NPS measures sentiment and willingness to advocate. It is not a substitute for a customer health score and should never be read as a renewal prediction on its own.
Net Promoter Score (NPS) is a customer advocacy metric built on one question: "How likely are you to recommend this product?" scored 0 to 10. Respondents scoring 9 or 10 are Promoters, 7 or 8 are Passives, and 0 through 6 are Detractors. NPS equals the percentage of Promoters minus the percentage of Detractors, producing a single number on a -100 to +100 scale. In B2B SaaS, 30 or above is considered good and 50 or above is considered excellent.
The six things to understand about NPS before you ship a survey, benchmark your result, or quote the number on a board slide. Each one is a place real operators either overstate the signal or ignore the structural flaws that make the number easy to misread.
NPS asks a single question: "How likely are you to recommend this product or company to a friend or colleague?" Respondents answer on a 0 to 10 scale. The simplicity is intentional. One standardized question allows comparison across time, segment, and competitor without the ambiguity of a multi-question survey.
Scores of 9 or 10 are Promoters, the enthusiastic advocates likely to refer others. Scores of 7 or 8 are Passives, satisfied but not loyal. Scores of 0 through 6 are Detractors, unhappy customers who may actively discourage others. The three buckets are the entire mechanism behind the score.
NPS equals the percentage of respondents who are Promoters minus the percentage who are Detractors. Passives are excluded from the calculation. If 50% are Promoters and 20% are Detractors, NPS is 30. The score ranges from -100, when every respondent is a Detractor, to +100, when every respondent is a Promoter.
In B2B SaaS, an NPS of 30 or above is widely treated as good, 50 or above is excellent, and above 70 is world class. Negative scores signal a product or experience problem that compounds across the base. Benchmarks shift by segment and vertical, so comparing a horizontal SMB tool to a vertical enterprise platform without context misleads.
NPS measures how a customer felt about advocating at the moment they responded. It is not a composite customer health score. Health scores combine usage, support load, invoice status, stakeholder coverage, and expansion signals into a renewal probability. A happy Detractor who still uses the product daily will still renew. NPS alone does not catch that.
Relational NPS is sent on a steady cadence (quarterly or annually) to the whole base to track trend. Transactional NPS is triggered after a specific event: onboarding completion, support case closure, a renewal. The two produce different numbers for the same base. Mixing them in one trend line is the single most common NPS reporting mistake.
NPS looks deceptively simple because the output is one number, but the mechanics behind that number carry structural decisions that materially change what the score represents. The six cards below walk through the question, the buckets, the math, the scale, the follow-up, and the reporting cadence that together constitute a defensible NPS program.
The canonical NPS question is "How likely are you to recommend [product or company] to a friend or colleague?" The wording is kept standard across companies and years so the resulting scores remain comparable. Rewording the question (asking about a feature, a department, a specific moment) breaks that comparability and should be labelled distinctly.
Respondents score 0 to 10, with 0 labelled "not at all likely" and 10 labelled "extremely likely." The 11-point scale is wider than a 5-point Likert so the three buckets fall out cleanly. The scale also captures the common survey behavior where enthusiasm clusters at the extremes.
Only scores of exactly 9 or 10 are Promoters. An 8 feels close to a 9 but does not count. The threshold is deliberately high because the metric is designed to measure genuine advocacy, not polite satisfaction. Detractors (0 through 6) are drawn broadly for the same reason: silent dissatisfaction harms the business as much as vocal dissatisfaction.
NPS is not an average of the raw scores. It is the percentage of Promoters minus the percentage of Detractors. Passives (7 and 8) are counted in the denominator but not in the numerator. This structural choice is why the score can be negative and why a modest shift in the mix of scores produces a large shift in the headline number.
Every mature NPS program follows the score question with an open-text prompt: "What is the main reason for your score?" The verbatim comments carry the actual insight. The score tells you the ratio. The comments tell you what to fix. A program that captures only the number without the comment is throwing away the most valuable half of the response.
A single NPS reading in a quarter is noisy. The signal lives in the trend over multiple quarters and in the breakdown by segment, product line, and tenure. Boards look at the direction and the slope. Teams look at the segment that moved. Reporting a single aggregate number without the trend or the slice invites the question "compared to what?"
The NPS score is a sentiment and advocacy indicator, nothing more and nothing less. The six cards below describe the honest interpretation: what the number signals when it moves, what it conceals when it holds steady, and the specific operating decisions it should and should not drive. Mature revenue teams pair NPS with retention metrics and health scores for the same reason.
A rising NPS means more of the responding base is willing to recommend the product right now. That matters for referral programs, word-of-mouth growth, case-study supply, and community engagement. It is a leading indicator for inbound demand that originates from the installed base.
A customer can be a Detractor on NPS and still renew because switching costs, embedded workflow, or a signed multi-year contract keeps them. A customer can be a Promoter and still churn because a budget cut or an acquisition closed the account. NPS alone has never been a reliable churn predictor. Pairing it with a usage-based health score closes the gap.
Low response rates (common in B2B at under 30%) mean the respondents are not representative. Happy and unhappy customers both over-respond; satisfied-but-indifferent customers under-respond. A score computed on a 10% sample of enterprise accounts tells you about the 10% who answered, not the entire base. Weighting by segment and reporting response rate alongside the score prevents the misread.
An aggregate NPS of 32 can be the average of enterprise at 58 and SMB at 6, which is a very different business than enterprise at 36 and SMB at 28. Every mature program publishes NPS sliced by segment, by product, by tenure cohort, and by region. The aggregate is only safe to quote once the slices have been checked for a hidden break.
