Much of what a CFO does can rightly be described as science: numbers, spreadsheets and reports all live in the empirical, analytical realm of cold, hard facts. While much of accounting does involve making assumptions, there is one part of the CFO's job description that might more accurately be described as art than science. 

Managing analysts' expectations is about color, texture, and the chiaroscuro of suggestion, mandated, of course by the rules that govern fair dealing and access to information. Yet research released recently by McKinsey shows that over the past 25 years, analysts' expectations have been consistently over optimistic, especially when it comes to earnings forecasts.

The report shows that most years, analysts predict corporate revenue growth of between 10 percent and 12 percent, when in actual fact the average growth comes in at 6 percent. Moreover, only twice in 25 years have forecasted earnings actually been less than actual earnings, in the early 1990s and in the period 2003-2006, both periods of strong general economic growth.

The McKinsey study also shows that since the introduction of Reg FD in 2001, there has been no real increase in the accuracy of analysts' forecasts. Moreover, more companies are actually shying away from giving any guidance at all. The business structures under which equity analysts toil have been widely criticized and are largely to blame for this chronic lack of accuracy. But perhaps CFOs also need to look at what their role might have been in this farrago.

Companies are obviously keen to paint their business in as positive a light as possible. And natural human nature is quick to err on the side of optimism rather than pessimism. But what the report shows, is that when the world economy takes a massive tumble, as it did in 2000-2001 or in 2007-2008, companies are very slow to realize the massive impact it will have on their own business.

CFOs naturally have a very good idea of how their companies are doing in and of themselves; but they seem less clear about what major global downturns will do to their numbers. Perhaps after the sobering effects of this recession have worn off, CFOs need to take greater heed of the world around them. In that way, the messy business of meeting analysts' expectations can become more scientific and less artful.