I went to a convenience care clinic yesterday. My regular doctor was on vacation and I thought that I might have strep throat.

As I sat in the examining room something struck me. It was the word “EXCELLENT”. It was everywhere. One sign read:

EXCELLENT SERVICE
Our goal is to provide you with EXCELLENT service.
Please let us know what we can do to make your visit EXCELLENT.

A second sign, from the charge nurse, also reinforced that they were striving for EXCELLENT service and that if I didn’t receive EXCELLENT service I should contact him or her.

Normally this would have just struck me as odd. However, having worked in healthcare, I had a suspicion as to what was happening.

It turns out this is a trick that is used to increase customer satisfaction scores. Patients, customers, or anyone else, when given the opportunity to rate you on a survey, will mirror the language that they heard during the interaction.

My guess is that in a few weeks I’ll receive a customer service survey. The highest rating on the survey is probably “EXCELLENT”. In the hospital that I worked, the key words were “VERY GOOD”. Our nurse managers were taught to say “Very Good” repeatedly when talking to our patients. Even our elevator wished you a “Very Good” day when you exited in the lobby. The consultants who worked with us at the hospital assured us that our scores would go up if we would consistently use that phrase. To a certain degree, they were right.

Here’s the problem. Our service wasn’t changing (in a meaningful way). If two people have identical experiences but one rates the experience higher because of this subliminal scripting, was that experience really better? Some may argue that it was since satisfaction is about perception. Whether it is or not, I think this strategy misses the mark. While changing perceptions is important, the goal of measuring satisfaction is to ensure that your people, processes, and systems create a good experience for your customer. This is a case where people lost sight of the metric as a proxy for a goal. They began to make decisions and take actions for the sake of the metric rather than the goal. Instead of teaching people how to create a better patient experience, we were teaching them how to get the patient to mark a different number.

The real question we should have asked was whether we believed that our current service was adequate. If we thought our service was good, why were we spending any time trying to get our patients to give us a higher score? We should have just set the target accordingly. Conversely, ff we felt that our service was inadequate, why focus on changing perception rather than the actual service? Keeping the service the same while focusing on improving perception provided little practical value especially to the patient. (Note: there was some value since the patient satisfaction scores are converted to a relative ranking against other hospitals. However, in my opinion, this just creates a situation where there is significant grade inflation as organizations game the system to increase their scores rather than taking real action. Perhaps this explains why healthcare, as an industry, is rated so low against other industries when it comes to customer experience).

This isn’t just true in healthcare. You’ve probably experienced this the last time you went to a car dealer for service. You come to pick up your car and receive a copy of the satisfaction survey that will be mailed to you by the manufacturer. It will inevitably say something like, “If you can’t give us a 5 on all questions, talk to the service manager”.

If it took three days for a repair that should have taken two hours, do you somehow get that time back? If the service agent was rude does the manager someone erase that part of your memory? Of course not. They can’t change the experience which is what the survey is measuring. The best they can do is compensate you for the poor service. That’s ok. Companies should compensate their customers for poor service. If you give them all fives you enable them to hit their target without reaching their goal. Chances are, the next time you come in for service not much will change.

When they ask me what they can do to get all fives, I tell them to fix their service and quality problems, and if they don’t recur, I’ll give them all fives on my next visit. That’s the right way to help them use their survey to improve service. If there is a question on responsiveness to my concerns, I’ll certainly give them a five on the spot. I’m not going to say that the other aspects of the service were good when they weren’t.

Gaming the system happens everywhere. I worked in one organization where managers would wait to fire employees until after their probationary period. This was because there was a metric in place regarding attribution in the first six months. The goal behind that metric was to decrease costs and improve the quality of the recruiting process. It’s expensive to hire and orient employees and losing employees in the first six month is a problem. However, by keeping the employees beyond their probationary period, these mangers were making things worse. Not only did the company continue to bear the cost of poor performers, it also had added costs since it’s harder to fire someone after their probationary period. In addition, by keeping the employees, it masked the original recruiting problem for which the metric was designed to track.

Measures are proxies. Proxies aren’t bad as long as you remember that they aren’t the goal. They are a tool to help you determine whether you are achieving a goal. Playing to the metric without understanding or attending to the real goal will only provide short-term benefit.