Two Keys To More Effective Story Point Estimation

Some teams find story point estimation frustrating or feel it is an arbitrary exercise. In years of working with Scrum teams, I’ve found two things that help them more successfully and effectively estimate using story points. Number one, establish (or dust off) your definition of done. And number two, find at least one reference story.

Establish or dust off your Definition of Done

A Definition of Done (DoD) sets expectations for what needs to be completed before the work can be considered done within a sprint or iteration. It includes all of the development, testing, documentation, integration, automation, or other activities the Scrum team needs to complete.

Ideally, the Definition of Done brings the team’s deliverable— new code, modified code, a part of Enterprise Resource Planning integration effort, a business intelligence report, etc.—to a potentially shippable state. The work could potentially be shipped assuming there is enough value to the customer in receiving it. Having a Definition of Done aligns everyone on the team about the aspects they need to discuss when estimating a new story.

Find at least one reference story

A reference story is an existing story where the team understands what it took to complete. This story is then placed on the chosen scale. e.g., PBI #510 is 5 points on the Fibonacci scale. Other stories are then compared against this story.

Are they the same effort/complexity? More? How much more? Having one or more reference stories helps the team align and give them a base against which to compare future stories.

A Definition of Done and a reference story help ground the team in the work that needs to be considered when estimated and give them something concrete to compare future stories against.