I hear companies claiming they have x% or y% of customer satisfaction, proudly claiming Net Promoter Score (NPS) of so much and that many returning customers. While these figures seem great at first sight, this doesn’t tell us much.
Read further to understand why satisfaction and loyalty scores must be interpreted carefully. We will once again use the example of the Belfius bank to illustrate this article. You will in particular discover some essential rules to respect when designing satisfaction questionnaire.
Is the satisfaction survey biased ?
As we have seen previously with Belgium-based Belfius bank, satisfaction scores can be easily manipulated, leading to extremely high satisfaction figures.
While this is difficult to assess for non specialists, one key question to ask yourself when faced with satisfaction survey scores is whether the survey was conducted in an unbiased way.
How can you do this ? If you have access to the survey itself this is pretty easy. Here are some rules to respect when designing a satisfaction questionnaire :
- Is the question really necessary ?
- To obtain the information you’re looking for, will one question be sufficient ?
- Wouldn’t it be good to confirm the results of one important question with another, slightly differently formulated, question later in the survey ? (this is called triangulation by the way)
- Will the respondent be able to answer the question asked ?
- Will the respondent be willing to answer the question asked ?
- Is the formulation of the question clear and unbiased ? Have you used words that would lead the respondent to answer in one way rather than in another ?
What are the satisfaction levels in the industry ?
Different industries command different customers expectations, hence different customer satisfaction levels.
The American Customer Satisfaction Index is a good reflection of this. When analyzed by industry it shows remarkable differences. Subscription Television services have gotten the lowest satisfaction scores (around 63-64 out of 100) since included in the ACSI (2001). On the other end television and video players have been systematically above 80 and have reached 87/100 in 2016.
To come back to the example of Belfius, we showed how unrealistic their satisfaction scores where by just comparing them to the rest of the industry. The discrepancies obtained spoke for themselves and completely discredited their results.
Are you measuring the satisfaction of non-customers ?
Most satisfaction surveys are sent to customers. Yet one amazing source of insights is forgotten in the process : non customers.
While it is interesting to measure the satisfaction of those who purchased from you, I think it’s much more interesting to measure the dissatisfaction from those who didn’t.
What I use to do when I get the opportunity is to ask our prospects why they did NOT chose us (yes, this happens !). While it is a difficult experience (because you’ll be faced with your weaknesses) I think this is a very rewarding experience. Not repeating failures is the best way to improve your business.
Coming back to Belfius, we did some market research in June / July to assess Belgian banks promises, and the results were extremely interesting. We found out that out of the ING, BNP Paribas Fortis and KBC were extremely responsive to our demands while Belfius showed some very dissatisfying practices (no follow-up, high response time, lack of answers) eventually leading to completely discarding them.
Keep in mind those 3 important principles when conducting or assessing a satisfaction survey :
- assess whether or not the survey itself (i.e. the satisfaction questionnaire biased or not)
- evaluate how satisfaction scores obtained look like when compared to the satisfaction levels of the industry
- evaluate the impact of surveying non-customers