GitHub Copilot Metrics
Important Notes:
- Average metrics are presented as a daily average over the selected time period (e.g., the daily average for the last month).
- Per GitHub Copilot's documentation, data is available only for teams with five or more active Copilot licenses on a given day.
High-Level Ratios
These metrics provide a top-level view of how effectively your organization is leveraging its Copilot investment.
| Metric | Definition | How to Interpret |
|---|---|---|
| Adoption Rate | The percentage of developers who were active in any capacity on a given day. (Daily Active Users / Daily Active Developers) | A high adoption rate indicates that licenses are being utilized and developers are at least seeing suggestions. A low rate may signal issues with installation, awareness, or perceived value. |
| Engagement Rate | The percentage of active users who took a specific action, such as accepting a suggestion or using chat. (Daily Engaged Users / Daily Active Users) | This measures deeper value than adoption. A high engagement rate means developers are not just seeing suggestions, but actively using suggestions as well as other features, like IDE chat. |
| Code Acceptance Rate | The percentage of code suggestions that were accepted by a developer. (Total Code Acceptances / Total Code Suggestions) | This is a direct measure of the quality and relevance of Copilot's code suggestions. A high rate indicates that Copilot is generating useful, high-quality code that aligns with your team's needs. |
User Activity (Daily Averages)
These metrics break down the user base into different activity levels and feature usage.
| Metric | Definition | How to Interpret |
|---|---|---|
| Daily Avg Licensed Users | The average number of users with an active Copilot license on any given day in the period. Only available for the entire organization. | The baseline for measuring the potential user pool and calculating license utilization. |
| Daily Avg Active Users | The average number of licensed users who had any Copilot activity, including passive actions like seeing a suggestion. | Measures the overall reach of Copilot within your licensed user base. |
| Daily Avg Engaged Users | The average number of licensed users who performed an intentional action, such as accepting a suggestion or asking a chat question. | This is a key metric for understanding how many developers are deriving value from Copilot, beyond installing an extension. |
| Daily Avg Passive Users | The average number of users who were active but not engaged (i.e., they received suggestions but did not accept any or use other features). (Daily Active Users - Daily Engaged Users) | A high number of passive users may indicate that suggestions are not relevant, that developers are not trained on how to use the tool effectively. |
| Daily Avg Engaged Users - IDE Chat | The average number of users who sent at least one prompt to Copilot Chat within their IDE. | Measures adoption of the conversational AI feature for coding, debugging, and learning directly in the editor. |
| Daily Avg Engaged Users - IDE Code Completions | The average number of users who accepted at least one code suggestion within their IDE. | This is the core metric for code generation usage. |
| Daily Avg Engaged Users - Web Chat | The average number of users who sent at least one prompt to Copilot Chat on github.com. | Indicates usage of Copilot for tasks outside the IDE, such as general research or code explanation on the GitHub website. |
| Daily Avg Engaged Users - PR Summaries | The average number of users who generated at least one pull request summary. | Measures the adoption of Copilot's features for streamlining the code review and documentation process. |
Per-User Averages
These metrics normalize total activity by the number of active users to show the average behavior of an individual.
| Metric | Definition | How to Interpret |
|---|---|---|
| Daily Avg IDE Chats per Active User | The total number of IDE chats divided by the number of daily active users. | A high number indicates that active users are heavily relying on the chat feature for problem-solving, research, and planning, making it a core part of their workflow. |
| Daily Avg Web Chats per Active User | The total number of web chats divided by the number of daily active users. | Measures the average engagement with Copilot on github.com per user. |
| Daily Avg Code Acceptances per Active User | The total number of code acceptances divided by the number of daily active users. | This shows, on average, how many useful suggestions each active developer is getting per day. |
| Daily Avg PR Summaries per Active User | The total number of PR summaries created divided by the number of daily active users. | Indicates how frequently the average developer is using Copilot to automate documentation and code review of their pull requests. |
Volume Metrics
These are the raw counts of events and are useful for understanding the overall scale of Copilot's activity.
| Metric | Definition | How to Interpret |
|---|---|---|
| Total IDE Chats | The total number of chat sessions initiated by users in their IDE. | A raw measure of conversational AI usage. |
| Total Web Chats | The total number of chat sessions initiated on github.com. | Volume of chat usage outside the IDE. |
| Total Code Suggestions | The total number of times Copilot generated a code suggestion. | The denominator for calculating the overall acceptance rate. |
| Total Code Acceptances | The total number of times a developer accepted a code suggestion. | The numerator for acceptance rate. |
| Total Code Lines Suggested | The total number of lines of code that Copilot proposed. | A raw measure of the volume of code being suggested by AI. |
| Total Code Lines Accepted | The total number of lines of code from suggestions that developers accepted. | Illustrates the volume of code that developers are accepting and then refining. |
| Total PR Summaries Created | The total number of pull request summaries generated. | A raw measure of how much documentation and code review work is being automated. |