Is NPS Still Relevant? Why Businesses Should Not Abandon It
More organisations are moving away from Net Promoter Score because it has been overused, misapplied and expected to explain more than it can. That does not make NPS irrelevant.
NPS still provides a useful measure of customer advocacy. The problem begins when it is treated as a complete measure of customer health, account risk or team performance.
Businesses should not abandon NPS. They should use it more carefully and combine it with the other signals needed to understand the customer relationship.
What does NPS measure?
Net Promoter Score measures how likely a customer is to recommend a business, product or service. Respondents answer on a scale from zero to ten and are grouped as:
Promoters: scores of 9 or 10
Passives: scores of 7 or 8
Detractors: scores from 0 to 6
The percentage of detractors is subtracted from the percentage of promoters to produce a score between minus 100 and 100.
Its value lies partly in how demanding it is. A customer who is reasonably satisfied is not automatically considered an advocate. That distinction can be uncomfortable, but it helps separate acceptable experiences from genuine customer loyalty.
Why are organisations moving away from NPS?
Customer relationships have become more complex. A customer may interact with sales, onboarding, product, support, billing, account management and leadership. Their view of the organisation is shaped by all of those experiences.
Businesses also have access to far more customer data than they once did. Product usage, support interactions, call transcripts, CRM notes, adoption, complaints and renewal activity all reveal something about the health of the relationship.
Against that backdrop, one survey score can feel limited. NPS can indicate whether a customer is willing to recommend the business, but it cannot explain the reason on its own. Nor can it reliably predict renewal, measure product adoption or assess the performance of a particular team.
These limitations do not make NPS ineffective. They simply define what it can and cannot tell you.
How has NPS been misused?
One of the most common mistakes is making Customer Success, Support or Service teams directly accountable for NPS.
Those teams influence the customer experience, but they do not control every factor shaping advocacy. Product quality, pricing, reliability, implementation, commercial terms and strategic decisions may all affect the score.
When people are held accountable for a measure they cannot fully influence, the survey process can become distorted. Teams may coach customers, survey only those likely to respond positively or delay requests following a poor experience.
This behaviour is not always driven by bad intent. It is often a rational response to a poorly designed incentive.
When that happens, NPS no longer reflects the customer relationship. It reflects how effectively the survey process has been managed. The metric has not failed; its governance has.
Has survey fatigue weakened NPS?
Customers are now asked for feedback after almost every interaction. Support cases, purchases, onboarding sessions, deliveries and renewals can all trigger another request.
The problem is no longer only whether the question is useful. It is whether customers are still willing to answer thoughtfully.
Survey fatigue lowers response rates and weakens data quality. Even a well designed question produces poor information when it is asked too frequently, at the wrong time or of the wrong person.
Replacing NPS with Customer Satisfaction, Customer Effort Score or an AI driven survey will not solve that problem by itself. Businesses must also consider when they ask, whom they ask and whether they act on what customers tell them.
Is NPS still relevant in B2B?
NPS remains relevant in B2B, but the respondent matters.
A B2B account rarely has one customer perspective. End users, administrators, internal champions, economic buyers and executive sponsors may experience the relationship differently.
An end user might recommend the product while the economic buyer questions its return on investment. A senior sponsor may support renewal despite frustration among users because the product remains strategically important.
A single response cannot represent the entire account. Businesses need to know whose sentiment they are measuring and how much influence that person has over adoption, expansion and renewal.
NPS can contribute to account health, but it should not be treated as definitive evidence that an account is safe or at risk.
What should businesses measure alongside NPS?
There is no single measure of customer health. It is multidimensional and may include:
Customer advocacy
Satisfaction and effort
Product adoption
Time to value
Realised customer outcomes
Support history and operational friction
Stakeholder engagement
Renewal and expansion
Commercial risk
Qualitative feedback
Different measures answer different questions. NPS indicates advocacy, Customer Effort Score identifies friction and Customer Satisfaction assesses a particular experience. Product usage reveals adoption, while renewal and expansion data show commercial outcomes.
None provides the whole picture alone. The goal should be to create a customer measurement system rather than find one perfect score.
Customer measurement must evolve as the business grows
A smaller business may understand its customers through direct conversations and a limited number of measures. As it grows, the number of products, segments, channels, markets and stakeholders increases. Senior leaders also become further removed from customers.
The measurement model must evolve with that complexity.
A growing organisation cannot expect one survey score to explain thousands of customer relationships. It needs to connect sentiment with behaviour, operational performance, commercial outcomes and qualitative feedback.
Context remains essential. A decline in product usage may be insignificant for one customer and a serious risk for another. A poor support score could indicate a service issue or expose a deeper product problem. A high NPS may reflect genuine advocacy or simply the goodwill of a user with no influence over renewal.
Reducing customer health to one number or dashboard colour removes too much of that context.
How can AI improve customer measurement?
AI can help businesses make better use of the customer information they already hold.
Support tickets, call transcripts, survey comments, complaints, CRM notes and renewal conversations contain valuable insight, but much of it has historically been difficult to analyse consistently at scale.
AI can identify recurring themes, detect changes in sentiment and connect feedback across channels. It can also help explain movements in NPS, satisfaction, adoption or renewal risk.
The best use of AI may not be asking customers more questions. It may be extracting more value from information they have already provided.
AI still has limitations. It can misread tone, miss context and overstate confidence. Existing data may also overrepresent customers who complain or engage frequently while overlooking quieter customers.
Its role should be to surface patterns and prioritise investigation, with people providing the context and judgement.
How should NPS be used?
NPS should be collected selectively, consistently and without influence.
Businesses should be clear about:
Why the survey is being sent
Which stakeholder should receive it
When it should be sent
How frequently customers should be asked
Who can access the results
How the feedback will be investigated and acted upon
The score should then be interpreted alongside behavioural, operational, commercial and qualitative evidence.
Teams should primarily be measured on outcomes they can materially influence. Support teams might be assessed through customer effort, resolution quality, repeat contacts and escalations. Customer Success teams might focus on adoption, time to value, risk identification, stakeholder engagement and agreed customer outcomes.
NPS can inform the broader view, but it should not determine an individual team’s performance or an account’s health on its own.
Do not replace NPS with another favourite metric
The debate should not be framed as NPS versus Customer Effort Score, NPS versus AI or NPS versus a customer health score. That simply replaces one form of overreliance with another.
The better question is whether the organisation has a measurement system capable of reflecting the complexity of its customer relationships.
NPS still has a place within that system. It should be used less frequently, interpreted more carefully and given a clearly defined purpose.
The organisations that understand their customers best will not be those that discover the perfect metric. They will be those that combine the right signals, interpret them in context and act on what they learn.