Track: Game changing methods and practices / Time: Monday 13:20 - 14:10 / Location: Amalienborg

We have to estimate the size of projects for forecasts, for portfolio-related decisions, and to prepare quotations for customers. But how can we reliably estimate a project's size without identifying and analyzing every single user story? And does it make sense to analyze in so much detail when a substantial portion of the effort is likely to be wasted? Software sizing is an essential part of estimating our resource needs for a project.  In order to forecast time and budget, we need to be able to quantify “what” we are building.  This is because the resources required are related to “how much” software is built. Randomized Branch Sampling (RBS) represents an approach that can enable us to accurately size a prospective project by focusing on just a small portion of the planned tasks. It was developed so an individual only had to count the number of fruit that existed on a limited number of branches of a fruit tree to arrive at a highly reliable estimate of the actual number of fruit that could be harvested from the entire tree. The same technique can be applied to Agile and Lean projects by examining a subset of user stories and tasks across a range of proposed epics in order to estimate total project size. This session will give attendees practical knowledge how to use Scrumban Randomized Branch Sampling (RBS) when planning Agile projects. The technique presented will produce accurate results for most projects with relatively little effort.