How should a business be measured?

For a long time, the answer has been “more.” Ever since Frederick W. Taylor did time studies of steelworkers with a stopwatch in 1900, the measurement of business activity – called “Greater Taylorism” by Walter Keichel in his business history “The Lords of Strategy” – has grown ever more central to management. One result of this drive to quantify and analyze has been that senior executives often create numerous profit centers, or isolated groupings of both revenues and expenses nested within large businesses.

The two benefits are obvious. First, profit centers allow these executives to make better decisions. In organizations whose various revenue and cost accounts are not linked, poor economic performance can be hidden by positive results elsewhere, and decision-making is clouded. Second, profit centers help make accountability clear. By giving managers direct profit and loss responsibility, companies can incentivize activity that measurably contributes to the bottom line.

So in most large companies, different business divisions and geographic regions are organized as distinct profit centers. Increasingly, product lines, key customer accounts, or brands are treated as mini-businesses as well, like at Procter & Gamble, where global brand managers have P&L responsibility. For that matter, why not functions too? Some organizations are establishing transfer prices for supplies and services between business departments (such as manufacturing and sales), and then measuring, and rewarding based on, the income of each.

There’s just one problem. We optimize what we measure. And the entire logic of profit centers rests on the assumption that maximizing the pieces will maximize the whole.

Unfortunately, this shortcut often isn’t true. Exceptional, sustainable results derive from great strategy, and great strategy isn’t additive – it relies on the way individual pieces fit together in a system, so that the whole is greater than the sum of the parts.

Decision scientists know this – in their models, it’s the difference between finding the local optima (the best result within each “neighborhood”) and the “global optimum” of the whole system. For that matter, football coaches know it too; no professional coach would argue that the best way to win a championship is to focus on maximizing each individual player’s performance statistics.

Yet this is exactly how many businesses are run. Rather than sacrificing certain parts for the good of the whole, companies essentially force each division to stand on its own. This approach undermines strategic fit, which, as Michael Porter put it, “requires the integration of decisions and actions across many independent subunits.”

For a coherent strategy to work, then, the organization executing it must be measured as a whole, rather than as parts. In other words, if a company is to have a single strategy, it must be driven by a single P&L.

This may sound like an extreme position. Yet some of the world’s most successful companies operate this way. Apple famously has only one P&L, for which its CFO, Peter Oppenheimer, has direct responsibility. And while each of its major hardware product lines is priced to make a significant profit, it bundles in all its key software upgrades, products, services, and platforms for free. CEO Tim Cook explains the logic:

“We manage the company at the top and just have one P&L, and don’t worry about the iCloud team making money and the Siri team making money. We want to have a great customer experience, and we think measuring all these things at that level would never achieve such a thing.”

It’s Apple’s single-company mindset that lets it give away industry-leading software and cannibalize its own products, which in turn has led to its unprecedented success. But that’s not to say a single P&L is always the right answer. Instead, a company should have as many P&Ls as it does distinct strategies. P&G’s Gillette shaving brand has a very different strategy from its Bounty paper towel brand, and Gillette has a different strategy in India than in North America. But although Gillette sells its razors and blade cartridges separately, these products fall under a single strategy. P&G’s profit centers reflect these boundaries.

Of course, companies should still measure a division, product, or function’s profitability (to the extent it can be done accurately) – that’s just good management. But this shouldn’t be the primary basis upon which managers are held accountable for their decisions, or they won’t enact a strategy that looks beyond their narrow interests. Amazon wouldn’t be able to underprice and over-market the Kindle to achieve their larger strategic objective of selling content if the Kindle product manager’s main objective was to maximize hardware profits. Nor would “free” look like such a great price point for Google’s Android unit.

So measure carefully – because if you reward each area of your business for acting in its own best interest, you just might get what you wish for.