The Misuse of Metrics

This past week, I had a case of influenza. It reminded me of an incident many years ago when I began my career as an accountant.
I joined the cost accounting department for GD Searle & Co., then an independent, more than one-hundred-year-old, family-controlled public biotechnology company, now a Pfizer subsidiary. The CEO at the time was Daniel Searle, a great-grandson of the founder. Searle was known for releasing the first birth-control pill, Enovid, and for developing Metamucil, Dramamine, Celebrex, and aspartame.

The leader of my cost accounting team was Glen, who was in his forties.  Glen had worked for Searle for many years and was very hard-working and knowledgeable.  Every year at the company’s Christmas party, which occurred at lunchtime in the company cafeteria, all employees who didn’t miss work that year received a special pin from Daniel Searle.  Searle called the names out of the people who had won for the year, for 5 years, for 10 years and so on. Glen and a small group of people were long-term winners. Glen prided himself on never missing any work days (excluding vacations and holidays.)  He was very proud to wear his pin to work, and receiving it from Daniel Searle was his year’s highlight.  Other awards, such as years of service to the company, were also given out.

My desk was behind Glen’s in an open-air room containing about 30 or 40 workstations. We all reported to a manager named Wally, who ran the cost accounting department.

It just so happened that year that there was a pandemic caused by what was then called the Hong Kong flu. One morning, Glen came to work with boxes of tissues and was sneezing and coughing. It was apparent that he was suffering from the flu. I told Glen he might go home and rest since he had already “clocked in” for the day. Glen wouldn’t hear of it and continued to work—and to sneeze and cough. He then came to work the following day and the day after in worse condition.

A few days later, I contracted the Hong Kong flu and was so sick I could not get out of bed. It took me about one week to feel well enough to return to work. Others in the department also contracted the flu. There was an elderly lady named Ida who needed hospitalization. Thankfully, nobody in our department died from it, as it killed more than one million people globally.

However, Glen kept his record of not missing any workdays intact, even though most of the department missed many days.  This is an unfortunate example of the unintended consequence of rewarding people for showing up for work, regardless of illness. I am sure that Searle’s intention was not to reward people for showing up at work ill and infecting many others.

A few years later, I managed a Technical Support team for a software company.  I visited a call center at another company and observed a young man taking a call who earnestly tried to resolve a customer’s problem.  After a while, the support person asked the caller if he could immediately call the customer back.  The person agreed, so the tech support person called the customer back.  He then proceeded to resolve the problem.  When I asked him afterward why he interrupted the call to ask permission to call the customer back, he informed me that he is measured by how long it takes him to resolve a problem.  If it takes more than five minutes, he receives a demerit of some sort, so to avoid that, he stops the call at over four minutes and then calls the person back, thus also incurring a telephone charge for the company.  He would have disconnected if it had gone over four minutes and asked to call the customer again.

As an accountant and a sales manager, I believe in metrics and accountability. However, these two examples show that metrics are sometimes used to measure the wrong activity and may be harmful.
In the first example, Searle should have changed the award to reward all perfect attendance records, excluding excused absences. It should also have measured his productivity in his accounting role, such as the number of items he processed that week, month, or year.

In the second example, the goal should have been simply resolving the customer’s problem and ensuring their satisfaction rather than annoying them with unnecessary callbacks. The goal should have been complete customer satisfaction, not quickly resolving calls within a defined time limit, with penalties if you don’t achieve it.
I have never found a negative incentive that worked to increase productivity, customer satisfaction, or employee morale.