Improve team’s efficiency and performance

The Leiga Efficiency visualizes the requirements of proposal, design, development, and release cycles from the perspective of the LigaAI project workflow. This helps the team understand and evaluate their response efficiency and delivery capability more clearly.

Feature

Click the navigation icon on the left, enter the efficiency dashboard and select the Work Item tab. Then you can view the workflow data filtered by project name, work type and time spans.

Metric

Metirc	Definition	Calculation
Lead Time	The time costs from proposal to final completion of certain work types.	The 85th percentile time for all works transit from proposal to final completion in the selected period (sorted from slowest to fastest).
Cycle Time	The time spends on a work from “In Progress” to “Completed”.	The average time for all works transit from “In Progress” to “Completed” in the selected period.
Throughput	The number of completed works in the selected period.	The number of works that have been dragged to the “Completed” in the selected period. For example: If a team completes 80 work items in one month, the team’s throughput is 80.
Efficiency	The average work processing efficiency per day in the selected time interval.	Efficiency = Throughput/Cycle Time
Lead Time Distribution	The time distribution from proposal to final completion of certain work types.	The Lead Time distribution. X-axis: Delivery time, Y-axis: Number of deliveries
Work in Progress	The number of WIP every day.	The number of processing works every day.
work Distribution	The distribution of different work types.	The distribution of different work types in the selected period.
Node Dwell	The time spent on a work at each node.	The time that a work item stays at a node during the selected period, such as the time costs from entering state A to changing to state B.

Understand your engineering team's velocity

Through tool integration and automatic collection of development process data, using the Cycle Time metric as a framework, visualize the average time taken by the development team from development to release.

Feature

Click on the Development tab, select the Git project and the time period you would like to view, then you are able to grasp the average working time for each Development process.

Hover over the icon in the upper right corner, you can quickly understand the current risks and work item the team faces and adopt the problem-oriented optimized strategy.

Currently, GitLab is supported, and further Git tools will be supported in the future.

Metric

Metric	Definition	Calculation
Cycle Time	The average time it takes for a single engineering task to go through the delivery process from “code” to “production”.	Cycle time is the sum of the average working cycle of each stage. (Cycle Time = Coding Time + Picking Time + Review Time + Deployment Time)
Coding Time	The average time it takes from the DoR-satisfied requirements arriving at the technical team to completing coding, ano initiating a Pull Request.	The time it takes from the first push to the Pull Request is sent.
Pick up Time	The waiting time from the initiation of a Pull Request to the start of a code review. The shorter the picking time, the closer and more efficient the cross-functional technical team collaboration, and the healthier the review process.	The time it takes from a Pull Request is sent to the code review starts (counting since the first comment is made).
Review Time	The time it takes to complete a code review and merge a pull request.	In general, the first comment that is made is considered as the start of a code review, and the confirmation of the review outcome serves as the end of a code review.
Deploy Time	The time it takes to deploy code to production.	Starting from confirming the review outcome and ending with synchronizing new code to the master branch successfully.
Cycle Time Distribution	A measure and display of the overall development cycle.	X-Axis: Number of days Y-Axis: Number of delivered tasks
PR Change Volume	The number of lines of code modified in pull requests. (Smaller-scale PR change volume usually indicates shorter review times.)	The average number of lines of code modified in pull requests every day.
Review Contribution	The contribution of code review during the selected period. (To avoid bottleneck situations where review requests pile up on one or a few people.)	The time and scale of code reviews by each reviewer, as well as their contribution ratio.
Deployment Frequency	The frequency of code deployment during the selected period. (A high deployment frequency indicates a stable and healthy continuous delivery pipeline.)	The number of times that code is merged into the master branch each day during the selected period.

Team Health

The Leiga Efficiency not only provides real-time cycle time calculations and visualizations for teams, but also offers industry benchmarks to help teams diagnose and identify problems quickly.

Feature

Click on the Trend tab to compare your team’s current metrics with industry data, then you are able to quickly understand the team's current metrics in the industry. What's more, by comparing your team’s current and past performance, you can observe if your team is developing in a positive direction.

FAQ

1. What are the permission settings for this module?

During initialization, we grant permissions to organization owners and organization administrators. If other members require access to performance content, the above roles can configure permissions in the permission management section. Data permissions are limited to the user’s visual range in the project they are associated with.

2. How to define the start of work?

For projects that use iterations in ligaAI, the start time is when the work enters the iteration. For non-iterative projects, the start time is the time the work is created in the project.

3. How well does Git tool support?

Currently, we support Gitlab. We will expand and support more Git tools according to customer feedback and demands to collect more behavior data related to Git.

Cross Project