Risk of manipulation of performance data
A company's or unit's selection of metrics can sometimes lead managers to "game" the system so as to meet the targeted performance.
For example, consider a car company that aims to be "the best-rated customer service car company in the industry." This company will have metrics related to customer satisfaction, customer loyalty, and the number of customer complaints.
New-car salespeople who work in this company know that their compensation depends on what customers write in the feedback forms that corporate headquarters sends them after a sale. Immediately upon closing a deal, salespeople may "coach" the customer on how to fill out the form. They may even appeal to the customer's compassion: "My income depends on what you say on the form. I hope you'll help me." Result? Annoyed customers who feel manipulated by desperate salespeople—and a lack of valid feedback to the company about its salespeople's performance.
Here are additional examples:
Your direct reports (or even your peers in other departments) may purposely low-ball targets so they can easily exceed them and get bonus compensation and recognition. Or a sales team leader may exaggerate the tough selling environment so you'll set low sales targets and he can look like a star performer without having to work too hard.
The lesson here is that measurement systems can change employees' behaviors—sometimes in ways a company never intended! Avoid this scenario in your own group by making sure to do the following before you set targets:
- Take time to find out how your people might be altering their behaviors in order to meet targets you've set for the performance metrics you're using.
- Consider whether these behavior changes might be causing unintended—and unwanted—consequences for your group and your company.
- If you ask some of your direct reports to provide input on target setting, try to verify the rationale behind their suggestions. Don't just take them at their word; ask them what sources they're using to form their opinions about targets, and verify the reliability of those sources.