CNET también está disponible en español.

Ir a español

Don't show this again

Tech Industry

The perils of great expectations

Why do we try and sum up the performance of a company in a number or two? Perhaps because we like clear indicators of who is a winner, and who is a loser.

It seems we're all just searching for a way to keep score.

The beauty of sports is that there is always a winner. The beauty of life is that it's not like sports. Sometimes you win, and sometimes you lose. Sometimes you get two-thirds of what you expected after leaving something on the table to get that far, and sometimes you fall short of a goal but find yourself better for the experience.

That's why we have sports, to amuse ourselves with contests that fulfill a need for clearly defined winners and losers but that don't turn the world on its side or redistribute billions of dollars. We expect our teams to win, and we're saddened when they don't (thanks a lot, Yadier Molina) but there's an inherent assumption that's it's all just for fun.

Back in the real world, however, it's a little messier. The sad admissions made by Dell yesterday underscore how those lines can blur.

Dell revealed Thursday that its internal accountants gamed a system--with the knowledge of senior management--designed to reward companies that hit their numbers. It's usually a single number that determines the stock market's reaction to a financial earnings report: earnings per share, or net income divided by the number of shares outstanding.

No matter what positive news might be contained in an earnings release, if EPS comes in even a penny below estimates, watch that stock price fall the next day. Just last month, the current darling of the tech stock crowd--Google--reported earnings per share 3 cents lower than expectations. Despite a 58 percent jump in revenue, a 28 percent increase in profits, and a whopping overall EPS of $3.56, the company's stock fell 5 percent the next day, and it hasn't recovered.

Everyone's guilty in this game. The stock market buys and sells based on that number, perhaps the revenue figure as well, but with little regard for the broader picture. A media report will talk about a company's brilliant execution, or start looking for scapegoats, based on a 2-cent swing in EPS. Executives are heavily compensated with stock options that make them obsessed with keeping the price as high as possible.

The most egregious periods of Dell's earnings manipulation prevented the company from missing its earnings per share targets by a few cents. Would that have taken away from the fact that during that period, Dell's overall performance was still far stronger than the rest of the PC industry, and maybe the entire tech industry? Dell executives apparently thought so, and they probably would have been right had the real numbers been reported.

This, of course, doesn't excuse Dell's actions, which illustrates the thorniness of the problem. Missing those targets would have served as a warning sign for investors that the company was fallible after all, perhaps giving them some time to bail out before Dell's myriad problems were unable to be concealed by creative accounting. Some metric is clearly needed to assess performance, but what?

Another reason to love sports is statistics: clear barometers of who is good, and who is not. If you get a hit 3 times out of 10, complete 60 percent of your passes, or put the ball in the hoop 2 times out of 5, we like you.

The problem is we're doing the same thing with massive corporations, organizations with thousands of employees and customers whose performance can't be summarized by a single number, even two. And the reward--or punishment--attached to those numbers doesn't correspond with reality.

I'm not smart enough to have a solution at the ready. But let's take the sports out of business. There are always going to be winners and losers in a capitalist system, and investors need to have a way of measuring whether they are getting a proper return on their outlays. The business community needs to figure out a broader way to draw conclusions about a company than a single number that is apparently very easy to manipulate.

Then, let's take the sports out of politics.