NPS – some reflections

NPS Loudspeaker

In my previous blog post “Recommend to a friend”, I brought up the background to the measurement of customer loyalty with the Net Promoter Score, NPS [2]. The method is easy to use and popular among commercial companies as well as in the public sector. But, all methods based on a relatively simple model has its weaknesses. In this post, I am sharing reflections and some criticism of the NPS approach.


The NPS-method relies on the attempt to measure customer loyalty to the company. However, loyalty is an emotional and value-charged expression, having a various meaning in different cultures and for different people and situations. Loyalty is something that shows you do not betray the other and you are there to help and support [5]. Synonyms for loyalty are, for example, solidarity, faithful, reliable, loyal, and honest. Loyalty is something that must be earned and may require some reciprocity; if I am loyal to someone, I expect to get something in return. The opposite of loyal is unfair. It is also worth noting that loyalty is also used to describe a relationship between people of different status, for example, between a boss and a subordinate.

Recommend to a friend

It is a sensitive matter to recommend a company and a product to someone you care about. Fred Reichheld states that two preconditions must be fulfilled [1]:

  1. You, as a current customer, must trust that the company you recommend will provide products and services that meet your friend’s needs. The goods and services must be priced fairly in relation to their features and benefits. See more in my post User Experience, UX.
  2. You, as a current customer, must experience that the company cares about, understands, and appreciates its customers. You must be satisfied with all the interactions you have had with the company and its staff. The company’s values must match yours. Read more in my post on Customer Experience, CX.

The first prerequisite is based on facts and aligns with your brain while the second relates to your feelings and your heart.  Fred Reichheld concludes that both conditions must be met to ensure that you can recommend the company to a close friend. In the end, it is all about your credibility. You must be sure that your friend gets a good value for money and is treated well.

Critical observations

Many critics of the NPS-model consider that the reality is more complicated than a single numerical score. Some critical observations:

  • The link between satisfied, loyal customers and the business’ growth and profitability can be verified using different methods of measurement. The NPS model is only one of several models possible. Some critics believe that the “probability to recommend” does not measure anything new compared to other conventional loyalty-related measurements [3].
  • Customers who remain and are loyal are not always happy buyers. In some sectors with high replacement barriers and switching costs, demanding customers tend to stay and be loyal since a change of supplier is both costly and challenging. Some clients may have no choice – in any case, not right now [3]. An unhappy customer can also be loyal and remain with the hope that the company will improve its products, services, and support.
  • Truly loyal clients can be infatuated, blinded by the brand and the image that it creates. These customers buy everything the company offers without reflection or comparison and will, with enthusiasm,  recommend it to friends.
  • A customer that ‘s hard to deal with and complains, and makes his/her voice heard may be the most important customer. It is, in any case, a sign of commitment. If you treat this customer with respect and listen to him/her, you may gain the most long-term loyal customer.
  • Timing is critical. It is easy to ask the ultimate question, but it is crucial to ask it at the right time. The likelihood of recommendation will vary depending on when the question is asked [1].
  • NPS has certain limitations in B2B. Here, the ultimate question may be replaced by “How likely is it that you continue to purchase goods and services from the company?” [3].
  • Clients tend to get too many customer surveys to respond to, which may lead to “survey fatigue.” As a consequence, the response rate may decrease, and replies given without much consideration. Survey fatigue affects particularly “passively loyal” customers, and may lead to responses only from dissatisfied customers. A solution to mitigate survey fatigue is to agree with clients in advance about their participation and the mutual importance of the investigation [3].
  • The NPS method is less useful in certain industries and for particular products. Examples are products we buy less frequently and complex customer offerings where it is hard to distinguish between competitors. Utility providers may have difficulty using the NPS-method since customers rarely talk about or recommend their power provider. Banks are an example of a multiple service provider and where it is tricky to distinguish between competing banks and their products and services [4].
  • In all surveys, the quality of the outcome depends on the quality of the provided information. For example, the extent to which customers are prepared to enter the probabilities of the end of the scale – a zero or a ten – varies between cultures. The NPS-method is primarily developed and adapted for Anglo-Saxon cultures. In Sweden and Japan, for example, we tend to give fewer zeros and tens versus customers in the USA. In Swedes are probably more inclined to indicate the probability of the recommendation in the range of 7 or 8 – there must always be room for improvement.


Employer Net Promoter Score, eNPS, is an extension of the NPS-method used as a measure of employee loyalty. In the survey, employees are asked to answer the question “How likely are you to recommend the company as a place to work?” In reality, employee loyalty is more complicated than a single numeric score, and eNPS receives similar criticism as NPS surveys with customers. Critics say that a recommendation should be included as part of a recurring and more comprehensive study on the employees’ view of the company as an employer.


To choose the company to work with is a much bigger and more important decision than to select a supplier for a product or service. Therefore, eNPS tends to be lower than the NPS. Employees put greater demands on their employers when responding to the question “How likely are you to recommend the company as a place to work” than a customer makes when they recommend a supplier [1].


A balanced use of NPS is a valuable and useful method with which to measure customer loyalty. If an increased NPS leads to an increase in sales and higher profits may, however, also depend on other circumstances. Nonetheless, the simplicity of the measurement and the possibility of comparing variations in the score over time are essential characteristics. Still, the NPS method should not be overestimated – because of its simplicity; it also has its shortcomings. And keep in mind that the NPS process should not be regarded as a model with which to produce only one (1) score – it must reflect the company’s overall the strategic approach to its customers.

The reliance on the method depends on the quality of answers. Therefore, the next time you get the ultimate question, “How likely are you to recommend the company to a friend or colleague?” make an effort to provide an honest answer. Do not hesitate to give a zero or ten!

Last but not least, deal with the customer in the same way as you would expect to be treated.

1) Reichheld, Fred; Markey, Rob (2011). The Ultimate Question 2.0: How Net Promoter Companies Thrive in a Customer-Driven World (Rev. and expanded ed.). Boston, Mass.: Harvard Business Review Press. ISBN 978-1-4221-7335-0
2) Net Promoter Score (NPS) is a registered trademark of Fred Reichheld, Bain & Company, and Satmetrix.
4) Discussion with Melanie Tilliet Stec on a fligth between Copenhagen and Stockholm. Thanks for your input Melanie!
5) Nationalencyklopedin,

Adobe Stock.

Leave a Reply

Your e-mail address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.