Several years ago I bought a large piece of merchandise at a national chain. I pulled around to their loading area and went in. I found myself in a room with a large flat-screen monitor, a bar-code reader, and several chairs occupied by several people. I scanned my receipt at the bar-code reader and saw my name appear on the flat-screen with the merchandise description and code.

As I waited, I noticed that a customer's status would move from "waiting" to "delivered." But it took a further ten minutes for the merchandise to be actually delivered to the customer. Here's what I think was happening. As the employee finished filling the last order he would turn to the next and click "Delivered." Then he would go hunt down the merchandise and haul it out to the loading area and deliver it to the customer.

I suspect the employees were gaming the system. I wonder if someone back at the corporate office was looking at reports that said that this particular store delivered merchandise in an averag of 9.2 minutes when it was actually closer to 20 minutes. Perhaps there was some sort of goal, objective, or metric that held employees accountable for delivering merchandise within a specific amount of time. If so, these employees (or perhaps their managers) had figured a way to look like they were a lot more efficient than they really were.

I hope this chain had more than one way of gathering data. Secret shoppers, customer surveys, and regional managers practicing "management by walking around" come to mind.

If you are measuring your employees on their behavior in an attempt to improve customer service and loyalty, don't depend on just one method of gathering data.

As Ronald Reagan said, "Trust, but Verify."