coding assistance, are seeking more granular usage data to better track their activity and inform their workflows. A recent discussion on the GitHub Community forum highlights this growing need, focusing on the desire for detailed software developer statistics related to Copilot Pro premium requests.

The Quest for Detailed Copilot Pro Usage Data

The discussion, initiated by user leonsama, articulates a common desire among power users: a comprehensive breakdown of premium request consumption. While the current billing dashboard offers an overview, the request specifically calls for:

• Timestamp-level breakdown: Pinpointing the exact moment premium requests are consumed.
• Model attribution: Identifying which specific AI model was engaged for each request.
• Hourly/daily aggregation: Grouped usage logs, ideally with export capabilities, to understand patterns over time.

Such detailed data would empower users to proactively track usage against subscription limits, identify which workflows or prompts trigger premium requests, and self-audit billing without needing support intervention. This level of transparency is crucial for anyone looking to optimize their how to measure productivity of software developers when leveraging AI tools.

Current Capabilities and Existing Workarounds

Community member Gecko51 provided valuable context, revealing that more data exists than is immediately apparent through the standard UI. For individual Copilot Pro subscribers, an API endpoint can provide some of the desired information:

GET /users/{username}/settings/billing/premium_request/usage

This API, accessible with a GitHub App user access token (requiring the Plan user permission via OAuth), returns daily aggregated usage items, including product, SKU, model, unit type, and quantity. This addresses the "model attribution" and "daily aggregation" aspects of the request, offering a glimpse into your software developer statistics.

However, it's important to note a few caveats:

• No Request-Level Timestamps: The API does not provide individual request-level timestamps.
• Individual Plans Only: This endpoint only returns data for self-paid individual plans; organization or enterprise-managed licenses will not show up here.
• UI Access: A visual overview of per-model usage and a rolling chart is available at github.com/settings/billing under "Premium request analytics."
• Export Limitations: CSV export of usage data is currently an enterprise-only feature, leaving individual Pro users without a direct export option.

Achieving Hourly Granularity (Workaround)

For those needing hourly granularity, Gecko51 suggested a clever workaround: cron-polling the API endpoint and diffing consecutive snapshots. This "duct tape" solution, while not ideal, allows users to reconstruct hourly usage buckets:

# in a cron job, every hour
curl -s -H "Authorization: Bearer $GH_USER_TOKEN" \
"https://api.github.com/users/$USERNAME/settings/billing/premium_request/usage" \
| jq '.usage_items[] | {date, product, sku, model, quantity}' \
> "usage-$(date +%Y%m%dT%H).json"
# Then diff consecutive snapshots to recover per-hour usage.

This method, though requiring some technical setup, provides a way for individual developers to gain deeper insights into their AI consumption patterns, contributing to a better understanding of how to measure productivity of software developers in an AI-assisted environment.

Moving Forward: A Call for Focused Feature Requests

The discussion concludes with a recommendation to open a more focused feature request specifically for "expose request-level Copilot usage for individuals via API + CSV export." Such a targeted request has a better chance of being tracked on the roadmap compared to general billing-dashboard threads. As AI tools become more integral to daily coding, the ability to access and analyze detailed usage data becomes increasingly vital for effective resource management and performance optimization, ultimately enhancing software developer statistics and overall productivity.