On this page:
What is Review Rate?
When teams fall under the pressure of needing to deliver quickly, it can be tempting to avoid the rigour of receiving feedback from peers to improve their code, tick the approval box and just merge it. The problem with this is practice is that it significantly increases the risk of introducing bugs into your code base, as well as making your code base less known and understood by other team members. Umano created Review rate to help teams strengthen their peer review practices and better protect their code base.
Review Rate is the percentage of pull requests reviewed and merged compared to the total number of pull requests merged in an interval.
Why does Review Rate matter?
Review Rate provides insight into the potential for risk entering your code base and is used as an indicator of a team’s governance related to quality assurance.
Review Rate is also an indicator of the health of your team’s learning culture as it highlights peer to peer learning, knowledge sharing and up-skilling.
What does a 'good' Review Rate look like?
Review Rate is presented as a percentage, where 100% indicates all pull requests for a particular interval have been reviewed.
Good practice indicates that as many pull requests are peer reviewed as possible, if not all.
A high Review Rate signals a culture of strong peer to peer learning, a shared understanding of new code being merged, and a reduced risk of introducing bugs into your code base.
A low Review Rate signals high practices of self-review. Whilst there are circumstances when this may be acceptable, it is not good practice.
How does TeamX measure Review Rate?
Review Rate is measured from the number of pull requests merged and reviewed in an interval relative to the total number of merged pull requests.
A review done by the author themselves is not considered as a review.
For example, if there were 4 pull requests merged in an interval, and only 2 were reviewed by a peer, then the team’s review rate for the interval is 50%.
TeamX surfaces each pull request merged in an interval, whether the pull request was peer reviewed or self-reviewed, along with who reviewed and approved the pull request prior to it being merged.
What drives your Review Rate?
Activities that contribute to and influence Review Rate include all the pull requests that are merged during an interval, and of those the number that were reviewed by a peer.
Your Review Rate is influenced by:
Total number of open pull requests
Total number of pull requests reviewed
Number of merged pull requests
Number of pull requests merged that were reviewed
Size of pull request
Number of issues addressed
Number of reviewers assigned
Tips for improving your Review Rate.
Be clear to identify the circumstances when a self-reviewed pull request is acceptable for your team, and nominate these times as the exceptions to the norm
Build your team’s culture of collaboration and spread the load by inviting all relevant team members to pick up a pull request opened for review
Ensure pull request reviews aren’t reviewed by your most experienced engineers by default, and encourage everyone to build a review practice