Let’s say this was our output for the previous Sprints:
- Sprint 1: 73sp
- Sprint 2: 110sp
- Sprint 3: 98sp
- Sprint 4: 131sp
- Sprint 5: 122sp
In this case, we can say that on average, we have delivered 107sp per Sprint. This is what we call “velocity”. Our velocity at the end of the 5th Sprint is 107sp per Sprint.
So, we can come up with these pieces of information:
- How much work should we commit to for the next sprint? Something around 107sp.
- We have 1000sp remained in our Product Backlog and we don’t think we can come up with new ideas for the backlog. When are we going to finish the project? 1000sp / 107sp per Sprint = about 10 sprints
Velocity always changes of course. For example, if our output is 140sp in the 6th Sprint, our velocity becomes 112sp. This is an opportunity to adjust all calculations and come up with more realistic forecasts.
It’s normal to have large changes in the velocity in the beginning of the project. It becomes stable after five to eight Sprints.
So, we know how to estimate, and who should estimate. Let’s say there are 6 developers and they are supposed to estimate. They have understood the item based on the explanations from the Product Owner, and they have discussed the techniques and approach. Now what? Should they vote one by one?