Beware Of The Secret Shopper Syndrome
November 7, 2007 Leave a comment
Lots of firms — from banks to retailers — use “secret” or “mystery” shoppers to evaluate if individual stores/branches are delivering good (or the right) service. With this type of an approach, consumers are sent to different locations to complete a specific task (or tasks). Then they are asked to fill out a survey based on their experience. The shoppers usually receive some small payment for their efforts.
This is a great way to get a meaningful sample with structured data about customer experiences. But unfortunately, sometimes this type of an approach can go bad — as it did for my wife during a recent trip to the post office to mail a package.
I’m can’t take credit cards; our machine has been broken for 3 hours…. My manager wouldn’t let me hang a sign out front where people would see it. We can’t mess with the decor, because a mystery shopper may come by and we’d lose points.
There’s something definitely broken here. In this branch of the post office, they are optimizing around the experience of a mystery shopper, not the actual customer.
Unfortunately, this problem is not confined to the USPS. And it’s indicative of a bigger problem: misaligned metrics. You can see this problem at a car dealerships whenever a salesperson tells a new car owner “you’ll be getting a survey and make sure to give me a perfect score” or when a salesman makes sure that only “friendly clients” get to fill out satisfaction surveys.
Whenever your metrics aren’t quite aligned right, it puts people in a position where they can — and often will — game the system. Who loses? Customers. Because instead of focusing on the customer and his/her experience, employees focus on optimizing some arbitrary metric — like consistent decor.
Don’t get me wrong, I think that data from sources like mystery shoppers can be very valuable. It’s just a matter of how the overall program (including employee incentives) is structured.
The bottom line: Focus on the needs of real, not mysterious, customers.