> For the complete documentation index, see [llms.txt](https://typo.gitbook.io/typo-help-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://typo.gitbook.io/typo-help-docs/platform/ai-coding/ai-impact.md).

# AI Impact

The **Impact** tab helps you measure the direct effect of AI usage on your team’s performance metrics.

We compare AI-generated PR metrics with non-AI PR metrics to identify trends and highlight how AI coding influences individual metrics, as well as overall team performance.

#### Enabling the Impact Module

To enable this module, please reach out to us at <hello@typoapp.io> and share how your team tags AI-related PRs.\
Based on your tagging method, we will configure the setup accordingly.

<figure><img src="/files/azYqJQIFnNJPKixgQIHN" alt=""><figcaption></figcaption></figure>

#### AI Prs

Shows the number of **AI-generated pull requests** compared to **non-AI pull requests** for the selected time period.

#### AI Code Type?

Shows how AI-generated code is distributed across different code change types. Typo analyzes AI-generated code when a pull request is created and classifies it into the following categories:

* **New Work** – New lines of AI-generated code introduced in the pull request.
* **Rework** – AI-generated code that replaces recently written code. If the original code being replaced was last modified within the last **30 days**, it is classified as **Rework**.
* **Refactor** – AI-generated changes made to older code that was last modified **more than 30 days ago**.
* **Churn** – AI-generated lines of code removed from the pull request.

#### AI Work type&#x20;

Shows the distribution of **AI-generated pull requests** across different types of engineering work, including:

* Infrastructure & Configuration
* Tests
* Interface
* Product Logic
* Documentation
* Others

Typo analyzes each AI-generated pull request when it is created and categorizes it based on the work type that contains the **highest number of AI-generated lines of code**. The pull request is then assigned to that category.

#### Impact on Metrics

<figure><img src="/files/9jNaZSkyRKyp0QfHeXzK" alt=""><figcaption></figcaption></figure>

* **PR Cycle Time Change:** This chart shows how your PR cycle time has changed when using the AI tool vs not using it. A positive number indicates an increase, while a negative number shows a decrease.
* **PR Review Time Change:** This metric tracks the change in the average time it takes for an AI PR to be reviewed vs the ones that are not generated using AI coding assistance.
* **PR Throughput Change:** This shows the change in the total number of merged PRs.
* **PRs/Developer Change:** This chart helps you understand if the AI tool is helping each developer complete more PRs.
* **Code Quality Change:** This shows the impact on the number of severe issues in the code health.
* **PR Size Change:** This metric tracks the change in the average size of your PRs, which can indicate whether the AI is helping to produce more concise code.

#### Team-Level Impact Table

This table provides a comprehensive summary of how AI is impacting each of your teams. It compares key metrics for each team to show the direct effect of AI on their performance.

* PR Cycle Time: The change in the average time it takes for a PR to be merged.
* PR Review Time: The change in the average time a PR spends in review.
* PR Throughput: The change in the total number of merged PRs.
* PRs/Developer: The change in the number of PRs merged per developer.
* Code Quality: The change in the quality of the code, measured by your pre-defined quality metrics.
* PR Size: The change in the average size of PRs.

{% hint style="info" %}
To enable this report for your account, navigate to **Settings > Integrations** and connect the AI coding tool used by your team.

You can add multiple coding assistant tools from the **Integrations** section as needed.
{% endhint %}