I’ve just finished reading the book Agent Zigzag by Ben Macintyre. It is the true story of Eddie Chapman, one of the most fascinating and effective double agents working for Britain during WWII. Chapman, a career criminal, was the only wartime agent pardoned for his crimes by the British. The Nazis so valued his efforts during the war that they awarded him the Iron Cross.
Eddie Chapman was clearly talented when it came to telling stories and charming friends, romantic interests, and his spymasters. Yet there were times when his deception should have been flagged and investigated more carefully by the Germans. One reason this wasn’t done lies in what the book calls “the central defining flaw of the German secret service.”
The Germans tied the fortunes of the spymaster to his spy. Dependent on each other and competing with other German spies and spymasters, there was great pressure to provide information about the Allies. Chapman’s Nazi controller needed to believe the misinformation Chapman was feeding him. Having a productive spy increased his monitory rewards and kept him away from the Eastern Front.
In contrast, within the British secret services, spy case officers shared responsibility for multiple spies. The British knew that if a spymaster’s self-interest was tied to his agent, he would not see that agent clearly and that might compromise the spymaster’s personal integrity.
This problem with tying individual rewards and punishments closely to the performance or satisfaction of a third-party is something we still see today in businesses, government agencies, the news media, and public schools. One of the most annoying manifestations of this type of warped reward dynamic are in the ubiquitous customer service satisfaction surveys. Too many companies have clumsily tied some aspect of compensation of front-line employees to the results of customer satisfaction surveys. This results in front-line sales and customer service employees coaching customers to always “give a 5-out-of-5.” From car dealerships to carpet cleaners they repeat the mantra that if the customer isn’t completely satisfied and isn’t able to give a 5-out-of-5 (or a 10-out-of-10), the customer should immediately tell them and they will fix the problem *before* the customer fills out the survey. This creates an environment where senior management is kept in the dark. Hidden are possible systematic issues that would be obvious if customers filled out surveys without influence. Any problems are either fixed with a series of one-of solutions by the front-line workers or are swept under the rug because the front-line workers have pressured customers to only give perfect scores in the surveys. A perfect score on a customer satisfaction survey is basically worthless to the company who is trying to improve.
Throughout business, when managers run their departments with little outside input or objective analysis, misinformation is bound to creep into reports. In schools, the pay-for-performance movement and the penalties imposed by No Child Left Behind created an environment that encouraged standardized-test cheating scandals.
Of course, as the Nazi’s found out, glowing reports that fail to mention the negative, will eventually catch up with you. If you are a business, perfect customer surveys will not stop your customers from finding other venders when they are truly unhappy. In the schools, kids who are below grade level will eventually feel the negative effects of that, regardless of their impressively doctored test scores. In order to get quality information, we must be much more careful about introducing self-interest variables into our search for the truth.