In early 1902[i], Paul Doumer, the French governor of Hanoi, began receiving complaints about the city’s rat population from other colonists. The rats had been there before the French (who had already been there for decades themselves) but the increased understanding of the mechanics of disease transfer had turned an unpleasant fact-of-life into an issue that needed to be tackled immediately.
Doumer, believing that the rat population could be tamed, set up a system where rat catchers would be compensated for each rat they caught and killed. The initial results were promising: for the first week of the program, rat catchers killed an average of over 1,100 rats per day.
Had the program been a success, we would expect the quantity of rodents captured to decrease, as the remaining rats became harder to find. Had the program been merely ineffectual, the number of rats caught would stay flat or grow with the number of rat-catchers as the limiting factor.
Instead, the number of rats caught skyrocketed. Daily counts topped 4,000 by May and 10,000 by June. On June 12, 1902, 20,114 rats were handed over to the French, nearly 20 times the initial catch.
Confused and concerned by the numbers, the French began to investigate. What they found horrified them. The Vietnamese has begun to breed rats for the bounty. Instead of reducing the number of rats, the French had underwritten a massive agricultural subsidy.
Brands look to reward and engage their most devoted fans and followers. The cultivation of supporters and promoters is crucial to brand success. But some brands have been frightened off from rewarding these key consumers because of the fear of ‘gamers’. This is an over-reaction: in our experience, incentives certainly influence activity, but well under 0.5% of users attempt to game our rewards programs to a meaningful extent.
People have been dealing with those trying to ‘game the system’ for as long as there have been systems. Fortunately, we have several big advantages over early 20th century colonists. At CrowdTwist, we are obsessed with the possibility of ‘perverse incentives’, the class of problems represented in the story above. Our analysis starts before the launch of a loyalty program and continues through reward redemption.
- Program Design: The most important way to counteract gaming is to design programs with gaming in mind. Primarily, this is a question of valuation. How much do we value the actions we are trying to incentivize? What is the value of the reward? How many rewards do we want to give away?
- Activity Monitoring: All of our rewards programs include automated data monitoring where potential gaming is tracked. The triggers are based on statistical analyses of previous programs, as well as ‘common sense’ conclusions. While we can’t disclose the exact measures, we have analyzed dozens of identifiers for gaming and use a diverse set of tests.
- Continual Iteration: Those looking to game the system continuously come up with new ways to obtain an advantage without contributing any value. From the US Mint to predictable product prices, externalities can impact how a program works in many ways. Continual adjustment is necessary to keep a loyalty program working as intended.
At CrowdTwist, our experience and familiarity with gaming infuses every step, from program design to reward fulfillment. We’re experienced at encouraging rat-catching and discouraging rat-farming, if you will. If you’re interested in learning more, shoot us an email and we’d be happy to continue the discussion.
[i] Michael G. Vann, “Of Rats, Rice, and Race: The Great Hanoi Rat Massacre, an Episode in French Colonial History,” French Colonial History Society, May, 2003