How to Improve Net Promoter Score®: The Complete Guide

How do you make important or day-to-day consumer decisions in your life? Whom would you consult if you were choosing a hairdresser's saloon or a new car to buy? 

83% of customers would trust recommendations from the people they know: colleagues, family, friends, etc. Often, we believe that people close to us can give us a better advice when choosing a product or service.

According to the same research, at least 66% of customers trust other consumer opinions posted online and according to another research, 58% of consumers said they have recently (within the past five years) began leaving more and more online reviews based upon customer service. 

That proves that referral marketing plays a more and more important role in business strategy. 

The question is, how can you measure it?

The Net Promoter: what is it?

The Net Promoter Score (or NPS) was designed by Fred Reichheld in 2003 to measure loyalty. That time, customer experience management was still unknown to most business doers. Soon after, NPS gained popularity and used to be implemented everywhere: from customer service interaction to every individual customer touchpoint. People have been obsessed with "the only number you need to know" ever since. But it is not a surprise because NPS gives a very simple and easy to track system to follow your most and least loyal customers.

In reality, most likely, you have already seen it:

Net Promoter Score is short and simple for customers to answer.

NPS consists of two questions: a scoring system (typically asked with "How likely are you to recommend your brand to a friend or colleague?") and the "why"-question. The scoring system divides all responders into 3 groups:

  • Promoters (answered 9-10): typically loyal and enthusiastic customers likely to recommend you,

  • Passives (answered 7-8): currently satisfied but easily tempted by your competition,

  • and Detractors (answered 0-6): unhappy customers, a high likelihood for negative word of mouth, high churn rate.

Promoters are significantly more loyal, so most businesses would do well to create promoters and decrease detractors.
— Bruce Temkin

You can calculate the Net Promoter Score by identifying the difference between the percentage of promoters and the percentage of detractors. For example, if 65% of your customers are promoters and 12% are detractors, NPS is 53.

👆Note: NPS can also be negative if the percentage of detractors is higher.

Net Promoter Score Calculation Detractors Passives Promoters

Many companies are focused on following the NPS as a number. No surprise they do, as according to this research of London School of Economics an average NPS increase of 7% correlates on average with a 1% growth in revenue. Isn't that a solid reason to think about how to increase your NPS?

⛔️Don't fall into this trap. 

If your company focuses on following the score only, you will very happy if the NPS goes up. But what will you do if the score starts to fall? Many companies panic: customer research is commissioned, task forces established, and people start to read the feedback just to understand what is going on.

Keep in mind that NPS only becomes a truly valuable metric, if its “why”-question is properly collected, analyzed and heard. 

The “why”-question is the golden source for your customer experience management. It shows you why a customer gave this or that score, opens up a dialogue and, in a way, lets you dive into your customer’s head.

NPS has been a good metric to benchmark and help brands understand the overall outcome of their experience. The real value is in the “WHY”. A lot of companies will chase the score but customer-centered companies will dive into the “why” to discover what is creating the experience and move to action to circle back to the customer or design new experiences. Here NPS becomes less about the score and actually about the experience.
— Diane Magers, CEO of CXPA

To summarise, NPS can be very efficient as it provides both the quantitative metric to set targets for and follow up and also the free text feedback for insight generation. At the same time, an NPS survey is simple and short enough for consumers to respond regularly. 

How to collect NPS: Best Practices

Here we've collected the best practices, tips and most common problems our customers come across when implementing and, later, using NPS.

1. First things first: Get that wording right!

The more popular NPS was getting, the more misused the metric became. The Net Promoter System is an open source system, which means you can change&customize the questions for your own purpose. Unfortunately, this sometimes leads to misunderstanding the customer feedback and, as a result, ignoring the voice of customers as a whole.

  • "How likely are you to recommend your brand/your product to a friend or colleague?"

Net Promoter allows you to switch between your "brand" to "product", if that follows the targets of your survey. If you ask about your brand, you will receive the measurements of brand loyalty, and if you ask about the particular product or service, you will more likely receive more product-detailed feedback.

Also, NPS doesn't tie you to "a friend or colleague"-wording. Easily, you could that to any role, which is suitable for your business. The options start from "a family member" and range up till "manager" or "classmate." 

Net Promoter Score Wording Friend Colleague Family Member
  • "Why did you give that score?"

