Thanks for your response Tom!

I agree that extra complexity is rarely a good thing, although my experience of using this approach has been welcomed by both those performing and those receiving the estimates, and actually I believe it has increased communication, as open and shared consideration is given to important influencing success factors.

Of course, if a mature team is working very happily and effectively with traditional scoring mechanisms and the scores provided are meeting the business need then it wouldn’t make sense to change it (unless the team are keenly interested in experimentation). Teams change over time however — in some environments more than others (but that’s a whole different topic)— and the maturity of that team can quickly regress as team members come and go, so dependency on maturity could be risky.

I have not suggested that the estimates are done by those not performing the work, but equally I would not assume that that persons who pick up the work were necessarily present when estimating was done. I have, however, often seen estimates done by individuals, with or without knowing if they will be the one to do the work, and it is in this case that I see a particular risk for time estimates to be given.

I agree about converting historic velocity to time. My article does stop short of detailing approaches to translating the data into costs, as this is a vast subject in itself and I didn’t want to dilute the article. The point my article aims to convey is that this can be greatly inaccurate when only looking at a single data point.