The operating playbook is simple. Call every Detractor within 48 hours of the response to understand the specific issue. Group the verbatim comments from all three buckets into themes to find the recurring product and experience problems. Follow up with Promoters to source case studies, referrals, and reviews. The score moves because the follow-up actions moved it, not because the score was reported.
Industry NPS benchmarks drift and are not always comparable because segment, response rate, and sample bias differ. The most honest benchmark is the companys own prior quarter. A score of 25 that moved from 18 is a stronger signal than a score of 35 that moved from 48. The direction is more actionable than the absolute number for most operating decisions.
NPS is only useful when the score, the verbatim comment, and the operating follow-up all land on the same account record where the renewal, the expansion pipeline, and the health signals already live. The six cards below describe the pattern used by B2B SaaS teams that treat NPS as part of the account view, not as a report that lives in a separate survey tool nobody opens.
Every NPS response is written back to the responding contact and the parent account in the CRM. The score, the bucket, the verbatim comment, and the response date are visible on the account view next to the ARR, the renewal date, and the open support cases. The CSM reads one record, not three.
A Detractor response triggers a workflow: a task on the account owner to call within 48 hours, a note on the next renewal play, a flag on the account health score. A Promoter response triggers a different workflow: a prompt to request a review, a referral, or a case study. The score is converted into motion the same day it lands.
The customer health score aggregates product usage, support load, invoice status, stakeholder coverage, and NPS into a single renewal-risk indicator. NPS is one weighted input. A Detractor on an account with falling usage and an open escalation is a different story than a Detractor on an account with climbing usage and a champion in place.
Because every response is tied to an account record, NPS can be rolled up by segment, industry, plan tier, product line, tenure cohort, CSM, and region. The segment that is quietly dragging the aggregate score down becomes a one-click report instead of a spreadsheet exercise across the survey tool and the CRM.
Every NPS summary in Strkr shows the response rate alongside the score. A 42 computed from a 60% response rate is a different number than a 42 computed from a 12% response rate. Making the base visible prevents the headline from being quoted without the context that would qualify it in diligence or a board review.
Each response is a time-stamped record on the contact and the account. When the quarterly score moves, the exact responses that drove the movement are inspectable. "Why did NPS drop six points this quarter" becomes a filter on the response log, not a reconstruction from a monthly PDF export.
Strkr captures NPS responses, writes them back to the account, triggers follow-up tasks on Detractors and Promoters, and rolls the score into the customer health view next to usage, support, and renewal signals. The score drives motion the same day it lands instead of sitting in a survey tool nobody opens.
The widely referenced guideposts for B2B SaaS are 30 or above is good, 50 or above is excellent, and 70 or above is world class. Scores below zero signal a product or experience problem affecting a meaningful share of the base. Benchmarks vary by vertical, segment, and company stage, so comparing a horizontal SMB tool to a vertical enterprise platform without that context will mislead.
NPS equals the percentage of respondents who are Promoters (scored 9 or 10) minus the percentage who are Detractors (scored 0 through 6). Passives (scored 7 or 8) are counted in the denominator but excluded from the numerator. If 50% of respondents are Promoters, 30% are Passives, and 20% are Detractors, the NPS is 30. The result ranges from -100 to +100.
Promoters scored 9 or 10 and are the enthusiastic advocates most likely to refer. Passives scored 7 or 8 and are satisfied but not loyal, vulnerable to a competitor. Detractors scored 0 through 6 and are unhappy enough to potentially warn others away from the product. The three buckets are deliberately asymmetric: 9 and 10 only on the promoter side, zero through 6 broadly on the detractor side.
No. NPS measures advocacy sentiment at a single point in time from the subset of customers who responded to a survey. A customer health score combines product usage, support case load, invoice status, stakeholder coverage, expansion signals, and often NPS itself into a composite renewal-risk indicator. NPS is one input to a health score, not a substitute for it. A Detractor can still renew; a Promoter can still churn.
CSAT (Customer Satisfaction) asks how satisfied a customer was with a specific interaction, typically on a 1 to 5 scale, and is usually transactional. NPS asks about overall willingness to recommend the product or company, on a 0 to 10 scale, and is usually relational. CSAT answers "did this interaction go well." NPS answers "how do you feel about the relationship." Many SaaS teams run both because they capture different signals.
Relational NPS is typically run quarterly or semi-annually across the whole active base to track trend. Transactional NPS is triggered after a specific event like onboarding completion, support case closure, or renewal. The two produce different scores from the same base and should be reported separately. Mixing relational and transactional responses in one trend line is the most common NPS reporting mistake.
A negative NPS means more Detractors than Promoters, which signals a product or experience problem affecting a meaningful share of responding customers. In B2B SaaS, scores from 0 to 20 are weak, 20 to 30 are mediocre, and anything negative is a signal to pause expansion spend and read the verbatim comments before investing further in acquisition. A low score with a rising trend is a different story than a low score that is falling further.
Yes, and it happens often. Surveying only happy customers, excluding Detractors from the sample, timing the send right after a positive moment, or asking leading questions all inflate the score. Mature programs defend against gaming by surveying the full active base, publishing the response rate alongside the score, slicing by segment, and following every reading with a verbatim comment review. A clean NPS program is a disciplined one, not an optimized one.
No sales call, no migration consultant, no four-month implementation. Enter your card, get 14 days of the full Pro tier, cancel any time before day 14 with zero charge. Spin up a workspace, import your CSV, and have something useful before lunch.
