> ## Documentation Index
> Fetch the complete documentation index at: https://doc.entelligence.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Code Review analytics
> Use pull-request flow, review effectiveness, churn, and code-quality trends to improve review policy and find risk.
Use Code Review analytics to answer three questions: **Are reviews moving? Is the feedback useful? Where is code risk accumulating?** Start with a trend, then open the underlying pull requests before drawing a conclusion.
## Review flow
The **Overview** tab tracks:
* Total and open pull requests
* Time to first review and time to merge
* Merge rate and pull-request velocity
* Active reviewers
* Review comments and average comments per pull request
* Pull-request size and backlog by author
Use these metrics to find a bottleneck: a growing backlog, slow first response, or a review policy producing too much back-and-forth.
## Code quality
The **Code Quality** tab brings together:
* Code churn over time and churn-impact summaries
* Churn hotspots and the most-churned files
* Reverted pull requests
* Three-month code turnover
* Bug reports by repository
* Pull requests with quality score, status, date, and ticket access
High churn is a prompt to investigate, not a verdict. It can reflect healthy iteration, a migration, generated changes, unclear ownership, or repeated rework.
## Review effectiveness
The **Review Analytics** tab shows:
* Average review cycle time
* Comments addressed and total comments
* Comment acceptance
* Average comments per merged pull request
* Merged pull requests by team
* Entelligence and human comments addressed
* Usage by user or team
Falling acceptance can point to noisy guidance, stale Learnings, or the wrong priority threshold. Rising cycle time can point to review volume, unclear ownership, or changes that are too large.
## Repository operations
Add or search repositories, open **Show details**, inspect enabled review behavior, and jump to repository pull requests.
Filter by time, status, and repository to investigate missing, running, or completed reviews.
Compare Entelligence and another review bot across acceptance, testability, security, and maintainability categories. Open the underlying comments before deciding which feedback was stronger.
## Ask better questions of the data
Some analytics expose **Ask Ellie**. Useful prompts include:
* Which repositories drove the increase in time to first review?
* Show the pull requests behind this churn hotspot.
* Which guidelines correlate with frequently dismissed findings?
* Where did automated reviews reduce human back-and-forth?
Validate the answer against the selected date range and source pull requests.
## Use metrics responsibly
* Improve review policy, repository ownership, and tooling.
* Compare trends across similar time windows.
* Pair counts with pull-request context.
* Do not turn one activity metric into an individual performance score.
## Next step
Review the [metric glossary](/reference/metrics) or tune [Code Review settings](/code-review/settings).