I had a conversation the other day with a senior ScrumMaster based in Europe. He had failed to see why I had persuaded my teams in the past to adopt a minimum viable product (MVP) approach to Agile. My reply was simply, "Risk management."
He didn't understand why risk management, so I described a likely scenario: What would happen if he multitasks, and as he's getting getting closer to the end of the sprint, for some bizarre reason the environment crashes or something else detrimental happens? He indicated that he had not experienced such a situation before and that he is a pro at multitasking (alarms went off!). I have been in IT since 1995 and cannot count the number of times that an environment has gone down inexplicably.
I have good reason for applying the MVP approach. Suppose you have a couple of PBIs (product backlog items) in the sprint, and you get the team to collaborate and work on one PBI at a time (also known as swarming). With all hands on deck, the work will soon be done during the sprint. And during this time, a few things are likely to happen:
- The team will have one or a few PBIs to show during the show-and-tell exercise.
- The amount of uncompleted PBIs will be reduced. (It is not as if the uncompleted ones will automatically be prioritized to the top of the product owner's list come next sprint planning exercise.)
- If something dire should happen, there is something to show to the stakeholders during the review, thus making them happy.
If you decide to multitask all the way through the sprint (I am not saying that this is not plausible), it can work, but there is still the risk of something dire happening. Maybe the environment goes down, and you haven't finished one PBI. What happens then? It's almost guaranteed that the stakeholders will not be happy.
I told the ScrumMaster that I will always be an advocate of MVP -- until I see something better come along. As the Spanish proverb says, "A wise man changes his mind, a fool never will."