Grooming, or Product Backlog Refinement if you prefer, has always been a troublesome, not to mention kinda-boring concept in Scrum.
And in recent years it’s somehow turned into a meeting, which it seems most people hate, but do anything, because apparently they were told to. Apparently unless you don’t do something in a meeting then it’s not Agile (how ironic).
Having practiced Scrum for years without such a meeting I found myself wondering what people were doing in it. Turns out, at least in the cases I observed, that it seems to be mostly a combination of Planning Poker™, followed by demands for more detailed acceptance criteria and then arguing. Whatever floats your boat I guess, but it wasn’t my idea of a good time.
This “Grooming Thing” was also creating some kind of weird division, in what I’d always assumed was a Cross Functional Team. Suddenly the world was divided into The People Who Did The Work and The People Who Prepared the Work for The Other People To Do.
Weirdly, it was only The People Who Did The Work whose time seemed to matter, and was thus tracked and reported on in great detail, the preparation people were apparently free.
It was all beginning to sound like Waterfall, but with stand-ups and some other new thing called JIRA® (very useful for tracking how The People Who Do The Work are spending their days)
I however was keen to get back to the collaboration and cross-functional team stuff. I had good memories, it was both fun and effective. Neat!
But damn, Grooming is tearing us apart. How could we get the band back together?
What we were missing was a A Bright Line for Grooming. A way of describing Grooming that didn’t involved creating Groomers and Groomees.
And then it hit me, it’s so simple.
Grooming is any work you do that results only in a change to the Product Backlog.
Oh, you’re an idiot, I hear you say, we already knew that, that’s not new.
But hear me out. Because a simple example might make all the difference in understanding what I’m talking about.
Imagine a team that starts a story, and for whatever reason (it really doesn’t matter) the story is not finished at the end of the Sprint. We asked for a feature, a deliverable, but we don’t have that. What we have instead is increased WIP. No cause for panic. It happens.
What are the relevant Scrum Rules in this situation?
- Don’t demo the feature (mostly because it doesn’t exist)1
- Don’t put the feature into production (again, because it doesn’t exist)2
- Re-estimate the story to completion and return it to the Product Backlog
So what has the team done exactly? What’s their output?
A New Estimate.
What do we call it when the team estimates a Product Backlog Item?
The Team just spent the entire Sprint on grooming that story.
Wait a minute I again hear you cry (I really should get that web cam fixed)
CODE WAS WRITTEN, TICKETS WERE CLOSED IN JIRA® THAT’S NOT GROOMING!!!!! THAT’S REAL WORK!!!!
That’s exactly my point, Grooming is an outcome not an activity. And above all Grooming is Work. It consumes both time and people3
Sometimes it’s obvious that we’re grooming, but other times we don’t know that we are until after the fact.
In my example, the only difference is that in this case the Team’s Estimate was based on coding, building and testing rather than (say) Planning Poker™.4
I don’t have my feature, but I do have an updated backlog, so you, my dear team spent the entire sprint grooming. ALL OF YOU.
What’s even more fun is that now we can finally make better sense of the Scrum Guide admonition that “A Team should spend no more than 10% of their capacity on Product Backlog Grooming”
It we take this rule into the my context for grooming it’s just saying “Hey Team! Try and spend 90% of your time on finishing things, not just starting or progressing things”5
To me that’s a far more useful interpretation than “Let’s disrupt the Sprint to ‘play’ Planning Poker® for 2 hours every Thursday”
And the best part? There is no longer an us and a them. There’s just us, because you never know when you’ll find yourself grooming.
- But that certainly doesn’t stop people from trying ↩
- If you’ve not finished a story, but you do have a feature you could demo and put it in production then you probably have another problem. You’ve definitely misunderstood a few things. ↩
- and souls if you do it the boring way ↩
- So yes, we’re also closer to finishing delivery of said feature. But the same is true of “other work” that many people relegate to the realm of Grooming, such as designing the User Interface”. It’s no different. Defining the end of Grooming as the start of coding is not particularly useful in my personal opinion. In fact I’d argue that if you’re doing anything to your Backlog Items that doesn’t bring them closer to final delivery, then you might want to think about why you’re doing those things! ↩
- A side effect of this is that you’ll need to start making things smaller if you want to achieve this goal. ↩