Are sprint commitments necessary?
I’ve had a draft called “Is scrum the best we can do” lying around for months and the chance to finish it is tiny. So I decided to extract parts of it: Some thoughts on whether sprint commitments are beneficial or not.
TL;DR: Maybe, it depends.
There are multiple arguments in favour of commitments. I’ll restrict this post to only looking at one: Focus.
One of the main advantages in favour of commitments is the increased focus. There is research that confirms what most people probably already know: Deadlines are a great way to fight procrastination.
The question to that is if the increased focus is worth it. Some things to consider:
People who are self driven tend to already have a high level of focus.
Turning up the pressure lever has a detrimental effect once the pressure gets too high. Leading to unhappy employees and worse performance.
Focus on the committed story might prevent people from trying out unknown solutions or more risky alternatives. This is one of the things the commitment is designed to prevent after all. But you’ve to be aware of it because it might be detrimental to the long term success of your company. Something I read a while ago that has been nagging me because I’m not sure to what degree it is true is “Innovation doesn’t happen by implementing the stories from the backlog”. I think it’s worth keeping in mind.
Due to team dynamics there is already some sort of commitment on a daily basis. Team members tell others what they’re working on, and the others expect to see results of that eventually.
I’d say whether commitments are beneficial or not depends a lot on the individuals in the engineering teams. Did they sign up for the overall cause? Do they believe in the product? Do they want to implement the stories on the backlog for more than the technical challenge they present? Are they the type that get things done and out the door?
Or do they easily get lost in unimportant details, looking for a perfect solution that doesn’t exist?
One lesson that most engineers eventually learn is that there is no one size fits all solution. It is time that we learn that the same applies to processes as well.