Release planning is in the center of the Agile Onion and is sometimes overlooked. When pressed for a release date or scope, the “planning” is often a best guess by the team and not always based on accurate data. But, there is a better way to predict releases using data you already have.
Lean and Agile development methods emphasize planning on five different levels. These methods can generate amazing high-level vision and 3-year plans when creating long term goals. It can also plan and commit to lower-level detailed iteration and daily planning. Yet when it comes to creating release plans, the methods often degrade into a best guess by the team, architect, or project manager. Inaccurate release plans are a challenging problem for industries where the stakeholders’ capacity to take frequent releases is small or non-existent. Predicting a release and sticking with it becomes vital when your stakeholders only take releases once or twice a year. In this talk, I demonstrate how to use your team’s historical data to accurately predict a release using the ActionableAgile™ tool. Monte Carlo method demonstrations are shown to forecast releases by date or number of features. We also review Cumulative Flow Diagrams and how to interpret them to identify areas of process improvement.