NPS Survey Questions: the core question, follow-ups & 20 examples you can copy
NPS survey questions: the core question, follow-ups & 20 examples you can copy
The NPS question is the one survey question everyone has seen — “how likely are you to recommend us?” — and the one most teams stop at. That’s a mistake. The score tells you where loyalty stands; the follow-up tells you why. And the exact wording you wrap around that 0–10 scale changes what you learn.
This post covers the standard NPS question, the follow-up that actually matters, and 20 copy-paste variants grouped by situation — relational, transactional, B2B, post-purchase, and more. If you need the formula and how to read the score, see [our NPS vs CSAT vs CES guide][PILLAR: customer-feedback-metrics-nps-csat-ces].
The standard NPS question
The Net Promoter Score survey has one scored question, and the wording is essentially fixed:
“How likely are you to recommend [company/product] to a friend or colleague?” Answered on a 0–10 scale, where 0 is “Not at all likely” and 10 is “Extremely likely.”
Responses split into three groups:
• Promoters (9–10) — loyal, likely to refer others.
• Passives (7–8) — satisfied, but will switch for a better offer.
• Detractors (0–6) — unhappy, and likely to tell people.
NPS = % Promoters − % Detractors. The result is a number between −100 and +100 (not a percentage). Use the NPS calculator to run it, and see [what counts as a good NPS score][SPOKE: what-is-a-good-nps-score] to read it.
Don’t rewrite the core question. The 0–10 scale and the “recommend to a friend or colleague” frame are what make the number comparable across time. If you change “friend or colleague” to “anyone” or swap the scale to 1–5, you’re measuring something different — and your trend line breaks.
What you can change is the subject: “recommend us” vs “recommend our product” vs “recommend our support team”. That choice depends on whether you’re running a relational or transactional survey (more on that below).
The follow-up question (this is where the value is)
The NPS score tells you the temperature. The follow-up question tells you what’s causing the fever — or the health. The standard follow-up is:
“What is the primary reason for your score?” — open text.
This single open-ended question is the most valuable line in the entire survey, because it gives you verbatim language from customers about what’s working and what isn’t. When you read 50 detractor responses and 30 of them mention “delivery time,” you’ve found your problem without running a focus group.
Tips for the follow-up:
• Keep it open-ended. A multiple-choice follow-up limits answers to the options you already thought of — and you’re surveying because you don’t know the answer.
• Ask it for every group, not just detractors. Promoter follow-ups tell you what to protect; passive follow-ups tell you what would push them over the line.
• One follow-up is enough. The NPS survey should take under 60 seconds. Every question you add past two cuts your response rate — [more on improving response rates][SPOKE: how-to-improve-survey-response-rate].
Some teams add a second follow-up tailored to the score. This works well with conditional logic:
•For promoters (9–10): “What do you value most about [company]?”
•For passives (7–8): “What would make you rate us higher?”
•For detractors (0–6): “What disappointed you the most?”
SurveyKar supports this with conditional logic on the Basic plan and above — the survey shows a different follow-up depending on the score the respondent just gave.
Relational vs transactional NPS — different timing, different wording
There are two ways to run NPS, and the difference changes when you send the survey and how you word the question.
Relational NPS (rNPS)
Sent on a fixed schedule — quarterly, every six months, or annually — to a broad sample of customers. It measures the overall relationship: “taking everything into account, would you recommend us?”
When to use it: - Board-level loyalty tracking - Comparing this quarter to last quarter - Identifying long-term sentiment shifts
Typical wording: > “Based on your overall experience, how likely are you to recommend [company] to a friend or colleague?”
Transactional NPS (tNPS)
Triggered by a specific event — a closed support ticket, a delivered order, a completed onboarding, a subscription renewal. It measures a single interaction: “after this experience, would you recommend us?”
When to use it: - Measuring the impact of a specific touchpoint - Catching friction right after it happens - Comparing support quality, checkout experience, or onboarding across teams
Typical wording: > “Based on your recent [support experience / purchase / onboarding], how likely are you to recommend [company] to a friend or colleague?”
The rule: keep the scores separate. A transactional score after a bad support call will be lower than a relational score from the same person. That’s not a contradiction — they’re answering different questions. Mix them and you get noise.
