The Customer Service Survey

Above-Average Customer Service

by Peter Leppik on Tue, 2005-09-13 01:00

According to popular wisdom, 80% of drivers think that they are above average (the actual number varies from 60% to 90% depending on the country). If the distribution of driver talent is close to normal, one would expect that only 50% of drivers are, in fact, above average.

(The statistics geek in me insists I point out that if there are a small number of really terrible drivers, then a little over 50% of drivers actually could be above average. The more relevant statistic is the median, since by definition exactly 50% of drivers will be better than the median. But I hesitate to state that "50% of drivers are above the median," as this makes me think of hordes of flying cars zooming over the center of the freeway. Or as Dan aptly put it, "Any driver who is above the median is not above average.")

Statistical puns aside, we see the exact same phenomenon in customer service: most companies think they provide above-average customer service, yet only about half of companies actually do.

Part of this is because it is only human to be optimistic and think we're doing better than we actually are. In the absence of hard data (most of us in the U.S. only take one or two driving tests in our lives), the natural inclination is to believe we're doing pretty well.

But another factor is that many customer satisfaction surveys are not designed to tell companies how well they're really doing--they're designed to tell companies what they want to hear. And most companies want to hear that their service is above average.

I've had to be the bearer of bad news on many different occasions. It isn't easy. We want to please our customers, but we also need to maintain our objectivity and the validity of our results. We use a set of purely statistical benchmarks, so there's no wiggle room to "adjust" the scores when someone does poorly. We don't have the ability to go back and say, "Maybe that call wasn't so bad after all," since the data is what it is.

But not all benchmarks are similarly objective. Any time a consultant or other professional is scoring call recordings (even if the scoring criteria are supposed to be completely objective), there is room for subjectivity to creep in. Even customer surveys need to be carefully considered, since slight changes can make a big difference. If question number one is "Was the agent friendly" (most people will answer "yes"), then the whole rest of the survey is likely to be slanted in a more positive direction.

If your objective truly is improving the quality of customer service, then the news you most want to hear is the bad news: what needs to be improved, what's frustrating people, and what isn't working. Bad news tells you where things are coming up short, and where you can have an impact. Good news rarely provides any action items other than "do more of the same."

But since we all want to be above average, sometimes the bad news can be hard to swallow.

Sorry...This form is closed to new submissions.