Skip to Content

What the VW scandal says about cheating at work

CHINA-GERMANY-AUTO-FAW-VOLKSWAGENCHINA-GERMANY-AUTO-FAW-VOLKSWAGEN

In the wake of the Volkswagen scandal, I can’t help but wonder at what point did Volkswagen decide to create illegal software that defrauds vehicle emissions tests. What was the tipping point in incentives and motivation that drove the decision? And applying this question in a broader way, how often are people incentivized to cheat at work, and what can be done about it?

Dan Ariely, a professor of behavioral economics at Duke University, writes in The Wall Street Journal that everybody has the capacity to be dishonest, but most people are not rotten to the core. But, more often than not, in a tough conflict between self-interest and integrity, even the best of us have the capacity to turn a little rotten; it’s what Professor Ariely refers to as the “Fudge Factor.” As Chief Design Officer of Badgeville, a gamification company, I have witnessed how employees of many companies fudge their performance to earn points and prizes. In my experience, in order to address cheating, you have to tackle two distinct issues. First, how do we recognize and investigate cheating? Second, if people are cheating, what strategies can curb it?

In general, anything is vulnerable to cheating. When you invent any game with rules, whether it be a point system for customer service performance, an emissions test for automobiles, or even a game like football or basketball, you unavoidably create ways to exploit the game or cheat.

The first step is to ask yourself this: does it matter? In Volkswagen’s case, I think we’d all argue yes, it does; the environmental impacts are huge and the company broke its trust with consumers. The impact of cheating can be more ambiguous.

A typical gamification program records and rewards particular desirable workplace behaviors. The opportunity then exists for someone to fake those desired behaviors. For example, if your intent is to promote more workplace conversations in an online collaboration tool, a side-effect may be more non-work conversations as well. But do you necessarily have a problem? On the one hand, you’re rewarding people for discussing off-work subjects at work, but on the other hand, the system as a whole is increasing all conversations, including the desired work-related conversations. Some users may even be getting rewarded for working less, but the net return for the company is more beneficial conversations overall.

The determining factor of whether you should act or not is, are the effects of cheating resulting in poorer performance, lost revenue, or raising accusations of unfairness among the ranks? For example, a salesman entering fake sales call transactions into the company’s CRM, or a services representative creating fake conversation threads in their company’s service community and telling friends about it to get artificial up votes, are examples of the kind of harmful cheating that cost the company money and credibility amongst its other employees.

Once you’ve established your cheating problem is indeed a problem, the second step is to determine why people cheat. What’s the reward? Is it points, privileges, money, prizes? Tangible results are a common cause but don’t underestimate the appeal of bragging rights and prestige. Sometimes, the reward and encouraged behavior aren’t well-balanced. Cheaters will quickly identify these mismatches and take advantage. In cases like this, simply removing the amount of reward may reduce the cheating.

Third, if the problem persists, you’ll need to investigate how people cheat. Do they enter fake data? Do teams of people ‘help each other out’? Are employees neglecting processes or cutting corners to ‘farm’ the tracked behaviors?

In most cases, extreme cheating should be obvious and stand out if you review the numbers. However, more subtle cheating, such as a salesperson who pads their numbers with a few extra calls – maybe to hit a bonus in your gamification system – can be harder to spot and your best source of policing will come from the complaints of coworkers.

Once you’ve confirmed the presence and scope of the cheating, don’t overreact. If you punish the whole company, you might kill your program altogether (and in most cases, if some employees feel compelled to cheat, it means you’re doing something right and likely many more employees are participating appropriately).

To counter the cheating, I recommend using a combination of these six approaches:

1. Limit behavior rates. If people are fabricating an action to obtain a reward, you can reduce the amount of times you will reward a user for that type of action. For instance, on a social media site, if one comment is worth points, you may introduce a ‘cool down’ period for scoring additional points for that behavior. For example, you can only get credit for writing a new comment once every five minutes.

2. Limit the count. Place a cap on how many times employees can earn credit for a behavior. For example, if updating a social media profile generates a reward, configure the game such that employees only get credit the first three times they update their profiles.

3. Crowdsource verification. For certain activities, an outside party can verify the quality of a behavior. For example, if you reward customer service reps based on the volume of questions they answer, they can cheat the system (e.g. rapidly copy-paste a generic answer to every question). However, if you only reward employees when a customer ranks their answers as “Helpful,” then you have essentially put a crowd sourced quality filter on the submitted answers.

4. Moderate verification. While this is more expensive than crowd sourcing, you could give an individual or team the responsibility of confirming behaviors, spot-checking oddities and responding to complaints of unfairness. Even with a handful of irregular spot checks, habitual cheaters will be discovered. Take Volkswagen as an example.

5. Quarantine cheaters. If the cheaters all belong to one team or geographical region, you may choose to manage and monitor them separately, particularly if the cheating is more of a cultural perception or difference in departmental values. This could save you from changing a gamification system that performs great throughout other parts of the population.

6. Remind people of honesty. In his WSJ article, Professor Ariely writes: “…simply being reminded of moral codes has a significant effect on how we view our own behavior.” A physical poster or a regular email about the rules of the game might cue people to be honest.

Within all of us, there is a seed of temptation to cheat. As business leaders, we have the power to reflect on our workplace, system and culture to create systems and environments that bring out the best in our people rather than the worst. Take away the ability to easily cheat and add a social pressure to not cheat, and you’ll significantly curb the incidence of cheating in your gamification programs and your organization as a whole.

Steve Sims is co-founder and Chief Design Officer at Badgeville.