Estimating for Iteration: Story Points and Thin Vertical Slices

A presentation on my two favorite parts of Scrum: estimating by complexity and building vertically.


Presented at:

Presentation designed using Reveal.js, a jQuery presentation framework.

Thank you to the Drupal Association and DrupalCon Seattle 2019 organizers for accepting and recording this session.

At Four Kitchens, we use Scrum to manage most of our projects. In doing so, we’ve learned to break down business requirements into thin vertical slices of functionality, usually written as user stories. This way, teams can ensure that planning, frontend components, backend functionality, and user training/demos come together in sync by being worked on with a unified focus, even as requirements fluctuate through the life of a product or client engagement.

Estimating these “pieces of work” by using story points side-steps the pitfalls of granular, hours-based estimation and equips Product Owners/Project Managers and Stakeholders to make informed prioritization decisions. Estimating is never perfect, but over time, a team will tend toward a predictable velocity which can be used to help plan sprints or releases.

After talking about these concepts and how to put them to work in theory, we will look at how Jira supports this method: using story points for sprint building, using reports for longer-term planning, and how sub-tasks can help aid a split workflow situation like component-driven design or the classic “backend vs. frontend” argument.

These are the two concepts I introduce to all my teams and clients as projects begin, so knowledge of Scrum or Kanban is not required! I encourage developers and managers alike to join for this conversation because the key to a successful project is a shared understanding of how it is structured. These techniques work for Scrum or Kanban style agile development and are not specific to Drupal or particular project management tools.