What Is Net Promoter Score (NPS)? Definition & Calculation Guide

A good NPS score depends heavily on industry, but as a general guide, any positive score means more promoters than detractors, above 30 is considered good, and above 70 is considered excellent, with software and ecommerce benchmarks varying widely by category and by how mature the specific market is.

NPS benchmarks by industry

Industries with high switching costs or strong brand loyalty, like some SaaS categories, tend to post higher average NPS than industries like telecom or ISPs, where customer frustration is common and switching is often difficult, not because of love for the provider. Comparing a company's NPS only within its own industry, rather than against a generic global benchmark, gives a far more honest read on whether the number is actually good or simply typical for a hard category.

Why NPS should be read as a trend, not a single number

A single NPS reading says less than the trend line; a steadily rising or falling NPS over several survey cycles reveals far more about whether the business is actually improving the customer relationship than any one score in isolation. A single quarter's dip can be noise, a seasonal effect, a rough product release, a support staffing gap, while a consistent multi-quarter decline is a signal worth investigating seriously.

An effective NPS survey is sent at a meaningful moment in the customer relationship, delivered through a channel the customer already uses, and always paired with a follow-up question asking why, since the score alone doesn't explain what to fix or where the biggest opportunity to improve actually sits.

When to send the survey

Common trigger points include a set period after onboarding, after a support interaction, or on a recurring cadence like quarterly, with the right timing depending on how often the relationship naturally changes. Sending an NPS survey too early, before a customer has had enough time to form a real opinion, tends to produce artificially neutral or overly optimistic scores that don't hold up once the honeymoon period ends.

Where to ask it (email, in-app, chat)

Email is the most common NPS delivery channel, though in-app prompts and post-chat surveys tend to get higher response rates since they're delivered in the moment rather than requiring a customer to return later. In-app surveys also tend to reach customers while their most recent experience with the product is still fresh, which generally produces more accurate, specific responses than a delayed email arriving days later.

The follow-up question that makes NPS actionable

The open-ended "why did you give that score" follow-up is what turns NPS from a number into actionable feedback, and skipping it is one of the most common reasons NPS programs fail to drive any real change. Even a low response rate on the follow-up question is worth having, since a handful of specific, detailed detractor comments usually points to the same one or two root causes far more precisely than the score alone ever could.

The most valuable NPS programs close the loop with detractors directly, turn promoters into referrals or reviews, and route the qualitative feedback back to product and support teams instead of letting the score sit untouched in a dashboard nobody revisits after the survey closes.

Closing the loop with detractors

Reaching out personally to detractors, especially ones who left specific feedback, can recover a relationship that would otherwise quietly churn, and often surfaces a fixable problem the business didn't know it had. A short, genuine follow-up message, acknowledging the specific issue raised rather than a generic apology, frequently does more to repair a relationship than any product fix delivered weeks or months later.

Turning promoters into referrals and reviews

Promoters are the easiest customers to ask for a referral, review, or case study, since they've already told the business they'd recommend it, making the ask a natural next step rather than a cold request. Timing the ask immediately after a high NPS response, while the positive sentiment is fresh, tends to produce a meaningfully higher conversion rate than a generic, untargeted request for reviews sent to the entire customer base.

Feeding feedback back to product and support teams

Feeding the open-ended comments back to product and support teams turns NPS into an ongoing source of prioritization input instead of a one-off survey exercise. Categorizing detractor comments by theme, a specific feature gap, a pricing complaint, a support frustration, often reveals that a small number of recurring issues are responsible for the majority of the low scores, giving teams a clear, evidence-backed place to focus.

The most common NPS mistakes are treating the score as a vanity metric with no follow-up action, surveying customers too frequently, and reading the number without ever looking at the qualitative comments that actually explain what's driving it up or down.

