What if the MMF doesnt fit within a 2 week Sprint?
Lets say the team comes to an agreement that a Sprint Length should be 2 weeks.
The output of a Sprint Planning Meeting is to have some kind of Sprint Goal and all the items that are needed to achieve this minimum goal.
Now what would you do if the defined MMF (items) doesn't fit in a 2 week Sprint? Scrum clearly describe that the product needs to be potential shippable after every Sprint.
Would you change the length of the Sprint?
I can imagine that Goal driven development can be hard to get inline with a fix Sprint. Perhaps sprint goal a fits in a 3 week sprint while Sprint goal b (next sprint goal) can easily fit in a 1 week sprint.
How is this best managed?
if you slice your pbi's vertically and keep pbi's independent of each other, you should be able to still have a potentially shippable product in the 2week sprint.
the point of having a fixed length is to develop a cadence for the team and to reduce overhead. the point of having a timebox for the sprint is so that there is time to inspect and adapt.
Some mature teams are able to deliver daily even though their sprint lengths are a week or two.
To be potentially releasable each item within the increment must meet the Definition of Done. A good DoD will take each item as close as possible to production readiness. The potential for release has nothing to do with whether or not the increment represents an MMF; rather, it is a function of whether or not each and every item is "done".
To put it another way: since each item must meet the DoD, each increment must (by implication) be potentially releasable.
If the increment *does* amount to an MMF then that potential (i.e. the chance that it actually *will* be released) is increased.
Hi Ian,
For me to achieve faster time-to-market it’s all about delivering potential shippable MMF. In other words how can we deliver minimum Business Value as soon as possible. For me the MMF consists of x increments that are delivered according to the DoD.
That would fit the term Minimum Market Feature, because if the feature is not potential shippable, than it’s also not suitable to go into the market. So for me, the MMF is also the potential release candidate.
I'm wondering how you think of this way of thinking.
Yes, an MMF will be a potential release candidate, but so is any well-formed increment in which all of the items meet the Definition of Done.
All you seem to be doing is raising the bar higher, and making a value judgement about feature sets. That's fine. Scrum isn't prescriptive enough to care about feature sets or MMF's.
My concern is that you appear to be phrasing Sprint Goals in terms of MMF's. That's a problem, because there isn't necessarily a direct correlation at all. It would be more realistic to phrase the goals in terms of the contributions they make to an MMF, and which would then will meet your own standard for release. I admit that this is hard to do, but the ability to craft meaningful Sprint Goals is a core skill in Scrum.