Feature-Based Capacity Planning

The views described below are part of Capacity Planning capabilities on Program level. They allow to compare demand as planned workload with available capacity of teams who do the work. It is especially helpful to Product Managers, Release Train Engineers, and System Architects as they determine whether the overall program increment (PI) goals can be achieved with the available teams.

Program Backlog is mostly a list of features, but can contain other items such as technology "enablers". Features are planned into Releases or Program Increments (PIs). Program Backlog view allows to distribute program backlog items (Features) across one or more Program Increments (PIs) and teams. We recommend to use this view to plan a next or track a current PI of a single Agile Release Train (ART).

Demand is summarized from estimated Features. Capacity for Releases/Program Increments is set manually or summarized from inner Team Iterations.

Use Cases

Capacity Planning board can be used to track progress toward a strategic roadmap. It helps Program Managers see a big picture and forecast upfront in what release which feature can be completed and delivered. This capability includes the ability to track releases by the number of assigned and completed features based on counts, high-level estimates, or story points. As result, the capability gives understanding what work items will be included in a scheduled release.

For example, if a release is overfilled, Program Managers may reschedule less important features forward to a next one. Also, Capacity Planning views help to detect bottlenecks and may signal that the number of allocated teams or the number of people in teams should be increased.

Prerequisites. Data Mapping

Releases or Program Increments (PIs) should be date-driven and incremental, that is, scheduled one after one without dates overlapping. Releases are typically 8 to 12 weeks long. They also may represent 2-3 calendar months or 1 financial quarter.

Typical feature should fully fit into a single release. When features are bigger, we recommend to split them into smaller sub-features, so that each of them could fit a release.

Program Backlog View Setup

To plan capacity in your Program Backlog, open or create a board view showing Features as cards and Releases as columns. Filter out past releases and empty columns if needed.

Customize cards to let "Initial Estimate" numeric field be shown for each Feature. As result you could quickly see demand of each card.

Demand versus Capacity comparison appears in every cell. Total demand of selected cards is calculated on the fly.

Red highlighting indicates that the release capacity is overloaded. Either increase capacity of the release or unassign some features from it.

Demand Calculations

Demand is summarized from assigned Features. It is measured in points. By default, the system calculates demand from Initial Estimates of the Features.

To enable accurate demand planning, members of agile teams should provide high-level estimates and assign the points to each feature. It is significantly important that all teams should use relative points of the same scale.

Multiple scenarios are supported in Targetprocess now and can be configured with the help of views, fields, and automations:

• Each feature demands 1 point by default. Capacity planning is count-based;
• Each feature is high-level estimated with points or t-shirt sizes scale. Estimates are provided before a feature is completely decomposed into estimated user stories;
• Each feature has estimation equals to total count of its user stories. Stories should be created in advance;
• Each feature has estimation summarized from its user stories' points. Stories should be created and estimated in advance;
• Sophisicated mix of these approaches.

Should you require to activate any of these scenarios please request our Support Team for assistance.

Capacity Calculations

Capacity for Releases/Program Increments is set manually or summarized from inner Team Iterations. It is measured in points.

When the board has no swimlanes, then capacity in each cell is retrieved from Capacity fields of Releases.

The fields should be configured and filled in in advance. User Guide for Administrators: Release Capacity: set up total calculation.

Team-based Capacity Planning: Features into Releases

Team-based capacity planning in Targetprocess has support for 1 feature : 1 assigned team use case.

"Almost equal" (≈) warning indicator means that the demand is calculated not accurately enough to manipulate mathematically. It happens when there are features with more than one team assigned. Team-based estimates for features are not supported in Targetprocess yet.

For accurate calculations we recommend to split cross-team features into multiple smaller single-team features. As result, each team could give its high level estimation to each feature, and whole Program Backlog can be planned considering estimates and available capacity.

When Teams are displayed as swimlanes, it is possible to calculate capacity of a release for each team as total capacity of its team iterations.

To ensure that calculations work properly, please check that your releases are split into team iterations for all the teams that participate in your Agile Release Train (ART). Each team iteration should have Capacity value set. We recommend to use historical Velocity values as target Capacity. Keep in mind that vacations and changes in allocations of team members may significantly affect team capacity.

Best Practices

Visual Encoding

To identify features that might require additional data adjustments, we recommend to activate the following visual encoding rules on your capacity planning board: