Patient satisfaction metrics are bad for healthcare. There. I said it.

I know. It is blasphemy to say something like that in the Lean world. The focus of continuous improvement should be on the customer. The customer tells us if we are doing something right by speaking with their wallets. The wider the wallet opens, the more satisfied the customer is.

Sure, there are surveys and awards and reports on patient satisfaction, but those metrics don’t matter much if a company is not selling the products people are rating highly.

Most companies would rather have a low rated product that sells really well than a highly rated product that sits on the shelf. That situation is uncommon, though. In most cases, the sales and product rankings are closely linked. Higher ratings tend to be correlated to higher sales.

Healthcare, though, would rapidly implode if everyone focused on patient satisfaction metrics.

Before I dive into that, though, I want to give you some perspective on what the healthcare industry is facing.

Imagine if your children got to choose your customer satisfaction metrics.

You’d get dinged for telling them to brush their teeth. For making them go to school. For not buying them everything on the toy aisle. For only having pizza 10 times in a week (remember, it can be for lunch and for dinner, and even for breakfast).

You get the point.

Now, think about what doctors face.

They encounter two major barriers to getting realistic patient satisfaction metrics.

The first is that patients don’t have reasonable expectations and confuse bad news with poor service. If a doctor wants a patient to be healthier—maybe a patient she has been seeing for years—and is extremely concerned with an increasing weight and smoking, she may become more blunt with each visit. Using the word “obese” to describe a patient who thinks he or she is just a few pounds overweight isn’t going to score many points. And telling someone that they have to quit a lot of the things they love is definitely not the way to score highly on satisfaction.

Or consider the anti-vaccination push. Most doctors strongly favor vaccinations. They have to look out for what they feel is the best interest of their younger patients, even if the parents disagree. Presenting facts against an emotional issue, again, is not the way to score patient satisfaction points.

And sometimes patients think they know a lot more than their doctors. Sometimes they do. Obviously, individuals have much more familiarity with their symptoms than a doctor does. But often, they make and incorrect self-diagnosis, or settle on an ineffective or inappropriate remedy. Patients want antibiotics, even if the doctor says it isn’t necessary. They want the drug they saw advertised on TV, even if it isn’t the right drug. Speaking of drugs, some want extra helpings of certain types of meds and might be irritated at a doctor cutting back on a supply of pain medication.

The situation for doctors is further compounded by the fact that they answer to insurance companies as well as patients. If they want to get paid for the work they do, they have to abide by the limits on the insurance plan. At least, they have to make sure the patient knows what the cost will be. And often, that is not easy to do in times of crisis.

Patients might want extra tests when they are in the ER. But then, come billing time, they challenge the need for those tests and the costs of those tests when the insurance company says they aren’t covered.

Plus, doctors might feel pressure about when to release a patient from the hospital. A patient might want to stay longer at the insurance company’s expense, while the insurance company is pushing doctors to justify the length of the stay that exceeds normal situations.

The first issue—of patients not wanting to be told things they don’t want to hear—is a challenge because of human nature. Long term improvement almost universally entails short term effort and sacrifice. And since patient satisfaction metrics are typically based on short term activities, primarily individual visits, patient satisfaction is a poor metric.

Now, there can be an argument made that doctors need to adjust to accommodate short-term patient needs while still focusing on the long-term impact. Perhaps they don’t directly use the word ‘obese’ but instead include a chart with the patient’s weight shown on a scale of healthy BMIs. But there sometimes can be no getting around the pressure. A patient who really wants antibiotics for a viral condition will be angry if they don’t get them. And if they do get them, what happens? In many cases, the timing of the human immune system fighting off a bad cold will match the duration of a course of an antibiotic pack. The patient will feel, incorrectly, vindicated, and will still think less of the doctor.

Some similar situations hold true in Lean organizations. Downstream customers often think they know best how the upstream supplier should be doing their work. They may want a part delivered in a way that is impractical or want a design change that makes the product easier to assemble but more susceptible to failure.

And they face the same situation with regards to payment. Since station 12 doesn’t pay station 11 for the work they do, they can ask for anything they want. Assembly doesn’t get a bill from paint or from fabrication, so their complaint/request list can be miles long. In most of the world, the response to a request for a special order is, “Sure, we can do that. It will cost $XXX.” Customers then decide if they really want the special order, or if was really just a ‘nice to have’ item.

To be honest, I really don’t profess to have a good solution to this patient satisfaction metrics issue for doctors. Sorry.

But as their issues relate to businesses, there are some remedies. Internal satisfaction scores can be helpful if they are in the form of specific metrics. And prioritization is a reasonable surrogate for wallets. Giving just a few metrics provides focus.

And you can make a customer rank-order requests, or ask for team members to support projects, or trade off tasks to rebalance work. All of these put a cost on the requests that make customers consider them more carefully.

The short of it is that customer satisfaction is great when there is a clearly defined relationship between a single provider and a single purchaser who is also the user and is also paying for the product.

When you add layers—distributors, for example—you cloud who a manufacturer serves. And when the person paying is different than the person selecting and the person using the product or service, again customer satisfaction gets murky.

The key is to acknowledge all the players and their needs, look for conflicts, and then problem solve to make the system get the best overall results. And fortunately for you, unless you actually are in healthcare, your situation won’t be nearly as complicated.