A project manager trapped between his manager and his manager’s manager

Dear Bob …

I lead a team of developers. We are working on a project that I suspect is a deathmarch. My manager is very smart, but he has a deep personal attachment to the project, and he is a bit of a prima donna.

His manager (a C-level) also suspects a problem, and he has asked me and some of my peers to sit in a meeting with him and my manager where we will try to see if we can reduce the scope of the project in order to make it more likely to succeed. I agree that the project does need to be scoped down.

The problem is that I am quite certain that sitting in this meeting will be career suicide for me — my manager is very attached to the project (some of my peers and I have breached the subject of scoping down — he didn’t take it very well). Also, he carries a great deal of clout and regardless of the outcome of the meeting, he won’t be leaving. There is a great deal of personal baggage between the C-level and my manager; I believe that this will not be a pleasant conversation.

If I oppose my manager, I’ll be caught in the aftermath — he’s not leaving, so I’ll likely either be dismissed or eviscerated. If I side with him, I’m equally compromised, as I will have sided against his manager (and I’d be lying — it’s very clear that we need to change something. I would tell you how long this project has been running, but it’s embarrassing)

I’m beginning to think that my only option is to abandon ship. That would take me off of a certain deathmarch, and I would avoid a personal conflict between myself and either my manager or his manager (I have been with the firm for a long time, and it’s a small community). It’s not the greatest time to leave, and I would prefer to find some way to salvage this situation. I just don’t know how.

– Feeling like a lemming

Dear Arvicolid …

This sounds like one of those situations that calls for a creative solution that lets both sides win, and you to leave the room with your skin attached. Fortunately, I’m pretty sure there’s a way to put one together.

But first: How you go about it is more important than the solution itself. In the meeting you have to find a way to let your manager and his manager spend time hashing things through, with you facilitating more than participating. In other words, your most important role in the early stage conversation is to clarify points, not to propose any of your own.

When the time comes to intervene, what you’re going to do is to give both parties what they want, but not exactly as they want it. You’re going to do it by recommending that the company break the current project into chunks, connected in a roadmap of successive releases.

The first release is scoped down to the bare minimum of functionality required to be useful and architecturally coherent. From that point forward there are two ways to go. One is to use a Scrum-like approach of monthly releases, with an application steering committee (or a single application owner if that’s politically acceptable) deciding what new features go into each monthly release — or in Scrum terms, managing the backlog.

The alternative is more project-like, delivering bigger chunks of additional functionality at longer intervals — three to six months is a reasonable range.

Which should you choose? Let the two managers reach consensus on this. So far as how to choose, I’d advise monthly releases if all additional features are small and granular, looking a lot like individual enhancements; and bigger chunks if the additional features look more like modules of significant size.

Regardless, the best way to organize any software project is to arrange the schedule so the business receives benefit as early as possible and as often as possible. The Big Bang might have been a terrific way to get the universe started, but it’s a lousy way to deliver software.

– Bob