By Tom Paquin

Is the customer always right? Obviously not. If a customer comes into a store and says that they are deserving of a 60% discount on their order because they had a hard time finding a parking spot, it’s okay to say ‘no’ to them.

I’d argue that any utterance, complaint, or assertion of any kind from a customer be met with a proportional amount of scrutiny. Because people don’t necessarily know what they want. Sometimes you need to show them what they want.

I will use a video game example, so skip the next two paragraphs if you don’t want to reach about Nintendo. Every few years, wide-eyed analysts and fans make broad assertions that Nintendo is about to release some sort of “dream” console, or game collection, or localization of a beloved Japanese game. These things never happen, and instead Nintendo releases modest hardware upgrades or $60 re-releases of mediocre games.

Nintendo does this rather than capitulate to fan demands because it is cheaper, it specifically targets the young demographic that Nintendo has always targeted, and because it has proven to be incredibly successful. Podcaster and journalist Justin McElroy frequently says, of Nintendo, “They won’t sell what you want them to, but you’ll want what they sell you.”

This is a similar strategy to the one employed by Apple, who, rather than acquiesce to the demand that they build a $500 “Netbook” (remember netbooks?) in the late aughts, built the exact opposite—a nearly $2,000 computer called the MacBook Air, that had something called a “Solid State Hard Drive”. And what type of computer has endured for over a decade?

Doing that requires you to understand your customers better than they do, and to study your customers in a non-binary fashion. The Simpsons has shown what happens when you just ask people what they want:

There is no value in getting a list of customer wants, you need to understand their behavior. In service, there are a variety of ways that you can do this.

One key consideration is to consider not just asset and employee output when measuring service processes, but also customer output. If assets are IoT-enabled, what is the length of time between exception and customer contact? When given an array of service times, how far into the future does the customer, on average, book an appointment for a specific need?

Then, of course, you can begin to benchmark elements against one another. What is the correlation between NPS and time from ticket to invoice? What length of product ownership yields the most repairs, errors, and scraps?

Finally, you can begin to involve customers directly in the service process, and see what attrition rates are with utilities like appointment assistance, “where’s my tech”, and other such tools. Doing so allows you to see what customers really value, anticipate their needs, and build service solutions that meet the contour of your business without needlessly cutting into your bottom-line. Because you know your business best, and your customers should want what you are serving them.