In my previous blog, I described how Net Promoter Score (NPS) works. Businesses like it because it is a simple metric that appears to indicate how well the business is doing in satisfying its customers. Satisfied customers produce positive word-of-mouth. Let’s quickly recap how NPS is computed.
NPS is computed from the responses of customers to one question: “How likely are you to recommend (company or brand X) to a colleague or friend?” Answers are recorded on an 11-point scale where participants are clustered as follows: 0-6 = Detractors; 7-8 = Passives; and 9-10 = Promoters.
NPS is computed by subtracting the percentage of Detractors from the percentage of Promoters. If 30% are Detractors and 50% are Promoters, NPS is +20.
Here are my top 5 concerns:
1. The scale is unfit for use. NPS uses an 11-point scale. All points on the scale, with the exception of 0, are positive. Zero means there is no chance of recommending the company or brand. Point 1 on the scale, therefore, must logically mean there is a small positive chance of recommending. Higher scores mean higher probability of recommending. There are no negative points on the scale to suggest that customers are likely to give negative word-of-mouth. A preferred scale would range from, say -5 to +5 where the minus score means that the customer is likely to give negative word-of-mouth and positive scores reflect positive WOM intentions.
2. The question is unfit for use. NPS measures intention not behaviour. Market researchers globally now accept that a customer’s intentions are not always accurate predictors of their behaviours. Contextual conditions often get in the way. For example, I may want to recommend a particular model of car to my friend, but I don’t do so because I know it would be too expensive for him. Do a web search on the link between intention and behaviour if you don’t believe me.
3. The clustering rule is wholly arbitrary. In NPS, 0-6 are deemed to be Detractors; 7-8 are Passives; 9-10 are Promoters. Why? Why not, say, cluster the groups are follows: 0-4; 5-7, 8-10? Are customers who score 7 or 8 really Passive, when they are at the more ‘likely to recommend’ end of the scale? Are customers who are at the midpoint of an 11-point scale (i.e. at point 5) really likely to be Detractors, that is, to speak negatively about the brand?
4. The NPS score is meaningless. The same NPS can be reached in different ways, and therefore reflects different market place conditions. Take the NPS of +20, for example. This score could be achieved by a number of splits, including the following: 60% Promoters/40% Detractors; or 30% Promoters/60% Passives/10% Detractors; or 50% Promoters/20% Passives/30% Detractors. In other words, a single score can mean different things.
5. NPS is not actionable. Let’s say your brand has a NPS score of -10, telling you that there are more Detractors than Promoters. So what? What action can you take? The score tells you nothing about how you fare against competitors and gives no guidance about the actions you could take to close the gap. Are you too expensive? Is quality too low? Is there some important element of customer experience that is unloved? Does the product not do as promised? Is after-sales service sub-standard? NPS as a single stat gives no guidance for action. If you can’t act on the measure, why would you bother measuring at all?