The "why"-question that brings you closer to understanding your promoters, detractors, and passives sometimes is changed to "tell us how we can improve" or "what you liked the most about us". We do not recommend this change as it leaves all the negative reasoning out, thus brings you further away from understanding the real Voice of Customer. 

  • The scaling system

Last, but not least, some of our customers asked us if they could change the response scale to 1-5, 0-7, etc. Is there a reason to do so? If you change the response scale, all the analysis and published literature would still be based on the 0-10 scale. You will lose the value of your data. For more than 10 years, the 0-10 scale has been proven to work.

In the end, one thing that matters the most is consistency.

After all, if you insist on changing the wording or scaling system of NPS, please do that with all your NPS activities. Don't mix different types of questions or scaling systems, as it's not only hard to compare, but also costs much more resources to maintain.

2. how to collect NPS

Collecting NPS in a right way ensures you reach your goals. It can also nicely increase your response rates if you know when and where you need to ask for feedback. That being said, let's think about... 

  • When to ask feedback?

Many companies often start measuring NPS at one touchpoint (which is often customer service, as it is the most obvious option) that they think is the most important. Then they fix the issues for that touchpoint and sometimes wonder why it doesn't bring any results or brings the results different from what they expected.

In order to avoid this mistake, the first must-do is to measure the overall experience of your customers. Don't go into measuring the success of your customer service or any other touchpoint before you have actually identified what problems and excitements your customers share about your products or services overall.

  • For that, first, identify the critical spot in your customer journey when you would get feedback on the journey holistically. When would the key part of the experience still be fresh in the customer's minds?

  • Only then you can ask about the most critical touchpoints if needed. If you have identified some critically important touchpoint(s), you can ask specific feedback about them as well (e.g. about your new mobile app). Do not ask feedback after each touchpoint either.

  • Also, don't forget to reschedule the NPS survey for your subscribers to take place again in 90-180 days depending on your business cycle.

NPS should be part of your ongoing customer experience management process where you collect feedback, analyze it, act based on it and finally close the loop with your customers.

  • Where to ask feedback? 

The answer to the question "Where to ask the feedback from your customers?" largely depends on what do you want to track with the feedback. Do you want to measure the overall experience? Do you want to measure the performance of one particular touchpoint? Do you want to measure the performance of an app, a product or the image of the whole company?

Here are some ideas you could use:

Where to collect customer feedback

The general rule is to always tie where you ask the feedback to the most natural placement available. Thus if you want to measure how satisfied your customers are with your app, it makes sense to ask for feedback in the app itself.

3. how to analyze the feedback

Don't end your feedback analysis by calculating the score. Understanding why your customers are happy or unhappy is the main reason why you should ask for feedback in the first place. 

We agree, that analyzing tens of thousands individual feedback comments might be not only tiring but resourceful and time-consuming. That's where text analytics technologies come into play.

Simple sentiment analysis of text analytics can divide a sentiment into three buckets: a sentence can be positive, neutral or negative. Machine learning technologies can detect the degree of sentiment as well: if someone hates your product, the negative sentiment is stronger than if he just dislikes it. Similarly, if someone loves your customer service personnel, the positive sentiment is stronger than if they just feel it’s good.

Let's take a look at this comment:

Net Promoter Score Why Question

The machine would likely give you the following results:

Topic                                  Sentiment

Camera software                 Strongly negative

Product general                   Strongly positive

Performance                  Somewhat negative

Design                            Somewhat positive

These results give you a better understanding of what the user was saying, together with the other thousands of comments the data become more generic, so that you could use it in your product&business development in the future. 

That's just one of the examples how the why-question of NPS enables you to understand the reasons behind the score. 

Now since you know why the free text feedback is a great source for insights, I could share a couple of our findings when it comes to the text feedback.

  • Consumers typically mention the most important things on their mind in the feedback.

  • Free text form is great in surfacing topics that most matter to consumers – getting all of the topics to appear with closed questions would require an extremely lengthy questionnaire.

  • Free text analysis typically includes categorizing the feedback to better understand which topics are talked about and assessing the sentiment of the sentences. This basic analysis can then be further utilized in e.g. calculating the impact of each topic on the NPS. 

Learn from the best, or NPS case studies

Often businesses pose the question: "How can we improve NPS?". That is the wrong question. What those businesses often forget, is that higher NPS wouldn't bring you any business results, if you don't understand or act on your customer feedback. 