Treating NPS as a vanity metric with no follow-up

A rising or falling NPS with no corresponding action plan is just a number on a slide; the value comes entirely from what a business does in response to it. Teams that report NPS quarterly without ever changing a process, a feature, or a support practice based on what it reveals tend to see the metric plateau indefinitely, since nothing about the underlying customer experience is actually being addressed.

Surveying too often

Surveying the same customer too frequently causes fatigue and declining response rates, which quietly skews the data toward only the most engaged or most frustrated customers, the people with enough motivation left to keep responding. A customer surveyed every month is far less likely to answer thoughtfully, or at all, by the fourth or fifth request, compared to one surveyed on a sensible quarterly or milestone-based cadence.

Ignoring the "why" behind the score

The score identifies who's happy or unhappy; the open-ended comments explain why, and skipping that half of the data leaves a business guessing at the actual fix. A team that only tracks the numeric trend can tell that satisfaction is falling, but has no way to know whether the cause is pricing, a recent product change, or a support quality issue without reading what customers actually wrote.

NPS is most useful alongside, not instead of, Customer Satisfaction Score (CSAT), which measures a specific interaction, and Customer Effort Score (CES), which measures how easy a specific process was to complete, since each captures a different layer of the customer relationship.

Customer Satisfaction Score (CSAT)

CSAT is typically asked right after a specific interaction, a support chat or a purchase, making it more immediate and interaction-specific than NPS, which measures the broader relationship. A business can have strong CSAT scores on individual support chats while still posting a mediocre NPS overall, which usually points to a problem with the core product or pricing rather than with the support experience itself.

Customer Effort Score (CES)

CES asks how much effort a customer had to expend to get something done, which tends to predict churn even more directly than NPS in support-heavy businesses, since high-effort experiences are a common reason customers leave. Where NPS asks about the relationship as a whole, CES zeroes in on friction, making it especially useful for diagnosing why a specific process, like onboarding or a support escalation, is quietly pushing customers toward the door.

NPS data serves a different purpose depending on which team is looking at it: product teams mine the comments for feature gaps, support teams use detractor scores to flag accounts needing outreach, marketing teams use promoters for advocacy campaigns, and leadership tracks the trend as a proxy for overall customer health.

Product teams

Product teams typically comb through detractor and passive comments for recurring feature requests or usability complaints, treating NPS feedback as an ongoing, low-cost source of prioritization input alongside more formal research methods.

Support teams

Support teams often use a low individual NPS response as a trigger for proactive outreach, especially for accounts that also show other risk signals like declining usage or a recent support escalation.

Marketing teams

Marketing teams frequently mine promoter responses for testimonial and case study material, since a customer who's already said they'd recommend the business is typically receptive to being asked for a public quote or review.

Executive and leadership teams

Leadership teams generally track the NPS trend line as one input into overall company health reporting, watching for meaningful shifts that might warrant deeper investigation rather than reacting to normal quarter-to-quarter fluctuation.

Teams generally implement NPS through one of three approaches: a built-in survey feature inside an existing support or product platform, a standalone dedicated NPS tool, or a simple manual spreadsheet-based process, with the right choice depending mostly on volume and how deeply the results need to integrate with other systems.

Built-in survey tools in support platforms

Many live chat and customer support platforms include a native NPS survey feature that can trigger automatically after a support interaction, which is often the simplest starting point since it requires no additional integration work.

Standalone NPS survey tools

Dedicated NPS and feedback platforms typically offer more advanced segmentation, reporting, and integration options than a built-in feature, which becomes worthwhile once a business needs to slice results by multiple customer attributes or feed the data into several other systems.

DIY spreadsheet approach

A manual approach, sending a survey link and tracking responses in a spreadsheet, can work reasonably well for a small business with low response volume, though it quickly becomes unwieldy as the number of responses and the need for trend tracking grows.

Choosing the right approach for company size

Smaller teams generally do fine starting with whatever built-in survey capability their existing platform already offers, upgrading to a dedicated tool only once volume or reporting needs genuinely outgrow that simpler starting point.