NPS is an outdated metric that gets customer satisfaction all wrong

Most companies will be familiar with the net promoter score (NPS). For many, it is the go-to metric for measuring customer satisfaction, and considered a box ticked.

Introduced by Bain & Co in 2003, the NPS comprises just a single question, while claiming to offer deep insights into how customers feel about a brand; its apparent strength lies in its simplicity. 

Recently, however, research has called the reliability of the NPS into question. Worryingly, it has been found to produce misleading results, overstating the number of customers who are unhappy with a business. This suggests many businesses are not getting the insights they need. Worse still, it may be driving a wasteful misallocation of resources, as businesses seek to address problems that don’t exist.

A quick but ineffective primer

We’ve all been asked the question immediately following an interaction with a brand: “On a scale from 0 to 10, how likely is it that you would recommend us to a friend or colleague?”

The score we give gets us categorised as a “promoter”, a “passive” or a “detractor”.

A nine or 10, and we’re labelled a “promoter”. This is interpreted to mean that we’re likely to recommend the brand or product.

Customers who respond with a seven or eight are deemed “passives”. They are thought unlikely to refer others, and if they do, their recommendations will be qualified or unenthusiastic. At the same time, “passives” are considered unlikely to criticise a brand.

“Detractors” are customers who respond with a score of six or less. They are taken to be genuinely unhappy, to the extent that they are likely to actively criticise a brand to friends and family, or on social media.

To calculate its NPS, a business subtracts the percentage of “detractors” from the percentage of “promoters”. For example, if 40% of your customers are “promoters”, 50% “passives” and 10% “detractors”, your NPS is 30.

Where NPS gets it wrong

The problem is that research suggests that most people categorised as “detractors” are unlikely to criticise a brand. Many may, in fact, be loyal customers.

ForeSee surveyed 20,000 people categorised as “detractors” by the NPS. Just 1% said they would communicate their “bad” experience to others. Further questioning revealed that 60% of “detractors” were loyal customers who had been using the brand for more than two years and intended to continue doing so. 

Overall, the NPS was found to overstate the number of a brand’s detractors by up to 270%. This is a major problem, suggesting serious reliability issues with the NPS.

The impact of this problem has been exacerbated by trends over recent years. Since the NPS was developed, consumer voices have been amplified by the advent of social media, giving unhappy customers the capacity to do serious reputational damage. This has driven an increasing focus on identifying “detractors” and taking steps to mitigate their concerns. However, given what the research says about the unreliability of this aspect of the NPS, much of this activity (and investment) is potentially misdirected.

An alternative to NPS

One solution is to stop using the NPS. Given its lack of reliability and potential to drive poor decision-making, this would seem to be a no-brainer.

The emperor’s wearing no clothes, right?

nlighten recently did a snap poll on LinkedIn and found that a majority of businesses prefer the client satisfaction score (CSS). This asks clients or customers: “On a scale from one to five, how satisfied were you with [an interaction or product]?” Ratings are given on a typical five-point scale: “very dissatisfied”, “dissatisfied”, “neither satisfied nor dissatisfied”, “satisfied” and “very satisfied”.

But what about interpreting the results? After all, this is where the NPS lets businesses down.

The key lesson here is to not read too much into a single number, derived from just one question. Understanding how clients and customers feel is just not that easy. A better approach is to look at quantitative data of this type as a guide, which can then drive further research.

As for the NPS, potentially it still has its place, so long as you don’t read too much into the results.

*Nathalie Schooling is the CEO of nlighten