Many companies have adopted NPS (net promoter score), as a means to measure the effectiveness of their CX programmes. Companies that improved their NPS scores found that overall business performance improved. Heck, NPS has even been backup by a Harvard Business Review article. Everything published in HBR must be true – right?

Companies have approached me, however, with an interesting problem. Several years after having implemented NPS, companies are now beginning to find out that despite having high NPS scores their business performance (sales, profits, and other financial metrics) are heading south. So what has happened?

Working with several firms from a number of different regions (US, UK, Europe, Middle East), belonging to a variety of business sectors (from FMCGs, Telecos, consumer electronics, real estate to big oil), we’ve made some important discoveries which I wanted to share with you.

NPS is not the magic potion that cures all

First, lets look at NPS in-depth. NPS asks one question, and one question only… “How likely is it that you would recommend [brand] to a friend or colleague?”. Based on this question a score is presented. A high score means that you’re doing good, and a lower score means you need to improve.

Likely to recommend, and actual recommendation are two different things

There are several issues with this approach. First, you are asking customers about their likeliness to recommend. This can be very different from actual recommendation. For example, you go to a restaurant, have a lovely meal and enjoy the customer service. You would naturally give a high score, maybe a 9 or 10 for recommendation. When you get home, your tummy starts rumbling – food poisoning. Not only will you not recommend this restaurant, but you will most likely do the opposite, tell people not to go there. But guess what, the NPS metric says you’re doing good since the same customer had given you a high rating.

There is another problem with this approach. Research has found that over 80% of the customers that say they will recommend the brand, never actually do recommend it to others. There are customers who say they would never recommend your brand to others, but keep returning themselves? What if you’re the only brand within a 100 mile radius? What if you’re the cheapest (i.e. lowest priced product)? Customers may be saying they would recommend you, but why are they doing it? Are they saying they would recommend because you’re good, or because you have the lowest priced products?

Over 80% of the customer that say they will recommend the brand, never actually do recommend it to others.

The netpromote.com web site states that “NPS, measures customer experience and predicts business growth”. The last time I checked, customer experience was a much more complex variable than just recommending the brand. What happened to other factors such as satisfaction, quality of the product / service, etc. Relying on NPS to give you an accurate prediction of how well your CX / CEM programme is doing is not very useful. Its like saying, because the girl at the pub smiled at me, she wants to marry me. We’re not really living in a dream world of a Bollywood movie. Similarly, how can we say that this one score will predict not only great customer experience but also business growth?

Another issue with NPS is the way its calculated. The categories they’ve developed appear to be fairly accurate. Customers that rate you 9 or 10 ought to be the promoters. Those within the 7 to 8 range should be passive, etc. But why are we adding these scores up? Actually you end up with three different scores. Each one needs to be treated separately. The three categories roughly correlate with the types of loyal customers.

Happy customers are usually less likely to tell others. In fact, (again backed by research), customers that are unhappy are more likely to tell others. Does everyone remember the guy whose guitar was broken by United Airlines? I’m sure United would have given an equally amazing service to customers. However, we don’t really see customers singing songs (posting them on YouTube and generating millions of hits) about the wonderful service they’ve received. Hence, a promoter is not really equal to a detractor.

Extremely loyal customers, who would never switch to a different brand are the ones that are emotionally attached to you. These customers, arguably, would fall under the promoter category. A second category of loyal customers like you brand, prefer your brand, are highly satisfied with your brand, but may switch, are classed as attitudinally loyal customers. These customers could be classified as passive, under the NPS framework. Finally, there are customers who only buy from you because they have no other choice. Maybe you are the only brand available, or offer the lowest priced products. These customers may not even be satisfied with your brand. In loyalty terms, we are only measuring their behaviour (returning to the brand) and hence classify them as behaviourally loyal customers. The NPS framework calls them detractors. (You can read more about loyal customers in this great article).

There are indeed several different types of customers. Each has their own characteristics. Each one needs a unique approach to managing. A one size fits all approach, as with the single figure NPS score, does not show you what is really happening, nor does it help you in dealing with the customers in the most effective manner.

While the NPS is a useful metric, it most certainly is not the magic number you need to be looking at. Customer Expereince, is much broader than this. If you want to talk about your NPS or Customer Experience programme, feel free to get in touch.

Help! Our NPS is not working, what do we do?