

A Great Place to Upskill
Learn what Net Promoter Score (NPS) means, how to calculate it, real examples, and proven ways to improve it for your product or business.

Imagine asking a friend two different questions about a restaurant. First: “Did you enjoy your meal?” Almost everyone says yes, even about a mediocre experience, since it costs nothing to be polite. Now ask a second question: “Would you recommend this restaurant to a friend?” That one is harder to answer casually, because a recommendation puts your own reputation on the line. If the friend has a bad time, that reflects on you too.
Net Promoter Score is built entirely around that second, harder question. Instead of measuring polite satisfaction, it measures something much closer to genuine conviction, whether a customer trusts a product enough to stake their own credibility on recommending it.
In this guide, you’ll learn what NPS actually is, where it came from, how to calculate it, a step-by-step framework for improving it, how it compares to related metrics, which tools support it, and the mistakes to avoid.
Net Promoter Score (NPS) is a customer loyalty metric based on a single survey question, “How likely is it that you would recommend this company, product, or service to a friend or colleague?”, scored on a 0-to-10 scale. Respondents are grouped into three categories based on their answer, and the final score is calculated by subtracting the percentage of detractors from the percentage of promoters.
NPS was developed by Fred Reichheld, a Bain & Company Fellow, and introduced to the world in his December 2003 Harvard Business Review article, “The One Number You Need to Grow.” Working with the research firm Satmetrix, Reichheld tested dozens of survey questions against actual customer behavior across 14 industries and found that “likelihood to recommend” outperformed every other question, including direct satisfaction questions, at predicting repeat purchases and referrals.
According to Wikipedia’s entry on Net Promoter Score, Reichheld, Bain & Company, and Satmetrix jointly hold the registered NPS trademark, though the metric itself is free for anyone to use.
For Product Managers, NPS is one piece of a broader measurement picture, not a replacement for it. It’s closely tied to the metrics covered in ProductSpace’s guide to key product metrics, and getting real value from NPS depends on the same instrumentation discipline covered in Product Space guide to product analytics fundamentals, since a score without behavioral context tells you very little on its own.
Real-world example: A project management tool might see its NPS drop from 45 to 30 over a quarter. Digging into the detractor comments specifically, rather than just watching the number fall, could reveal that a recent pricing change is the dominant complaint, turning a vague “loyalty problem” into one specific, addressable issue.
NPS matters because it distills customer loyalty into a single, trackable number that’s simple enough for an entire organization, from support agents to the executive team, to rally around and act on.
Predicts word-of-mouth growth: Reichheld’s original research found the recommendation question outperformed other loyalty questions at predicting actual referral behavior
Simple enough to track consistently: one question is far easier to collect and monitor over time than a lengthy, multi-question survey
Segments customers actionably: promoters, passives, and detractors each call for a different, specific response strategy
Creates a shared, company-wide language: NPS is simple enough that support, product, and leadership teams can all discuss it without translation
Surfaces at-risk accounts early: a declining NPS often shows up before churn does, giving teams a window to intervene
Benchmarks against competitors: since NPS is so widely adopted, it’s one of the few loyalty metrics with broad industry benchmarks available
Product Managers use NPS as one signal among several when prioritizing what to fix or improve next. Customer success teams use it to flag at-risk accounts, particularly low-scoring detractors, for proactive outreach. Marketing and brand teams use it to gauge overall customer sentiment and track it over time.
Executives use it as a simple, board-friendly metric to track customer loyalty trends at a glance, something over two-thirds of Fortune 100 companies reportedly do today.
Data and analytics teams build the surveying and reporting infrastructure, closely tied to the categories covered in ProductSpace’s guide to essential Product Management KPIs.
Loyal enthusiasts who will keep buying and are likely to actively refer others.
Characteristics: highly satisfied, willing to put their own reputation behind a recommendation.
Best practice: turn promoters into active advocates through referral programs, testimonials, or case studies, rather than letting their enthusiasm go untapped.
Satisfied but unenthusiastic customers who are vulnerable to competitive offerings and unlikely to actively promote or criticize the product.
Characteristics: neither loyal advocates nor active detractors, easy to overlook since they rarely complain.
Best practice: dig into what’s holding passives back from becoming promoters, since converting even a portion of this group meaningfully lifts the overall score.
