Christmas sprint quandry

[Jason Field]

Hi all,

I have a question I'm thinking about at the moment which might be of interest to the group. We have been running sprints for a while now as a tool for charting our development process (via a release burn down chart), and we're getting some useful statistics out to help estimate our deliveries. The question I am pondering is what to do about Christmas? Since everyone will be out of the office for about 2 weeks during the holidays, how do I factor that into our team's velocity? If I cut the 2 weeks for Christmas out of the timeline, will I be getting a fair reflection of the impact of holidays on the development schedule? Or do I schedule a longer sprint (4 weeks rather than our usual 2), and take the hit on our velocity to more accurately gauge our progress long term?

My current thinking is that route #2 is the way to go to ensure the long-term statistics stack up, but I'd love to hear if anyone else has come up with a good solution.

Any thoughts gratefully appreciated :o)

Cheers,

Jason

[chrs]

Jason,

I find that each sprint is different anyway, with varying amounts of
things distracting the team: support issues, courses, holidays,
sickness etc. So when planning the next sprint I look at how many days
were lost due to all this stuff and take a guess at how many will be
lost in the upcoming sprint too.

It means that your teams "true" velocity is how much you could acheive
if all that external stuff didn't get in the way. Henrik Kniberg
mentions this in his (freely downloadable!) book:
http://www.infoq.com/minibooks/scrum-xp-from-the-trenches
He calls this "Focus Factor".

So I would say that the Christmas period is just a time when there's
more than the usual amount of stuff distracting your team from the job
in hand...

HTH

Chris