Good Strategy/Bad Strategy: The difference and why it matters

by Richard P. Rumelt

The logic of the situation is counterintuitive to many people—the faster the uptake of a durable product, the sooner the market will be saturated. Many managers find these kinds of forecasts uncomfortable, even disturbing. As a client once told me, “Professor, if you can’t get that bump out of the forecast, I can find a consultant who will.”

The problem at AT&T was not the competence of individuals but the culture—the work norms and mindsets. Bell Labs did fundamental research, not product development. The reaction to a request for demonstration code was as if Boeing engineers had been asked to design toy airplanes. Just as in a large university, the breakthroughs of a tiny number of very talented individuals had been used to justify a contemplative life for thousands of others.

The first step in breaking organizational culture inertia is simplification. This helps to eliminate the complex routines, processes, and hidden bargains among units that mask waste and inefficiency. Strip out excess layers of administration and halt nonessential operations—sell them off, close them down, spin them off, or outsource the services. Coordinating committees and a myriad of complex initiatives need to be disbanded. The simpler structure will begin to illuminate obsolete units, inefficiency, and simple bad behavior that was hidden from sight by complex overlays of administration and self-interest.

The problem with treating strategy as a crank-winding exercise is that systems of deduction and computation do not produce new interesting ideas, no matter how hard one winds the crank. Even in pure mathematics, the ultimate deductive system, stating and proving an interesting new theorem is a profoundly creative act.

Good strategy work is necessarily empirical and pragmatic. Especially in business, whatever grand notions a person may have about the products or services the world might need, or about human behavior, or about how organizations should be managed, what does not actually “work” cannot long survive.

The concept of cost is tricky. People talk as if products have costs, but that is shorthand easily leading to confusion. Choices, not products, have costs. The cost of choosing to make one more unit of a product is sometimes called marginal, or variable, cost. The cost (per unit) of choosing to produce at a fixed rate for a year is called average cost. The cost per unit of choosing to build a plant and produce at a fixed rate is called long-run average cost. The cost of filling a rush or special order has no particular name, but it clearly exists. There really is no such thing as the single correct “cost” of a product. It all depends on the decision—on what is being compared to what.