Code Cycle Time

Track and improve the speed at which you are able to ship code.

Purpose

Being effective in completing tasks and shipping features depends a lot on the cadence and flow of the development process. One indicator of the development flow is the cycle time of the pull/merge request (PR). The cycle lead time tracks how much time is spent in the different PR stages, such as development, waiting for a review pickup, the review time itself, and merging the PR.

The shorter the cycle time, the more responsive the development team can potentially be.

Improvement Actions: Long cycle times might result from too large tasks leading to large or complex code commits that in turn, lead to long review and merge times. Large/complex PRs also tend to be picked up for review with greater delays as reviewers need to block out more time.

Other causes can come from underspecified tasks. For this more detailed and broken down planning can help to make improvements. Additionally, improving resource constraints and giving developers fewer parallel activities with less interruptions will help to reduce the overall cycle time.

Explanations

This report tracks the PR lifecycle for teams and repositories. Views can be aggregated across repositories and across teams for a holistic approach. Some of the charts in this report include:

Code Cycle Time: The cycle time from the first commit in a PR until it is closed/merged in, i.e., being done.

Stages: Breakdown of the cycle time into PR stages of:

• Development: time from the first commit to opening the PR.

• Response: time from opening the PR to start of a review.

• Review: time from first review response to approval of the PR.

• Integration: time from approval of the PR until its merge/closure.

Distribution: Shows all the PRs clustered into different cycle time buckets. This enables you to quickly identify outliers and investigate why certain PRs might have taken a long time to complete. Common causes can indicate specific work patterns or process problems to be addressed.

Breakdown by Stage: Shows the lead time trend over time broken down by PR stages. This gives indications about work pattern consistency and changes in efficiency. It might be that review delays are causing long cycle times or merge problems.

Good to Know

By default, all the PR stages and activities on them are imported as meta-data from your connected repository providers such as GitHub, GitLab, Bitbucket, Azure DevOps, etc. For consistency Logilica uses the GitHub naming conventions of Pull Requests.