Most organizations consider customer service as a value. They’ll tell you it’s important to them to treat their customers well. But how do they know what their customers are thinking, how do they accurately and reliably measure how they are doing?
Typically when asked how companies measure customer service we hear one of three things:
- They have no measure
- They have an unreliable one (but they don’t know it)
- They have an unreliable one (and they do know it, but that’s what you have to put up with in customer service)
One of the most common ways we see unreliable customer service surveys is in the use of email or “back of receipt” surveys for data gathering. You’ve probably experienced this before – you go to a store and at the bottom of the receipt there is a website where you can take an online survey on your experience in that store. If you take the survey, (and I do now, since I’m fascinated with how others get their customer service metrics – but my guess is most of you probably do not) you may be asked to answer anywhere from 10—50 questions on your experience. This survey data is used to rank the stores and to get feedback on all the elements of the customer service experience.
While this customer service data may be useful in understanding individual customer experiences, there is a significant flaw in gathering customer service data this way: The data is gathered from a skewed sample and is not representative of the population of your customers.
Think about it, who takes these surveys? At FranklinCovey we have learned that the majority of people who take these surveys are rarely representative of the whole customer base. Too often, organizations confuse the sheer number of surveys with the quality of the surveys. This is where random sampling with high response rates is so important – a random sample of your customer base will more accurately predict the experience of all of your customers, than just those people who took the time to go online and take your survey. Why? Because you don’t know if those people who went online have the same opinions as those who did not. And this is particularly true when you offer an incentive to take the survey. Not only do you get a skewed sample, but you usually get higher scores because sub-consciously people hope that saying nice things will make it more likely that they’ll win.
Good sampling technique is critical to get an accurate, reliable customer service metric, otherwise you’ll end up only hearing from those who love you, who hate you, or who wanted to win a prize. This is the first part of FranklinCovey’s Customer Loyalty Practice value proposition, that we provide an accurate, reliable metric at every client store/location, every month.