When executives on a leadership team get serious about making their organization more customer-centric, they often begin by examining the company’s incentive systems. In most cases, they quickly determine that the key performance indicators and scorecards on which employees are rewarded favor financial performance over customer loyalty. It then seems obvious that one of their first big symbolic moves should be to incorporate some form of customer feedback directly into everyone’s incentives, from senior executives right through to the front line.

As the creators of one of the most popular customer feedback metrics, you might think we would applaud this move as a requirement, or at least as a key to success in becoming more customer-centric. But we don’t. At least not the way most companies do it.

Let us be clear. We see embedding customer feedback metrics into incentive compensation as a natural and well-intentioned decision. Businesses are accustomed to paying for results, after all. Companies want to emphasize that they take feedback seriously, so they put their money where their mouths are. “If we are really serious about this,” they will say, “then we have to include it in the incentive formula. Otherwise, no one will really focus on it.” We see this all the time when we help companies become more customer-centric using the Net Promoter System^SM. They want to link the Net Promoter Score^SM to employee pay.

But an early link between Net Promoter Scores and compensation has almost always proved to be a mistake. Despite what executives often tell us—“You don’t understand. Our culture is different. We are a metrics-focused organization”—establishing the link too quickly has caused problems in virtually every situation we have studied.

Some of the problems are practical. In many cases, the incentives lead to behaviors that are the exact opposite of what was intended. For example, we saw employees of one retailer spend lots of time arguing about the appropriateness of the score, the technical details of how it was collected, the particulars of the calculation and the analysis of what was driving the score. But the score should be just an indicator of success and progress, not an end in itself. The system is about more than the score; it’s about using feedback to learn and to improve the customer experience. You want people to discuss the feedback they receive, dig out the root causes of customer concerns, take action to fix problems and find new ways to earn enthusiastic praise. When the score determines people’s pay, they’re more likely to focus like a laser on the metric itself. They’ll talk not only about how it was calculated but why it isn’t right and why they shouldn’t be held accountable for it— when all the time they should be focusing on learning how to earn their customers’ adulation.

Worse still is the temptation to game the scores. Instead of taking Net Promoter® objectives seriously, employees may figure out shortcuts to higher scores, the way car dealerships habitually beg for top-box ratings. (See below "A cautionary tale: The feedback system that became a national joke.”) Some may cross the line into unethical behavior. In one case, Enterprise, whose approach was the basis for the original Net Promoter System, heard rumors that team members at a few branches were changing phone numbers on the records of unhappy customers. As a result, callers seeking feedback never reached those individuals. (At Enterprise this kind of behavior is called “speeding” and is grounds for dismissal.)

To make it even more challenging, newly adopted metrics such as the Net Promoter Score may bounce around for a while, until practitioners figure out solid feedback collection techniques, appropriate sampling methodologies and so forth. Employees often feel that it isn’t fair to tie their compensation to so mercurial (and apparently uncontrollable) a measure. Moreover, they may not yet have learned the connection between scores and behaviors. So variations in scores in the early days seem arbitrary— the last thing employees would want their bonuses tied to.

Call center reps at St. George Bank in Australia had exactly this reaction. They believed many bad customer experiences could be traced to product, policy or operational issues outside their control. To address their concerns, St. George’s parent company, Westpac Group, created a metric that encouraged the reps—called “bankers” within Westpac—to maximize the number of Net Promoter surveys collected. Only after a year, when the volume of surveys collected allowed bankers to see that they could indeed influence both detractors and promoters, did Westpac Group incorporate the Net Promoter Score into the bankers’ incentive compensation scheme.

Even if you could somehow avoid all of the behaviors above—overemphasis on the score, gaming and a natural suspicion of any incentive-linked metric—you still would be sending the wrong message to employees by immediately linking compensation to changes in the Net Promoter Score.

To see why, think about what the Net Promoter System is trying to accomplish. It builds the voice of the customer into a company’s daily operations. It establishes the fundamental goal of creating loyalty and advocacy among customers—turning as many as possible into raving fans. It sends a message that the very purpose of this company is to enrich people’s lives by providing them with great products and services and an exceptional experience. The companies that gain the most from the system are the ones whose employees truly care about enriching customers’ lives in this way.

