How to Get Code Review Metrics in Bitbucket
Code review is one of the cornerstone processes in the software development workflow. While it ensures the quality of merged pull requests, many teams experience code review bottlenecks that prolong delivery time and block pull request authors from moving forward with their tasks. This is where code review metrics can come in handy: they can help engineering managers and teams track code review progress and promptly remove blockers as soon as they arise. In this article, we will explore what code review metrics exist, which ones could address the needs of your team the best, and how to get code review metrics in Bitbucket Data Center with the help of the Awesome Graphs app.
Awesome Graphs is a reporting and data-providing tool that helps over 1,6K engineering teams leverage development analytics in Bitbucket. It provides out-of-the-box reports as well as CSV and REST API data export for custom automated solutions.
How to select code review metrics
While there are many code review metrics that can be tracked, not all of them will be equally useful for each team. The principle of quality over quantity works in the metrics world, too! So, when choosing a metric or metrics to track, consider the following questions that can help guide you in the right direction:
- Has your team experienced any recent delays during code review? What exactly was holding you back at that moment?
- What are the biggest challenges in your current code review process? Which metrics could help better identify these issues?
- Are you looking to improve code review speed, quality, or both?
- How collaborative are the team members during code review?
- Are you satisfied with the review quality? What makes an excellent code review in your team?
Answering these questions should help you narrow down the list of metrics that will suit your team’s exact needs and address its challenges. Now, let’s explore what metrics are available in Bitbucket and how to set them up using the Awesome Graphs app.
Code review metrics in Bitbucket: overview
While native Bitbucket Data Center functionality doesn’t include engineering statistics, it is possible to bring your code review metrics vision to life. Awesome Graphs for Bitbucket allows you to track a variety of metrics, including some of the core code review metrics: Resolution Time, Cycle Time, Reviewer Load, and Review Participation. For each one of them, let’s take a look at the tracking opportunities in the app.
Resolution Time
Defined as the time period from the creation of a pull request until its merge, Resolution Time is a metric that helps many teams around the globe streamline code review. If you are looking for a quick way to get Resolution Time insight, consider using Resolution Time Distribution Report.
This ready-to-use histogram groups pull requests based on the time between their opening and merge. It also features useful statistics above the graph: median, interquartile range, minimum, maximum, and sample size. The graph allows you to zoom in on a group of pull requests by simply clicking on any bar: the summary stream below the graph will adjust to show only the pull requests included in the bar you selected.
Cycle Time
Alternatively, if you would like to take a broader look at the pull request lifecycle, you can opt to use Cycle Time Report for your reporting needs.
This graph enables you to track the full PR cycle from the first commit to the merge. Using Cycle Time Report, you can track average cycle time and evaluate each particular phase:
- Time to Open: Time from first commit until a pull request is created
- Pickup Time: Time from when the pull request is created until the review begins
- Review Time: Time from the start of the review until the pull request is approved by all reviewers
- Time to Resolve: Time from the pull request approval until it is merged or declined
Check out this article on how to track cycle time in data export and for each individual pull request for complete insight into cycle time analytics.
Reviewer Load
Sometimes, optimizing the code review process requires teams to rebalance pull request assignments. The easiest way to gauge each reviewer’s workload is to evaluate Reviewer Load. Pie Chart Report, available in the Awesome Graphs app, can easily help you find reviewers with the most load on the team.
With data conveniently presented in both table and visualized format, Pie Chart Report can help you evaluate the amount of pull requests that require attention from top reviewers. Clicking on a particular sector of the chart will enable you to get a list of all the reviewer’s pull requests in the Summary stream below the graph.
Review Participation
Tracking how collaborative team members are in providing feedback to PR authors is also a crucial code review metric. Awesome Graphs allows you to work closely with this data, evaluating pull request comments from both the author’s and reviewer’s perspectives. We have written a full article on PR comment analytics opportunities in Bitbucket. Make sure to check it out for more insight on how to measure Review Participation.
How to find code review metrics in Bitbucket Data Center
To gain access to code review reporting, activate your free trial for Awesome Graphs for Bitbucket and install the app in your instance. You will be able to find reports discussed in this article on both project and repository levels and even configure custom teams to enjoy code review analytics in one click.
If you would like to venture beyond code review metrics and explore other app opportunities, check out our resources further:
