Typically, worrying about customer service isn't a job for the CFO, but Reichheld thinks it should be, and thanks in part to work done by Reichheld himself, things could be starting to change. Quick, simple surveys are being used by a growing number of companies to keep tabs on whether customers are happy or not. As a result, this vague and fluffy subject is gaining a much sharper edge. In some cases, the results are being used as a way to forecast future sales or profits or make hard-nosed decisions about pay and promotions. Professional investors are also deploying these new metrics to evaluate opportunities: In one case, they actually justified a $90 million investment decision.

The biggest name to convert to this approach is corporate trend-setter General Electric Co., which ran a presentation on the subject for 300 Wall Street analysts on July 26 (see box). Software company Intuit, which sells a range of tax software that includes QuickBooks and TurboTax, switched on to the same approach a little earlier and the customer satisfaction numbers have become a key metric for the company's finance organization, says Robert Lawson, Intuit's vice president of investor relations. "We know that word of mouth is our most important method of generating sales, so our finance organization tracks the customer numbers closely as a way to anticipate revenue growth in the longer term," he says.

It's about time, according to Robert Shaw, a professor of marketing at Cass Business School in London and director of the Value Based Marketing Forum. He argues that finance organizations have a responsibility to try and understand what is driving their companies' revenues and profits–a task they have tended to shirk in favor of the far simpler job of analyzing and controlling costs. "They don't see much of a pattern in the revenues that they can pull apart, which gives rise to a lot of misplaced fatalism that not much can be done to influence revenues anyway," he says.

What companies have lacked is a credible metric, says Bain & Co.'s Reichheld–something that could be used to hold businesses accountable, provide staff with incentives for improvement and act as a proxy for future sales growth. "Everyone with a brain in their heads knows that customer satisfaction is a key component of value. They just don't have a good way of measuring it," says Reichheld.

That may not sound like a controversial claim, but Reichheld is a controversial figure at the moment. At Bain & Co., he developed a simple measure of customer satisfaction called the Net Promoter Score (NPS). This is the metric now being used by GE and Intuit, among others. The growing popularity of NPS has made Reichheld public enemy No. 1 for the market research firms who conduct the lengthy, complex customer satisfaction surveys that have, until now, dominated the scene–but, Reichheld charges, have failed to have any impact beyond the isolated world of the marketing department. Since NPS started to win converts, Reichheld has been rubbished and threatened with lawsuits. His critics have even called partners at Bain to complain about his work. Cass Business School's Shaw–who agrees with Reichheld about the limitations of traditional market research–calls NPS "utter hogwash."

NPS is essentially a customer satisfaction survey stripped of the clumsiness that Reichheld claims has made these surveys such an ineffective management tool. Whereas the traditional surveys can take up to half an hour to complete–and even longer to analyze–an NPS survey has just one question: "How likely would you be to recommend us to a friend?" The respondent is asked to rate their response from one to 10, with any score between one and six being classed as a "detractor" while a nine or 10 is classed as a "promoter." Subtract the percentage of detractors from the percentage of promoters, and you have your NPS. It sounds almost criminally simple, but this simplicity is one of NPS' key strengths, says Reichheld. Complex surveys are so demanding that few customers can be bothered to answer them, making it hard to get a representative sample. They also take longer to conduct and longer to analyze. By the time the whole process is finished, the marketing department has a wealth of information about how a small sample of customers was feeling a few months ago. It may be interesting, but it's not really actionable–which is why senior executives tend to ignore these surveys, says Reichheld. In contrast, NPS surveys can be conducted so quickly that the results are available to individual branches within days. That gives the company a chance to actually do something about unhappy customers.

Enterprise Rent-A-Car has been employing a similar approach (which it calls the Enterprise Service Quality index, or ESQi) for the last decade. Rapid growth resulted in the company overtaking Hertz as the biggest car rental company in North America in 1996, but the company's founder was becoming increasingly troubled by anecdotal evidence that customer service standards were slipping. Attempts to find out what was going wrong started with a 17-question survey in 1994, which was mailed out to customers. Two years later, the company had narrowed the survey down to one question which it found had the most bearing on customers' actual behavior, and had also started conducting the survey by phone.

Critics of this new breed of customer survey contend that they are just too simple, that they don't tell companies why customers are unhappy or what to do about it. "The criticism that this is just one question is pretty common, but it's superficial," says Reichheld. "Knowing the score is just the first step. You also have to have a process for following up with your customer, identifying the problem and fixing it."

That's exactly what Enterprise has been doing. Dan Gass, Enterprise's vice president for customer service, explains that no employee is eligible for promotion unless the branch, city or region that they represent has ESQi numbers that are above the company-wide average. As such, branch managers take it seriously when their latest survey results show a drop in customer satisfaction. It's standard practice for branch managers to personally contact all customers who are not completely satisfied with the service they received and try to find out what went wrong. When Enterprise adopted these policies in 1996, the company's average score for completely satisfied customers was below 70%. In February of this year it hit 80% for the first time. Simply put, says Gass: "Our overall customer service serves as a bellwether for the health and well-being of our corporation. If our customers are completely satisfied, our business continues to grow."

It's not just companies on the scale of Enterprise that are using simplified customer metrics–smaller companies claim to benefit as well. Riggs Cat, a heavy-machinery dealership in Arkansas with annual revenues of around $230 million and around 20,000 customers, started using NPS six months ago, says Jon Bryant, the firm's marketing and customer relationship manager. He says that the company's CFO helped provide some of the impetus for the new approach as a way to keep tabs on the frequent billing disputes that delay payments of receivables. As surveys are conducted, any signs of a dispute are immediately sent to Riggs Cat. Now, says Bryant "We are resolving these issues much faster. The response of our CFO has been that we've gained so much more from the surveys than just finding out how willing customers are to promote us."

It's still in its early days at Riggs Cat, but Bryant says that discussions have already started about how to use the NPS as a component of compensation. At the moment, 5% of pay for all rank-and-file staff is determined by the company's market share and profits. Bryant says that this proportion will likely remain the same, but NPS will be added to the mix. In doing so, the company hopes to drive customer service improvements in some of its more far-flung stores. At the moment, NPS scores vary considerably from one location to another, and Bryant says the company sometimes wonders "whether the outlying stores, which contribute less in terms of overall revenues, are a little disconnected from corporate directives." The hope is that publicizing NPS scores and tying them to pay will help make customer service a higher priority across the company.

The credibility of NPS has also been given a shot in the arm by the willingness of some professional investors to use it when making investment decisions. Late in 2003, Summit Partners– a venture capital and private equity firm that has raised $9 billion since 1984–was running the rule over an opportunity to take a stake in Chicago-based online stock and options broker optionsXpress. Summit had no problem obtaining the info it needed about the broker's financials, products, market and people, but "the one unknowable was what retail customers thought of optionsXpress and its service," says Bruce Evans, a managing partner with Summit in Boston. "The growth rate suggested that customers liked the company but we wanted a definitive answer." Summit surveyed 1,600 optionsXpress customers via e-mail and received 400 responses revealing a customer base that was "emphatically happy about optionsXpress," Evans says. "For us, the NPS data added a lot of momentum to the deal." Soon after, Summit paid $90 million for a 30% stake in the company–a decision the firm's investors have never regretted. OptionsXpress went public in an issue led by Goldman Sachs in January 2005 and currently has a market capitalization of roughly $2 billion, says Evans. Since then, Evans says that Summit has used NPS on a number of other occasions: "It provides really good, really usable data."