User stories part 3: Using spikes to help estimate user stories

A user story containing unknown elements should be split into a spike to investigate the unknown elements plus a user story to develop the functionality. This enables the customer to prioritise the research separate from the implementation of the new functionality. Without a spike to substantiate the estimate for the user story, the customer may incorrectly assume the estimate to be valid and prioritise the user story accordingly. Facing a spike and an associated user story, the customer should prioritise the spike ahead of the user story to obtain a more reliable estimate for prioritising the user story.

Next post in this series:
User stories part 4: Collaborating to write acceptance tests

Previous posts in this series:
User stories part 1: What is a user story and who writes them?
User stories part 2: Adaptive planning

References:
[1] Mike Cohn's Agile Estimating and Planning

Tags: agile, extreme programming, user stories, spike

posted by Simon Baker @ permalink
blinklist : del.icio.us : digg : furl : simpy : spurl : reddit