When starting a project, one of the choices that have to be made is the iteration length. While a 4-week iteration feels very short, once in the rhythm, it is quite long. In a lot of cases, you wish to receive feedback earlier. Our team decided to give 2-week sprints a try. At first, we thought it would create a lot of overhead. Twice as many planning meetings, reviews, retrospectives…
But because they are simply smaller in size, I have the feeling the overhead remains the same as in 4-week sprints. We now have a stronger focus on the sprint goal, and receive feedback every 2 weeks instead of every month, which is a huge benefit if you have a lot of uncertainty.
To be able to do this, we had to review some practices:
· The product owner must keep the features small in size, so they can be easily fit into a 2-week iteration.
· It must be possible to write acceptance tests in advance (ex. FitNesse).
· The customer has to feel comfortable with the new schedules. He has to be able to attend twice as many review meetings.
· All absences must be known during the sprint planning. Even one day of unplanned absence can endanger the sprint.
Make sure to stick with your choice of iteration length. It is confusing for a team to have to constantly switch. If the length remains the same, it will feel like a natural rhythm, and the end of an iteration will smoothly transition into the start of the next.