Among all the metrics that companies use to measure their performance, one that has got much attention in the last ten years is for sure the NPS, or net promoter score
The NPS measures the willingness of customers to recommend your product or service to others. It is a score on a scale from 1 to 10, where customers evaluate themselves. Based on the given answers, customers are classified into three categories.
Detractors, who gave a score from 1 to 6, are basically people that are not satisfied with your product, and are far away from recommending it.
Passives, who gave a score from 6 to 8, are people that are ok with your product but are not big fans either.
Promoters, who gave a score from 9 to 10, are so excited about the experience with your product that would recommend it to others, acting as your ambassadors.
The Net promoter score is then calculated by subtracting the detractors from the promoters.
It sounds like an interesting indicator, right? In fact, many of the largest companies in the world use it extensively. For some others, the NPS is even part of the strategic goals. So how come we say the NPS does not work?
First of all, by asking customers if they would recommend your product, you are asking them to imagine something that they might do in the future, and not to describe something that already happened. It is proven that by doing this, you are adding a lot of bias to the results, which become unreliable
Secondly, we shouldn’t forget that the NPS is still a survey, and usually, there are other questions together with the main one. By adding other questions to the NPS we are adding a layer of complexity for the respondents, and we add even more bias to the answers, which are not free anymore but guided
Third, it is hard to learn something from it. Even if we decide to take the risk and trust the NPS, we would hardly extract any insights, unless we add an open text question. In that case the challenge becomes reading all the free text answers. If you don’t do that, the NPS would product only quantitative information, but won’t tell you “why” someone is happy or not with your product
It’s hard to accept that today’s most iconic marketing metric it is unreliable and obviously overvalued! However, you can still run experiments to verify the quality of the results. For example, you can change the main question from “will you recommend us” to “have you recommended us?” In this way, we would talk about the past, and It would become much easier for customers to answer.
If you still want to run the NPS, at least make sure it contains only two questions, where the first is the score from 1 to 10, and the second is a free text, where customers are free to write whatever they want
If you don’t want to use the NPS anymore, you should consider alternative sources of feedback, such as reviews, social or interviews! Remember that voluntarily generated feedback is by far the most insightful, and it’s also the cheapest to get!