As a business school professor, my compensation is based in part on how satisfied my students are. The corporate world follows a similar logic. CEOs get paid based on how shareholders value their company’s stock. Companies such as Cisco Systems, Siebel Systems and Sun Microsystems link bonus compensation to customer satisfaction.

While most companies design employee compensation based on a "pay for results" scheme, they rarely apply the same logic to getting paid by their customers. When you think about how companies get paid for their products and services, you realize that it often has very little to do with customer success. Ideally, the more value you create for your customers, the more you should be rewarded. Value capture for the seller should be directly linked with value creation for the buyer. But that’s not what happens in practice.

Consider a commercial printing company that does direct-mail catalogs. The more catalogs it prints for its customers, the more it gets paid. Yet its customers define success in terms of the dollar sales generated per catalog mailed. From their perspective, the more effectively they target customers, the fewer catalogs they need to mail. So the printing company actually profits from the ineffectiveness of its customers.

Similar perverse logic applies to brokerage companies such as Ameritrade and E-Trade. They make their money on trading commissions, so they have an incentive to make customers trade more frequently. However, investment advisers will tell you that most investors are better off with a "buy and hold" strategy rather than trading frenetically in an effort to time the market. The more money that brokerage companies make from trading commissions, the less likely that their customers get superior returns.

Recognizing this problem, companies such as Charles Schwab and Merrill Lynch have moved away from commission-based compensation. Instead, they base their fees on a percentage of the client’s assets. But even this doesn’t align the client’s interest with the interest of the brokerage firm. Ideally, my broker should get paid more if I get superior returns on my investment. How about charging customers fees indexed to investment performance? Brokerage companies could agree to be paid based on their performance relative to a benchmark such as the Dow Jones Index or the Russell 2000. Mutual funds could base their fees on their percentile performance within a style class (value, growth or large cap). So why hasn’t any investment firm implemented this approach? Maybe it has something to do with the fact that investment advisers and actively managed mutual funds rarely beat indexes.