Discussing software development metrics at my place of work, a colleague (Derek Morrison) came up with a neat concept – a way of measuring Business Value.
We’re using Scrum as an agile management framework, estimating in points and measuring Velocity to help plan future Sprints.
For those not familiar with Velocity, it’s the total estimated cost (in effort or points) of the features that are 100% complete (i.e. signed off and delivered) in a Sprint or iteration. If you’re familiar with Earned Value Analysis in more traditional project management methods, it’s basically the same concept.
Each feature, or user story, listed on the Product Backlog has two Fibonacci scores. One for Size and one for Business Value. This is potentially a useful way to prioritise the Product Backlog, as a simple formula (value/size) can be used to order the backlog with the quickest/biggest wins first.
Derek’s idea was to measure the Business Value of what we’re delivering, as well as Velocity for the team’s planning purposes. We referred to this metric as Business Velocity.
We haven’t started using this metric in our teams yet, so it’s too early for me to comment on how it’s working in practice. However I was very interested to see James Shore’s comments on the same.
Personally I think it has merit. So long as – as with all metrics – people don’t play the system.