Your NPS calculator page breaks down both approaches with worked examples.
20 NPS question examples by situation
Copy these and adapt the bracketed parts. Each is paired with a suggested follow-up.
General / relational
[table]
Post-purchase / e-commerce
[table]
Support / service
[table]
SaaS / product
[table]
B2B / professional services
[table]
Onboarding
[table]
Renewal / churn risk
[table]
In-store / walk-in (India context)
[table]
India tip: For walk-in businesses — restaurants, salons, clinics, coaching centres — send NPS on WhatsApp within an hour of the visit, or put a QR code on the bill. The recommendation question works better fresh. SurveyKar’s WhatsApp surveys and QR walk-in features are built for this.
Mistakes that kill your NPS data
1. Rewriting the scale. A 1–5 or 1–7 scale isn’t NPS — it’s a satisfaction rating. NPS requires 0–10.
2. Asking two questions in one. “How likely are you to recommend our product and our support?” — the respondent doesn’t know which to answer, so neither score means anything. Pick one subject per survey.
3. Leading the respondent. “Given how much you love our product, how likely are you to recommend us?” primes a high score and wastes the data.
4. Surveying too often. Quarterly relational NPS is plenty for most businesses. Monthly NPS fatigues your audience and tanks response rates. Transactional NPS is fine on every event — it’s tied to a moment, not a calendar.
5. Ignoring the open text. If you only report the score and never read the follow-ups, you’re tracking a number that tells you nothing you can act on. The follow-up answers are your action list.
6. Mixing relational and transactional scores. They measure different things. Report them separately, or your trend line is noise.
How to run an NPS survey on SurveyKar
1. Start from a template. Use the NPS survey template for general NPS, or the SaaS product NPS template for product-specific feedback.
2. Add conditional follow-ups (Basic plan and above). Show different follow-up questions to promoters, passives and detractors — the examples in the section above work directly.
3. Send on WhatsApp or QR. In India, WhatsApp gets 3–5× the response rate of email for most SMBs. The WhatsApp surveys integration sends the survey link in a message; the QR walk-in feature puts it on a code at the counter.
4. Read the score live. SurveyKar scores NPS automatically as responses come in — no spreadsheet needed. The dashboard splits promoters, passives and detractors and shows the open-text answers alongside the score.
Frequently Asked Questions related to NPS
What is the standard NPS survey question?
“How likely are you to recommend [company/product] to a friend or colleague?” on a 0–10 scale. This exact framing — with the 0–10 scale and the “recommend” verb — is what makes NPS comparable across surveys and over time.
How many questions should an NPS survey have?
Two: the scored 0–10 question and one open-ended follow-up (“What is the primary reason for your score?”). You can add a third conditional follow-up for each group (promoter, passive, detractor), but stop there. Every question past three cuts your completion rate.
What’s the difference between relational and transactional NPS?
Relational NPS is sent on a schedule (quarterly, biannually) to measure overall loyalty. Transactional NPS is triggered by a specific event (a support ticket, a purchase, onboarding) to measure that moment. Keep the scores separate — they answer different questions.
Can I change the NPS scale from 0–10?
No — the three groups (promoters, passives, detractors) and the formula depend on the 0–10 scale. If you use a 1–5 scale, the cutoffs don’t apply and the number you get is not NPS. Use a [CSAT survey][SPOKE: csat-survey-questions] if you want a 1–5 or star-rating format.
Should I ask NPS on WhatsApp?
In India, yes — WhatsApp surveys consistently get higher response rates than email for mobile-first audiences. Send the link in a WhatsApp message within an hour of the interaction you’re measuring.
Try it
Templates - NPS survey template - SaaS product NPS template - Customer satisfaction survey template
Calculators - NPS calculator - CSAT calculator
Keep reading
All articlesWhat Is a Good NPS Score? How to Read Yours (Without Fake Benchmarks
The short answer is a set of general bands that the industry uses as a rough guide. The honest answer is that “good” depends almost entirely on your own previous score, not on a table of industry averages you found online.
Customer Feedback Metrics: NPS vs CSAT vs CES
How to Create a QR Code Registration Form (Step by Step)
Replace the paper register at your entrance: build a registration form, put its QR code on a poster, and check visitors in at the counter.