My First Agile Project Part 14: Did We Need A Coach? Does Anyone?

December 8, 2008 by mattgrommes

Picture courtesy of unoguy on flickr

In the previous part of My First Agile Project I talked about the recent tempest in a teacup regarding the failure and fall of Agile. A commenter on that post ended up crystalizing something I thought during the reading of various posts on the subject; should we have hired an Agile Coach to help us start going on Agile? This is an important question for me in trying to figure out what we could have done better as well as for other teams starting out on Agile as well. This is an issue for Agile itself as well, whether a coach is needed for teams to succeed or not. Read on for more on this question and whether I think all new Agile teams need a coach.

When we started this project, our vendor brought Scrum to us. They recommended we buy Ken Schwaber's Agile Software Development with Scrum and had one of their people give a presentation about Scrum. After that, we were on our own. Most of us read the book, me included, but none of us apparently internalized all the important parts of Scrum. As I've mentioned before, we didn't do a good job of keeping track of our backlog or estimating things to know how much work we were adding to the project. These two omissions were the main reasons we got offtrack so badly without knowing it.

Would a coach have helped us with this?

It only really became an issue after 6 months or so when we'd added a ton of work to the backlog. When we started we thought we were doing really well with Scrum and honestly I think we figured if we got off the path our vendor's people would help us back on. But I think they were as new to the process as we were. So if we'd had a real coach in the office to help us start, I still think we would have gotten off track eventually unless that coach drilled into us the importance of tracking the backlog over time like we should have been doing. I haven't (yet) gone back into Ken Schwaber's book to see what it says about grooming the backlog but if we had just read more about the Scrum process wouldn't that have done the same thing? So I'm not convinced that having a coach in for the start of the project would have helped.

What I think would have helped was a real hard look at the entirety of our process after 6 months or so. We should have done a "Half-way Retrospective" to re-read the book and do a hard examination of what we were doing. That's the point where I think a coach might have been more useful. It's hard to look at your own processes so sometimes an outsider, especially one with deep knowledge of how the process is supposed to go, can come in with fresh eyes and get you back on track. Maybe 6 months is a little too long but the point is that some time has gone by with the team working hard on the project. In my opinion the process of Scrum is easy enough that a small-ish team should be able to pick it up without outside help. After some time has gone by though, a correction might be in order, which is one of the main tenants of Agile anyway.

The question of having a coach at the beginning of the project is where I think coaching proponents, and the Agile community in general, needs to look. If you're saying a project needs a coach from the start, what are all the books for? Why is the process so complicated that it needs a coach? If you're starting a hundred-person organization on Scrum and need to do Scrum-of-Scrums or something out of the ordinary like that, then yes, a coach with experience in that process is a good idea. But for a smaller team, an internal IT department like us or a small software shop, I don't think a coach is needed. Or maybe I should say, a coach shouldn't be needed for the Scrum process. You might need a coach for making your team more cohesive or to help with management issues that might derail the Scrum process, but not for the process itself. If your team can't make a good go of it by reading a book like Ken Schwaber's, there's probably something else wrong and all a coach is going to do at that point is help point that blocker out. Like people say, Scrum helps expose problems in the organization that other processes might have hidden.

During all the recent brouhaha one of the criticisms of Agile I read over and over is that the pushing of coaching means the process is just a money-grab. I don't think this is the case at all and it's a pretty offensive accusation, actually. But the fact is that this notion is out there and it's something Agile should be confronting. If 'hire a coach' is the answer people keep hearing to solve their problems, it starts to look like either the process is too hard or the dreaded Consultants are just trying to attach their suckers onto our budgets. Maybe the Agile community needs to come up with a way of approaching the importance of coaching. A cohesive statement about what teams should think about when starting the process; when you probably just need X, Y, or Z book or when you should talk to a coach about your process up front. A response from the community in general or just from the "thought leaders" of Agile would be useful I think and clear up some of the confusion. If you're trying to sell Agile to your boss, a survey of your company that helps lead you to a book or coach seems better than starting out with "first we'll need to bring Coach X on for a week".

Obviously, this being My First Agile Project, my experience is limited. But I have read a lot and gone to the Agile conference so I'm not completely talking out of my hat here. If you think I'm wrong or misguided, please let me know in the comments. I hope I've made clear that I think the question of coaching is one that needs a discussion, not just to be blown off. Thanks for reading and I'll be back next week!

Permalink