Unhappy customers who can actively damage the brand through negative word-of-mouth and are the most likely group to churn.
Characteristics: by design, the bar for landing in this group is deliberately low, since Reichheld’s scoring intentionally sets a high threshold for what counts as genuine loyalty.
Best practice: treat detractor feedback as the most valuable signal in the entire survey, and follow up directly rather than only tracking the aggregate number.
Key Insight: The 9-10 threshold for Promoters is intentionally strict. A customer who rates a 7 or 8 might sound satisfied on a typical survey, but NPS treats them as merely Passive, since Reichheld’s research found that only genuinely enthusiastic customers, the 9s and 10s, reliably translate into repeat business and real referrals.
NPS = % Promoters − % Detractors
Passives are counted toward the total number of respondents but don’t factor directly into the subtraction, which means the score can range from -100 to +100.
Worked example: if a survey of 200 customers produces 100 Promoters, 60 Passives, and 40 Detractors, that’s 50% Promoters and 20% Detractors. NPS = 50 − 20 = 30.
A positive NPS is generally considered good, above 50 is considered excellent, and above 70 is considered world-class, though what counts as strong varies meaningfully by industry.
Send the survey at the right moment: timing it after a meaningful interaction, like onboarding completion or a support resolution, produces more accurate, actionable responses than a random, unprompted survey.
Always include a follow-up question: ask “why did you give that score?” immediately after the rating, since the number alone doesn’t explain what to fix.
Segment results by customer type: a blended, company-wide NPS can hide sharp differences between customer segments or plan tiers.
Prioritize detractor follow-up: reach out directly to low scorers, since this feedback is often the most specific and actionable in the entire dataset.
Look for patterns, not one-off complaints: a repeated theme across multiple detractors points to a real problem worth prioritizing on the roadmap.
Activate your promoters: turn enthusiastic responses into referral requests, reviews, or case studies while the sentiment is fresh.
Close the loop with respondents: let customers know what changed because of their feedback, which builds trust and improves future response rates.
Never link NPS to individual compensation: Reichheld himself has warned that tying scores to employee bonuses corrupts the data, since it incentivizes chasing high ratings instead of honest feedback.
Track the trend, not just the snapshot: a single quarter’s NPS matters less than whether the number is reliably moving up or down over time.
Avoid survey fatigue: Reichheld has also cautioned against over-surveying customers, since being bombarded with requests for ratings erodes response quality and rates alike.
The instrumentation and customer feedback discipline behind this exact NPS improvement process, alongside the broader product analytics toolkit, is covered in Product Space AI Product Management Fellowship, where you apply frameworks like this to a real capstone project instead of just reading the theory.
These are illustrative applications of NPS, not confirmed internal processes:
Apple: widely cited as an early adopter of NPS, reportedly using in-store and post-purchase surveys to track loyalty trends across its retail experience.
Zappos: frequently referenced in customer experience literature for building a company culture explicitly oriented around producing promoters through exceptional service.
A B2B SaaS company: segmenting NPS by account size might reveal that enterprise customers score notably lower than small business customers, pointing directly to a support or onboarding gap specific to larger accounts.
A subscription fitness app: noticing a sharp NPS drop immediately after a UI redesign, paired with detractor comments about navigation confusion, could justify rolling back or iterating on the change quickly rather than waiting for a slower churn signal.
Metric | What It Measures | Relationship to NPS |
|---|---|---|
CSAT (Customer Satisfaction Score) | Satisfaction with a specific interaction or feature | More granular and immediate; NPS measures broader, relationship-level loyalty |
CES (Customer Effort Score) | How easy it was to complete a specific task | Focuses on friction, while NPS focuses on overall willingness to recommend |
Churn Rate | The percentage of customers lost over a period | A declining NPS often precedes rising churn, functioning as an early warning sign |
Customer Lifetime Value (CLV) | Total profit expected from a customer relationship | Promoters tend to correlate with higher CLV, though NPS doesn’t measure revenue directly |
NPS works best as one input among several, not a standalone verdict. Product Space guide to key product metrics covers how a loyalty signal like NPS fits alongside retention, engagement, and revenue metrics for a complete picture.
