How to balance centralized and decentralized planning

The challenge is in mixing the correct amount of top-down analysis to provide the shape of things and set the goals, and bottom-up analysis to fill in the details

Dear Bob ...

Here's a question for you: Why does planning work for a corporation but fail miserably in governments? Or does it work for corporations? If so, when?

[ Also on InfoWorld: "The world's quickest course in project management" | Get sage advice on IT careers and management from Bob Lewis in InfoWorld's Advice Line newsletter. ]

We all believe that Communism failed because central planning does not work. Clearly, a country is too big to run by (detailed) planning.

We know that if a project is small enough, planning is totally a waste of time. That is, Nana needs to write a letter. No planning -- get paper, get pen, write. No Gantt charts. No management reviews.

Somewhere in the middle, most of us think that planning is useful. Is it? If so, when, where, how, and how much?

What are the bounds? How big is too big to plan? How small is small enough to reject planning?

- Balancer

Dear Balancer ...

Great questions. A few thoughts:

First of all, the post-Vietnam military has shown that it's possible to run a government agency effectively, and that includes planning -- not perfectly, but effectively. Without a lot of detailed knowledge, I'd hazard a guess that the U.S. military runs quite a lot more effectively than the average Fortune 100 corporation. It's possible. It just takes great leadership, which the military emphasizes and most other government agencies don't.

Which doesn't have much to say about your main question -- how to find the right balance between top-down and bottom-up planning.

I think the best planning requires top-down analysis to provide the shape of things and set the goals, and bottom-up analysis to fill in the details. The challenge is to balance the importance of coherence against the importance of tailoring solutions to the details of the situation.

The dividing line depends a lot on the nature of the beast. It's a lot smaller with an ad agency, for example, than for a manufacturer: A small team of engineers, and sometimes just one, can completely design a product and the manufacturing line needed to produce it. Every ad campaign, in contrast, is unique, and agency teams need a lot of creative latitude.

Which means in manufacturing, central planning can manage a much bigger company than in advertising.

There are too many determining factors to allow me to develop a quantitative answer (one might be possible; I'm not smart enough to develop it).

Instead, I'd say a good way of thinking about this would be to look at it in terms of matching the level of planning detail to the ability of one person to understand the situation.

So an executive who maintains control at a level of detail he or she doesn't understand in detail is making a bad mistake. But so is an entrepreneur who gives up too much control of situations he/she does understand in depth.

Put it differently: For me, at least, the time to give up on centralized planning is when trying to understand what's really going on out there starts to make my head explode.

Make sense to you?

- Bob