NPS Survey Questions: The Core Question, Follow-Ups and Examples

The standard NPS question is: “On a scale from 0 to 10, how likely is it that you would recommend [company] to a friend or colleague?” The scale runs from 0 (not at all likely) to 10 (extremely likely). Most businesses add one open follow-up, such as “What’s the main reason for your score?”, because the reason is what tells you what to fix.

That’s the whole core survey. Everything else in this guide is about getting the wording right for your business, choosing a useful follow-up, and connecting the answers to your reviews without breaking platform rules.

The core NPS question, word for word

The wording has stayed consistent since the metric was introduced, and there’s a good reason to keep it that way: consistent wording keeps your scores comparable over time.

On a scale from 0 to 10, how likely is it that you would recommend [company name] to a friend or colleague?

A few details matter more than they seem:

• Use the full 0 to 10 scale. Eleven points, starting at zero. A 1 to 10 scale or a 5-star scale can’t be scored as NPS.
• Label the ends. “Not at all likely” under 0 and “Extremely likely” under 10. Leave the middle unlabeled.
• Don’t color-code the numbers red, amber and green. It signals which answer you want and nudges people’s scores.
• Name the thing you’re asking about. Your business, a specific product, or a location. Be clear which.

If you’re new to the metric itself, our guide on what net promoter score is explains how promoters, passives and detractors are grouped and scored. Once responses come in, our walkthrough on calculating your NPS covers the formula and sample sizes.

Adapting the NPS question to your business

You can swap “friend or colleague” for the audience that fits, as long as you keep the same structure and scale and don’t change it between survey rounds.

Business type	Example wording
Local service (plumber, salon, clinic)	How likely is it that you would recommend [business] to a friend or neighbor?
Restaurant or hospitality	How likely is it that you would recommend [restaurant] to a friend?
Online store	How likely is it that you would recommend shopping at [store] to a friend?
B2B or SaaS	How likely is it that you would recommend [product] to a colleague?
A specific location	How likely is it that you would recommend our [city] location to a friend?
A single product	How likely is it that you would recommend the [product name] to a friend?

For B2B, ask the actual users as well as the person who signed the contract. They often answer very differently, and both views shape what ends up on review sites like G2. Our SaaS reputation management page covers how that plays out for software companies.

The best follow-up questions

Pick one open question, two at most. Every extra question lowers the number of people who finish the survey. These are the follow-ups that tend to produce useful answers.

The reason question (the default)

What’s the main reason for your score?

Neutral, works for every score, and produces comments you can tag by theme. If you only add one question, add this one.

The improvement question

What’s one thing we could do to make your experience better?

More action-focused. Useful when you already know your strengths and want a list of fixes.

Score-specific follow-ups

Many survey tools let you show a different follow-up depending on the score. That’s fine for the private feedback question, as long as the review invitation stays the same for everyone.

• Detractors (0 to 6): “We’re sorry we fell short. What happened?”
• Passives (7 or 8): “What would have made this a 10?”
• Promoters (9 or 10): “What did we do especially well?”

The permission question

May we contact you about your answer? (Yes / No)

This is worth adding, especially for detractors. It gives you permission to follow up personally, which is often how a private complaint gets fixed before it becomes a public review.

Optional extra questions, and when to skip them

Some businesses add one closed question to help sort the comments. Keep it short and only add it if you’ll actually use the answer.

• Touchpoint: “Which part of your visit stood out most?” with options such as booking, staff, wait time, price, result.
• Customer type: “Is this your first visit or have you been with us before?”
• Location or team: useful for multi-location businesses, though you can often fill it in automatically from your records.

Skip demographic questions you don’t need, anything that duplicates data you already hold, and long grids of rating questions. If you want a broader questionnaire, that’s a different survey. Our guide to the customer feedback survey covers longer formats.

A complete NPS survey you can copy

Here’s a short survey suitable for a local service business. It takes most people well under a minute.

1. On a scale from 0 to 10, how likely is it that you would recommend [business] to a friend or neighbor?
2. What’s the main reason for your score?
3. May we contact you about your answer? (Yes / No)

Then, on the thank-you screen, shown to every respondent:

Thanks for your feedback. We read every answer. If you’d like to share your experience publicly, you can leave a review on Google here: [link]

That thank-you line appears whether someone scored you a 3 or a 10. For more on review request wording, see our review request email templates.

When and how to send the survey

• Timing: for transactional NPS, send it shortly after the interaction, while it’s fresh. For a service visit, the same or next day is typical. For a product, wait until the customer has had a chance to use it.
• Channel: email works for most businesses. Text messages suit local services with a customer’s consent. In-app prompts suit software.
• Frequency: don’t survey the same person after every small interaction. Set a minimum gap between surveys to the same customer.
• Who gets it: everyone in the group you’re measuring, not only customers staff think were happy.

A worked example

This is an illustrative scenario, not a real client.

A small online store selling kitchenware adds the NPS question to its post-delivery email, sent a week after the order arrives. It uses the reason question as the follow-up and adds a permission question.

After a month, the comments from passives keep repeating one phrase: “fine, but packaging was excessive.” Detractor comments mostly mention a single supplier’s mugs arriving chipped. Neither theme had shown up clearly in the store’s product reviews yet, though two recent three-star reviews hinted at breakage.

The owner switches the mug supplier’s packaging, contacts the detractors who gave permission to offer replacements, and replies to the two product reviews explaining the change. Nobody is asked to edit a review. The follow-up question found the problem before it became a pattern in public.

Common mistakes with NPS questions

• Changing the wording between rounds. Even small changes make scores harder to compare.
• Leading questions. “How likely are you to recommend our award-winning team?” pushes scores up and makes the data less useful.
• Asking staff to coach customers. Telling customers “only a 10 counts” distorts the results and can irritate them.
• Too many questions. A long survey loses respondents and mostly hears from people with strong opinions.
• Gating the review link. The most costly mistake, because it risks platform action and hides the problems you most need to see.
• Collecting comments nobody reads. If nobody tags and acts on the answers, the survey is just a cost to your customers’ time.

Turning the answers into better reviews

NPS answers are private. Your reviews are public. The link between them is action: fix what detractors describe, tell people what changed, and keep asking everyone for honest reviews. Over time, the fixes show up in what customers write. Our guide on building a customer feedback loop turns this into a monthly routine, and our review management service can run the review side for you.

Frequently asked questions

What is the standard NPS question?

“On a scale from 0 to 10, how likely is it that you would recommend [company] to a friend or colleague?” The ends are labeled “not at all likely” and “extremely likely.” Keep this wording consistent so your scores stay comparable.

Can I change 'friend or colleague' in the NPS question?

Yes, many businesses adapt it to “friend or neighbor” or “colleague” to fit their audience. Keep the 0 to 10 scale and the same wording from one survey round to the next.

How many questions should an NPS survey have?

Usually two or three: the NPS question, one open follow-up about the reason for the score, and optionally a question asking permission to contact the customer. More questions tend to reduce completion.

Can I show different follow-up questions based on the score?

Yes, tailoring the private feedback question to the score is common. Just don’t tailor the review invitation. Everyone should see the same option to leave a public review.