A near-real-time dashboard illuminated the room with the call center team’s close rate. The incentive to meet and maintain the company’s goal was an impressive number upon their paychecks at the end of the month. They loved to see the gauge graphic shimmer when they ended their calls and celebrated en masse when they concluded the day on target.

Then the day came when the dashboard fell dim. The data visualization feverishly sought the cause and applied fix after fix to resolve the outage. The call center managers consoled the call center teams that any shortage would not penalize them while consistently applying pressure to the technicians to expedite the resolution.

It wasn’t until after sunset that the dashboard flickered upon the screen, resuming its display of the team’s performance. The team members had begun their journies home hours prior, but the call center managers remained on-site to witness the return of the dashboard. With their half-eaten protein bars melting in their hands, they could not believe their eyes. The teams, operating metric-blind, had achieved an all-time high close rate. The managers wondered if the organization could benefit from a more frequent dashboard outage.

This scenario is a true story that occurred several years ago and was an excellent illustration of Goodheart’s Law. In 1975, the economist Charles Goodheart stated that “Any observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes.” In other words, the manipulation of a metric is the direct result of using it as a means to measure and reward performance. The statement is not necessarily a judgment on the character of people, but rather a natural response to a miscommunication of intent.

The power of production targets is in their explicit description of a specific outcome. For example, my target is to produce a quantifiable volume of 300 lbs of nails. Upon achieving that target, I am promised to receive a 10% bonus. As a nail-maker, I quickly determine that I can meet that objective by lunchtime if I produce 30 – 10lb nails made of iron instead of five thousand galvanized steel nails. A busy morning, lunch, and cha-ching baby!

The weakness of production targets is in their explicit description of a specific outcome. For example, the client of the manufacturer of fasteners, the nail-maker’s employer, supplies their product for home builders. Their customer needs are for high volumes of light-weight and weather-resistant nails. The few 10lb nails, made of rapidly rusting iron, falls short of the intent of their production.

Leveraging production targets toward the effort of continual improvement rather than reward/punishment alters the nature of the metric. It is similar to the difference between day trading and long-term investment. The focus is on trends and analyzing the system that supports the process rather than blaming employees for a dip in production during a given hour.

The outcome of the metric-blind call center raises a couple of questions. Did the presence of the dashboard encourage sand-bagging as not to elevate expectations? Or did the focus shift from meeting a target to executing their process well to ensure a sale? Both are likely to be accurate, which speaks to the importance of thoughtful planning when developing production incentives. Are you experiencing the unintended consequences of overlooking the impact of Goodheart’s Law?