Jim asked me the other day why we (and most strategy consultancies) conduct all our mainline strategy projects in three phases. Really, it was a just for fun kind of question. We learned early in our strategy training that it makes a lot of sense to start with some kind of assessment (where are we now), then move to a strategy (where are we going) and wrap up with a plan (how do we get there). Projects are extremely different – different durations, activities, deliverables, people involved – but they almost always follow these basic three steps.

We run across clients who want to skip one of the phases, but it is pretty easy to make a case for doing all three. Without an assessment, you aren’t aware of all the issues (many of which are rooted below the surface) or the resources you have to work with. Without the strategy phase you have no overall vision by which to make difficult technology choices. Without a plan, people aren’t aligned about exactly how to make things happen.

Jim wasn’t really questioning why we do all the activities in the phases but more why we have to structure it in time as such. We pondered whether projects were closed off to new information by calling a phase complete. What we came to after a little healthy debate that really wore out his dog (since it resulted in a longer walk) was that a strategy project should be like a funnel, where you start with a hypothesis but a broad one and gradually hone in on the final recommendation. The checkpoints along the way (i.e. the end of the phases) not only put some rigor around the process to ensure you are making progress to that end, they also ensure the right amount of input and buy in of key stakeholders along the way. If you get agreement on what the problems and opportunities are, as well as the big picture of how you want to address them, the actual projects that need to happen are easy. And interestingly enough, we find that the hardest point to get alignment on is the assessment (counterintuitive to many since they think it is the most straightforward) and the easiest point is at the end of the plan (assuming we’ve had the right checkpoints along the way).

So the answer is yes. There are really good reasons to structure strategy projects in these chunks and in this order – although overlapping phases can be both necessary and beneficial. So for now, we will stick with this basic construct and find more interesting problems to tackle…