Changing from 2 week Sprints to 3 week
Hi Folks,
First time posting so go gentle!
I am a Scrum Master on a large Media project with devs in Germany and UK, The PO is based in Germany too. I am new to Agile since July and still learning but fortunate to be working with a very high achieving team. Our bi weekly Scrum Events in London require the PO and Tech Lead from Germany to come to London and have been doing this for 6 months. Last week the PO asked if in the new year we can change to 3 week Sprints and moving scrum events to Tues/Weds from Mon/Tues, the request is probably driven out of work life balance issues. However I will try and establish what exactly is the driver for the request next week.
We are a high performing team at the top of the curve and have great flow now after 6 months, I personally will not support the switch although I do support the switch to Tues/Weds if this helps ease the balance.
My question though is, does the PO have any jurisdiction to change when Scrum Events happen? From my POV I own the process so therefore surely I get the casting decision on if we change when events happen. Or, do you offer to the team to discuss in Retro to agree any potential changes?
I know this will probably lead to more questions but fire away, any advice is welcome.
Thanks
Paul
Rather than seeing yourself as the owner of the process, think about how you can help the Scrum Team to take ownership of what is theirs. This means coaching them to inspect and adapt their Scrum implementation.
Broadly speaking there are three things to take into account:
- the demand exerted by the Product Owner for releasable work, and for opportunities to empirically inspect and adapt the product in the context of its business environment
- the ability of the Development Team to actually develop and deliver work at a given cadence
- the need to synchronize with other teams and/or events where irreducible dependencies exist
Hi Ian
Many thanks for coming back quickly as I now have something to shape discussion on Monday. I would say points 1 and 2 are most relevant however 3 does apply. I will most certainly encourage the empirical approach as this has yielded best results so far.
Thanks
Paul
Have the team experimented with conducting the events remotely?
That certainly does have some drawbacks, but if travel is a burden, perhaps another solution would be to come over every 4 weeks, and conduct events remotely, 50% of the time.
There is value in conducting experiments that are easy to reverse. I would imagine it's easier and cheaper to conduct remote events once or twice, than booking several flights in advance on a 3 week cadence, only to find that the 3 week Sprint length does not work well.
I would like to support the points above.
One critical question that comes to mind: is this "large Media project" operating as a single-release type sort of work or -quite the opposite- are there regular releases (albeit to a small number of internal users maybe) at the end of the sprints - 2 releases/month? Hopefully it's the latter.
Now, with that (regular releases) in mind, I think it becomes evident that reducing the time to "market", feedback and learning opportunities, with 50% (2 weeks > 3 weeks) is detrimental to the product itself.
So, rather than artificially increasing the sprint length, it is advisable to addres the underlying issue by agreeing on compromises and alternate paths. Sure, physical presence is always the best, but, as pointed above, virtual presence (camera & sound) could do the tricks every other 2 weeks.