Sign up for our free weekly e-newsletter

"Eighty percent of the time you can't predict what people will do by how satisfied they are," Hofmeyr said. One problem is what he called the mutually compensating error: For every customer who says he'll do something and doesn't do it, there are an equal number of customers who say they won't do something, but then do it. For example, perhaps you query customers as to repurchase intent. If 30 percent say they plan to repurchase in the next year, and you measure repurchase a year later, you'll likely find that 30 percent did repurchase. But it's not the same 30 percent. Some of the customers who said they plan to repurchase, let's say 20 percent, in fact, do repurchase. The other 10 percent don't. But 10 percent of those queried who said they weren't planning to repurchase do make another purchase.

"Classical measures show an average level of potential business outcomes," Hofmeyr explained. "They don't show which individuals are truly loyal."

For this reason Hofmeyr recommends asking customers two essential pieces of information: how important you are to them, and how they rate you against the competition.

According to Hofmeyr, commitment is stronger than loyalty, and customers will stay committed when an organization is important to them. If your organization isn't important to customers, you can't create commitment, he said. Additionally, as crucial as it is to know how a customer rates your company in terms of satisfaction, likelihood to recommend or repurchase, etc., it's equally as essential--and often overlooked--to know how you rate against the competition. For instance, if a customer gives you an 8 on a 1-10 scale of likelihood to recommend, that's great. But if that customer gives your main competitor a 9 or 10, that's not so good for you, after all. "What matters is not your absolute score, but the score you get relative to other options," Hofmeyr said.

One caveat in the quest for loyalty: habituation. People get used to something (cool product features, free overnight shipping, what have you) and ask what's next. "'Customer delight' is the most dangerous marketing premise out there," Hofmeyr said. "Nothing has the power to delight a human being forever. We're doomed to be habituated to things that delight us."

The benefits of customer delight are valuable, he said, but companies shouldn't "processize" delight or customers will become accustomed to it and expect more and more. "Delight works better when it's rare and unexpected," he said.

Finally, Hofmeyr pointed out that business leaders shouldn't expect customers to "love" only their company. "Anyone who has children knows that love is boundless," he said. "So why do we think customers owe all their loyalties to us?"

Customers may not "owe" any one company all of their loyalty, but that doesn't mean you should end the quest for it--and for commitment. Loyalty and commitment are each a powerful competitive asset in their own way. So, Hofmeyr recommended four actions to help ensure success in that quest:

• Establish and measure relevant competition
• Take habituation into account
• Measure involvement
• Understand that customers have a right to value others, too