Can you buy this kind of caring with incentive compensation? You might as well try to buy friendship. Your best employees—the ones you want to keep, who create the most value and whom you most want to encourage— will put in extra discretionary effort because the system is giving them a chance to exercise their judgment, to master a task and to be part of a productive team. They are people who value the opportunity to enrich someone’s life. Their motivation is intrinsic, not extrinsic. They don’t do it for the money.

None of this is to say that improvements in Net Promoter Scores should never be tied to compensation. In the right context, incentive pay can reinforce and reward people for the right sorts of behaviors, and thus create a stronger system. But we urge companies to abide by a set of tough guidelines that have been shown to increase the odds of success and decrease the odds of unintended negative consequences.

First, wait until you have an established, stable system with reliable metrics. That requires identification of the right respondents, consistent and well-tested survey techniques, and sufficiently large sample sizes and response rates. Veteran Net Promoter practitioners such as Charles Schwab have found that even tiny variations in approach— a change in font size in the email, for instance—may affect results. Only when everyone in the organization can see that the feedback is accurately collected and measured will they believe in the reliability of the scores.

A second guideline: You need crystal-clear links between scores on the one hand and financial and strategic outcomes on the other. Changes in incentive compensation make sense to people only when those changes help the business succeed by encouraging the right kind of discretionary effort. Schwab was able to show that differences in Net Promoter Scores correlated highly with the growth and profitability of its individual branches, so it made good business sense to tie bonuses to NPS results.

Third, employees need processes and tools for understanding root causes. This requirement gets to the heart of the Net Promoter System: It’s the job of frontline employees and supervisors to determine why some customers wound up dissatisfied while others were delighted. Once they uncover the root causes, they can take action to improve the situation or else recommend changes to the higher-ups. But if they don’t understand the root causes, they have little ability to affect the feedback they receive, and they are essentially powerless to affect their incentive compensation.

A fourth essential is programs that facilitate organized learning. The more that frontline employees can learn from one another and from knowledgeable supervisors, the faster they will progress. Faster progress makes for a better customer experience, which then generates better ratings and higher bonuses. Higher bonuses encourage still more learning. It’s a powerful virtuous cycle, but it can’t get started unless opportunities for organized learning are there in the first place.

The fifth requirement is repeated communication. Companies often announce bonus or incentive plans with some fanfare and then forget to reinforce them. That doesn’t work in this context (or probably in any other). People need to be reminded of the incentive plan’s goals and motivations, how it is expected to function and what they can do to increase their own compensation. Just when you, as a leader, believe you have dramatically over-communicated about the intent of the incentives, you might just be reaching the lowest threshold for success. Keep going.

The sixth and final requirement? Strong anti-gaming rules and procedures. It should be a firing offense to talk to customers about the scores they will ultimately give you. One telecommunications company has begun to monitor its scores, and when analysts spot a pattern that suggests evidence of tampering they immediately audit the feedback. As Ronald Reagan famously said—quoting an old Russian proverb—“Trust, but verify.”

Verification is important to a successful Net Promoter System, but trust is absolutely central. Indeed, trust underlies the entire system. Companies trust their employees to do right by their customers, and to learn over time how to deliver an even better experience. Employees trust that their company really does have the best interests of customers at heart, and that they will ultimately benefit from being part of an ethical business. Customers trust the company to deliver great products and services, and they trust employees to listen to their feedback and act on it. Linking customer feedback and money too quickly often undermines all this trust. It encourages people to do the right thing for the wrong reasons, and it often backfires.

When the system is established and trusted, however, sharing in the benefits reinforces trust. That may seem like a tough balancing act, but it’s one that companies need to get right.

Fred Reichheld and Rob Markey are authors of the best-seller The Ultimate Question 2.0: How Net Promoter Companies Thrive in a Customer-Driven World. Markey is a partner and director in Bain & Company’s New York office and leads the firm’s Global Customer Strategy and Marketing practice. Reichheld is a Fellow at Bain & Company. He is the best-selling author of three other books on loyalty, published by Harvard Business Review Press, including The Loyalty Effect, Loyalty Rules! and The Ultimate Question, as well as numerous articles published in Harvard Business Review.

Net Promoter® and NPS® are registered trademarks of Bain & Company, Inc., Fred Reichheld and Satmetrix Systems, Inc.

Net Promoter System^SM and Net Promoter Score^SM are trademarks of Bain & Company, Inc., Fred Reichheld and Satmetrix Systems, Inc.