We’ve found that we keep neglecting projects that have high payoff, but aren’t urgent. You know the kinds of things - contributing to open source, building tools for internal use, improving our skills. At companies like 3M, Atlassian, and formerly at Google people have used 15% or 20% time to start projects that have turned into valuable business opportunities such as Post-it notes.
So we decided to try it. Jonny came up with a manifesto:
- that many important, valuable projects are never the top priority, meaning they are never completed.
- that in our current system there is not enough time for activities such as developer education, freedom to innovate and building our public reputation.
- that the business will benefit from a more innovative, educated and motivated engineering team.
- that developers’ enthusiasm for a project is precious and that they can be trusted to use their enthusiasm on projects with real business value.
- that the workload for each sprint should be reduced by 20%.
- that developers are expected to spend up to 1 day a week to work on their own projects.
- that developers are expected to complete what they committed to in the sprint.
- that developers are expected to demonstrate in stand-ups and retrospectives what they have achieved during their 20% time and record their projects on the wiki.
- that developers are free to choose what to spend their 20% time doing, as long as they can demonstrate business value in stand-ups and retrospectives.
We can measure our success…
- with our sprint velocity, as we believe more educated and enthusiastic developers work faster.
- with the number of new, innovative projects and the business value that they generate.
We’ll hold a retrospective in a month to see how it’s gone, and we’ll blog the results here.