NPS is dead! (The Problem and the Solution)

Net Promoter Score (NPS) has long been a popular metric for businesses to measure customer satisfaction and loyalty. However, as the business landscape and customer expectations have evolved, NPS is no longer the relevant metric it once was. In this article, we will explore the reasons why NPS is no longer a reliable indicator of customer sentiment and why businesses should consider other metrics.

Before we move ahead, let's take a trip down memory lane:

The Net Promoter Score (NPS) was first introduced in 2003 by Fred Reichheld, a consultant at Bain & Company, in his book "The Ultimate Question: Driving Good Profits and True Growth." The NPS system is based on a single question: "On a scale of 0-10, how likely are you to recommend our product/service to a friend or colleague?"

Customers who respond with a score of 9 or 10 are considered "promoters," those who respond with a score of 7 or 8 are considered "passives," and those who respond with a score of 0-6 are considered "detractors." The NPS is calculated by subtracting the percentage of detractors from the percentage of promoters.

Reichheld and his colleagues at Bain & Company conducted extensive research on the correlation between a company's NPS and its growth rate. They found that companies with high NPS scores tended to have faster growth rates than those with low NPS scores. This correlation led to the widespread adoption of NPS as a metric for measuring customer satisfaction and loyalty.

NPS quickly gained popularity among businesses of all sizes and industries, as it was easy to understand and implement. Many companies began using NPS as a key performance indicator (KPI) and it became a standard metric in customer experience management.

However, over time, NPS has been criticized for its limitations, including its reliance on a single question and limited ability to provide detailed feedback. Many experts believe that NPS is no longer the relevant metric it once was, and that businesses should consider other metrics, such as natural language processing (NLP) based language models, to gain a deeper understanding of customer sentiment.

The Problem!

One major limitation of NPS is that it relies on a single question: "On a scale of 0-10, how likely are you to recommend our product/service to a friend or colleague?" While this question may provide a general sense of customer satisfaction, it fails to capture the nuances of customer feedback. NPS doesn't allow businesses to gain a deeper understanding of what is driving customer satisfaction or dissatisfaction. Additionally, NPS doesn't provide any actionable insights or suggestions for improvement.

Another limitation of NPS is that it is a static metric. It doesn't take into account the dynamic nature of customer sentiment and can't provide any insight into how customers' opinions change over time. In contrast, newer metrics such as natural language processing (NLP) based language models can provide businesses with detailed and nuanced feedback that can help them identify patterns and trends in customer sentiment over time.

In addition, NPS is also limited in terms of the sample size. It's difficult to get a statistically significant sample size to generalize the results. Also, it's heavily influenced by the population who are willing to give feedback, which might not be a representative sample of all customers.

Furthermore, customers have evolved and their expectations have changed. They now expect businesses to be more responsive, empathetic, and personalized. NPS, with its single question, doesn't provide any insight into how well a business is meeting these expectations.

In conclusion, while NPS has been a useful metric in the past, it's no longer relevant in today's business landscape. Businesses should consider other metrics that provide a more detailed and nuanced understanding of customer sentiment. These newer metrics, such as NLP-based language models, can provide businesses with actionable insights and help them stay ahead of the curve in terms of meeting customer expectations.

The Solutions!

Instead of just measuring NPS, also measure TLM (Total Loyalty Metric). NPS helps you understand how likely someone is to recommend your business to others, and TLM helps you understand how likely someone is to do business with you again. This opens doors for categorizing your audience into the following bucket:

1. Will do business with you and recommend you to others (Pure Promoter)

2. Will do business with you but not recommend you to others (Forced Customer)

3. Will not do business with you but will recommend you to others (Distant Promoter)

4. Will not do business with you and will not recommend you to others (Pure Detractors)

Now, for each of these segments if you append a 'WHY?', I believe the insights will be astounding and 'actually actionable'.

The problem with NPS is that you cannot predict churn (despite a high NPS). And while a high NPS will lead to higher retention rates, it does not solve for loyalty.

The problem with TLM is that it will be highly skewed in the 'Unsure' and 'Somewhat Likely' zones as 'Very Likely' seems like a commitment for a repurchase. Customers want control and command in their hands while making a purchase decision, which will sort of lead to lower levels of commitment on this 4-point scale.

Let's call it NPLM - % in Bucket 1

This would help you measure the % of your customers that are with you because you did not incorrectly sell something, are happy with your services, and think that you are better than your competitors. Which is the ideal state any business wants to be in.

Given that metrics like NPS are a part of most investor/board presentations, NPLM would be a better metric for any investor to simply deduce if it is worth investing in a business or not.

1 drawback of NPS was you could not use it to benchmark performance against your competitors. With NPLM, you can. You can use NPLM as an internal performance metric for various customer-facing teams, but also use it as an external business performance metric.

Let's say one of your customers has an interaction with your customer support agent:

Q1. How likely are you to call our support agents again to resolve your queries? (TLM)

Q2. How likely are you to recommend our support agent to fellow <brand> customers? (NPS)

You have the ultimate customer satisfaction metric for your business!