Tool | Best For |
|---|---|
Delighted | Lightweight, purpose-built NPS survey collection |
Qualtrics | Enterprise-grade survey design and analysis |
SurveyMonkey | Flexible survey building, including NPS templates |
Medallia | Large-scale customer experience and NPS program management |
Intercom / Zendesk | In-app and support-triggered NPS surveys tied to real interactions |
Wootric / InMoment | NPS tracking paired with detailed sentiment analysis |
Linking NPS to individual compensation: Reichheld himself has specifically warned this corrupts the data, since employees start chasing scores instead of honest feedback
Treating the number as the whole story: NPS without a follow-up “why” question leaves teams with a score but no clear action to take
Surveying too frequently: over-surveying, what Reichheld has called a “tragedy of the commons,” erodes both response rates and data quality
Ignoring Passives entirely: this group is often the easiest to convert into Promoters with focused attention, yet gets the least
Using one blended, company-wide score: averaging across very different customer segments hides where the real loyalty problems are concentrated
Treating NPS as a perfect predictor on its own: some academic research has challenged whether “likelihood to recommend” is meaningfully better than other loyalty questions, so pairing NPS with other metrics produces a more reliable picture
Aspect | Advantage | Limitation |
|---|---|---|
Simplicity | One question is easy to deploy and track consistently over time | A single number offers no detail on the specific issue behind a score |
Benchmarking | Widely adopted, making cross-industry comparison possible | Benchmarks vary significantly by industry, making blind comparisons misleading |
Actionability | Segments customers into three groups, each suggesting a different response | Doesn’t identify the root cause without a follow-up question |
Predictive value | Reichheld’s original research linked it to repeat purchase and referral behavior | Some academic studies have questioned whether it outperforms other loyalty questions |
Net Promoter Score is based on a single question, how likely a customer is to recommend a company, scored 0 to 10.
It was developed by Fred Reichheld and introduced in his 2003 Harvard Business Review article, “The One Number You Need to Grow”.
The formula is NPS = % Promoters (9-10) minus % Detractors (0-6), with Passives (7-8) excluded from the subtraction.
A score above 50 is generally considered excellent, and above 70 is considered world-class.
Reichheld himself has warned against linking NPS to compensation and over-surveying customers.
NPS works best paired with a follow-up question and other metrics, not as a standalone verdict.
Just like the difference between asking a friend if they enjoyed a meal versus whether they’d stake their reputation recommending the restaurant, Net Promoter Score exists to measure the harder, more meaningful question.
It won’t tell you exactly what to fix on its own, but paired with a follow-up question and real behavioral data, it gives teams a simple, consistent way to track whether customers are becoming genuine advocates or quiet detractors.
If you want to go beyond reading about customer metrics and actually apply frameworks like NPS to a real product with mentor feedback, Product Space AI Product Management Fellowship covers analytics, customer research, and product growth hands-on across a 14-week, project-based curriculum.
How well do you know What Is Net Promoter Score (NPS)?? Let's find out with this quick quiz! (just 10 questions)
Question 1 of 10
A customer loyalty metric based on a single question, how likely a customer is to recommend a company, scored on a 0-to-10 scale.
Fred Reichheld, a Bain & Company Fellow, introduced it in his 2003 Harvard Business Review article, "The One Number You Need to Grow."
NPS equals the percentage of Promoters (scores 9-10) minus the percentage of Detractors (scores 0-6).
Any positive score is generally considered good, above 50 is considered excellent, and above 70 is considered world-class.
Promoters score 9-10 and are loyal advocates, Passives score 7-8 and are satisfied but unenthusiastic, and Detractors score 0-6 and may actively discourage others.
Reichheld himself has warned that tying scores to compensation encourages employees to chase high ratings instead of collecting honest feedback.
A follow-up "Why did you give that score?" question, since the number alone doesn't explain what's driving it.
Linking it to compensation, over-surveying customers, ignoring Passives, and treating one blended score as the whole picture.
Regularly enough to track trends over time, but not so frequently that it causes survey fatigue and erodes response quality.
Delighted, Qualtrics, SurveyMonkey, Medallia, and in-app tools like Intercom or Zendesk are all commonly used.
Learn what change management is for product rollouts, why most failures are people problems not tech problems, the ADKAR model, and mistakes.
Learn what a fake door test is, the real ethical tension behind it, why a click only measures interest, when to use one, and when to avoid it.
Learn what a product health dashboard is, why most fail without anyone noticing, the HEART framework, a gut-check test, and mistakes to avoid.