By Martha Brooke
Guest Columnist
Customer satisfaction just doesn’t matter. Jones and Sasser proved this more than ten years ago in their research published by the Harvard Business Review. Their discovery was that the customer who is a little bit more satisfied is NOT a little bit more likely to buy from you again. However, the customer who is TOTALLY satisfied is six times more likely than the merely satisfied customer to become your much coveted, multi-time buyer. Nailing customer-satisfaction measurement is crucial. Get it a little wrong and your plans and revenue projections will be sorely misguided. Get it right and your action plans will create real-world results.
Is your fundamental hypothesis correct?
Satisfaction surveys will produce false information if the original assumptions are incorrect. The most common incorrect assumption is the mistaken belief that a questionnaire asks about everything the customer thinks is important.
Are you asking what you really want to know?
One way companies try to figure out if their customers are sincerely satisfied is to ask: Would you refer us to a friend? Quite possibly, at one time, this question did produce clear answers, but the question is in such high use now that it has become a junk question. What about asking: Were you tickled pink? Did you get a warm and fuzzy? The point is, get the customer to think about if they really benefited from or enjoyed your products and services.
Is your measurement based on a misrepresentative population?
Generally, satisfaction programs favor a particular group of customers because information about a particular group is more accessible.
Are your sample sizes too small?
If your sample sizes do not accommodate a 90 percent or greater confidence level with an error rate less than 10 percent, you simply don’t have enough data to know what’s really going on.
Is your measurement based on you or on your customers?
Your satisfaction survey may use terms that make sense to you but not to the customer. For example, a hotel client asked customers to rate "housekeeping" but what the customers really cared about was "a clean place that felt good to sleep in at night." When this client changed their survey to use customer wording, their satisfaction rating declined.
Does your measurement allow for conflict of interest?
Even when there are no bonuses tied to quality measures, nearly every employee wants to excel at their job and get ahead. Because of this, sometimes employees evaluate themselves and their teams higher than the evidence justifies.
Is your measurement based on your customers’ priorities?
No matter how comprehensive your criteria set is, if quality attributes are not weighted based on customer expectations and needs, your scoring will produce false information. Find out from your customers what’s most important to them and by a factor of how much—then incorporate this information into your analysis.
You owe it to your company to find out what’s really going on with your customers. Armed with the truth, you can make changes that will help you to succeed.
Martha Brooke is president of Interaction Metrics, an independent customer-service research company located in Portland, Oregon. Visit www.interactionmetrics.com.