We asked the experts who achieved great results in customer experience, how they have put things together.

First, we have asked Diane Magers, CEO of CXPA, how to tie NPS, or customer experience to the business value. "You can correlate NPS to better business results: revenue, retention, - Diane noticed, - Focusing on customer behavior and measures can help you tell the more detailed business story. Here's one example: imagine that you want your customers to go online and self-serve instead of going to call center. Here you could measure the number of people who self-serve vs. call in and it has value. The online interaction might cost you 10 cents and a call might cost you 12 dollars (in the US). In this case, the better online experience is tied to the business value."

At the same time, the NPS segments: promoters, passives, and detractors can quite accurately predict how those customers would keep the relationships with the brand. To analyze the connection between loyalty and the Net Promoter Score, Temkin Group has conducted the research "Economics of the Net Promoter Score"  in 2016.

Statistics Value of NPS Promoters Passives and Detractors Across Industries

According to the research, promoters are more than five times as likely to repurchase from companies, more than seven times as likely to forgive companies if they make a mistake, and almost nine times as likely to try new offerings from companies. The same research also shows that promoters recommend a company to an average of 3.5 people.

Then we checked how the top performing companies do NPS and we found some surprising results.

  • Slack

Slack has set up records for growth and was one of the fastest companies to receive validation of $ 1 billion. All this time, NPS has been one of their core metrics. The company not only has aimed at high scores but followed up every single customer interaction and worked hard on making it just perfect.  

Slack Net Promoter Score

Slack not only aimed at positive feedback, but they also worked hard on turning every single user into a promoter. By doing this, together with fast quality positive word-of-mouth marketing, Slack has achieved its amazing results.

We have talked with Bill Macaitis, a man behind Slack’s NPS, find 7 new tips from Bill for your NPS survey here.

  • Mention 

Mention App was proud to share that with the help of NPS they managed cut churn by half in only 2 months. Similarly, they have individually following up every promoter, passive and detractor. 

The full process in described in  this presentation .

The full process in described in this presentation.


Promoters were offered a discount, passives received an extended free trial, and detractors were thanked and asked how things could be improved. Their recipe for a successful NPS campaign is facilitated automation, top answer rates, and qualitative handling.

  • Airbnb

Airbnb has completed a research on "How well does NPS predict rebooking?". The results were quite interesting, they found that customers with an NPS score of 10 were 13% more likely to rebook and 4% more likely to refer a friend than detractors (and that without word-of-mouth marketing). 

NPS Benchmarks

Although the subtitle says "benchmark", don't rush into comparing your scores with top-performing companies. Let's find your targets step-by-step. Although I repeatedly noted that you should never chase the NPS number, I didn't mean you shouldn't think about it at all. The Net Promoter Score is a great way to follow the trend, make a check up how you're doing and the "why"-question helps you to understand the drivers behind the trends.

So, how to take control of your Net Promoter Score?

1. make sure your Net Promoter Score is positive

Having NPS > 0 is already a reason to celebrate. That literally means you have more promoters than detractors and that's a good thing.

2. Aim higher every month/quarter

Compare your NPS to your past NPS. Is it higher? You're doing a great job. Is it lower? Find out why and how you can make it better next time. Building quality relationships with your customers is your and only your business, make sure you get it right.

3. Now you can check the industrial&regional standards.

Ideally, you should search for companies in the same industry and in the same region. Comparing your results across regions or industries might be a close-to-impossible thing to do, as the standards vary a lot.

For example, the US scores have been always higher than European scores because of the culture mentality. If you want to compare European and American NPS scores across the same industry, CheckMarket suggests counting 8 as a promoter score and 6 as a passive in Europe. The scores in Australia are typically even lower because of higher consumer expectations.

Now that you know, what score your company should aim at, let's take a look at the top-performing companies from around the world. 

Top performing companies in March 2018 according to  NPS Benchmarks

Top performing companies in March 2018 according to NPS Benchmarks


Although the Tesla score or 97 seems to be unrealistic (yes high scores are a standard among electric automobile industry), the score of Amazon has been proven along the way. In another very interesting research, Amazon was chosen as the most trusted brand among SMB vendors (there're also quite a few surprising industry benchmarks in the same research).

To sum it up, the main idea is to compare your NPS with yourself and yourself only at the first place, and then try to find trustworthy relevant NPS benchmarks in your industry and your region.