Don’t Get Blinded by the Numbers

Businesses love numbers. No strategy proposal is complete without a detailed spreadsheet translating recommended actions into profits. The ability to cook up those spreadsheets is one reason bright young managers go to business schools like Harvard and Stanford. They come out with finely honed quantitative skills and set to work applying them.

Over the past couple of decades this management-by-numbers game has gained currency. The huge amount of data captured by IT and the growing sophistication of econometric modeling encouraged nearly everyone to believe that a firm’s success was driven by the quantity of its data and the ability to model them. I’m reminded of Isaac Asimov’s science fiction trilogy in which a brilliant mathematician creates a model that charts the future of humanity.

Midway through Asimov’s series, though, the model breaks down—and the same thing is happening in the real world, forcing educators to rethink how we’re training tomorrow’s business leaders.

More and more we’re coming to see that strategy is as much about interpreting as it is about analyzing. When we look at the prospects of a car like the Mini Cooper, we must not only measure its fuel consumption and interior space but also take into account the beauty of its design and the exhilaration of driving it. We can quantify the revenue opportunities from impinging on the privacy of Facebook users, but we can’t measure the betrayal those users feel.

Although factors such as design and trust can’t be reduced to numbers, they can be interpreted and understood. In fact, only by understanding them can we make meaningful distinctions between alternative strategies or predict customers’ emotional responses to a change in direction.

The best executives grasp this reality well. Take the former P&G CEO A.G. Lafley, who attributes the most successful decision of his career to a qualitative judgment. In 1990, as the general manager of laundry products, he had to decide whether to move to compact detergents, which were transforming the market in Japan. The new format would cut P&G’s production, packaging, and transportation costs as well as retailers’ shelf-space and warehouse needs, all while maintaining revenues.

Unfortunately, analysis of consumer surveys suggested that only a small percentage of customers preferred the new format. But Lafley dove into the qualitative information, reading comments from hundreds of consumers. He found that although only a few expressed a desire for compact detergents, the rest were indifferent—not negative. He could make the switch after all, pleasing the small minority without upsetting the majority, and delivering a big win for P&G. “The analysis never tells you the answer,” he concluded. “The best it can do is to inform your judgment.”

Lafley isn’t alone. Increasingly, firms are putting aside data-driven approaches and borrowing from disciplines like ethnography. They study consumers interacting with products. They give consumers cameras and diaries to record their experiences. They even visit consumers’ homes.

This kind of approach requires completely new capabilities. The successful strategists of the future will have a holistic, empathetic understanding of customers and be able to convert somewhat murky insights into a creative business model that they can prototype and revise in real time. To do all that, they’ll have to be good communicators, comfortable with ambiguity and ready to abandon the quest for certain, single-point answers.

That’s what business school deans like me have to work on teaching in the years ahead.

by Roger L. Martin - Harvard Business Review