Guest Experience and Hospitality

Measuring Guest Loyalty With NPS

How restaurants calculate Net Promoter Score, how many responses they need, how NPS compares with simpler scores and which trends signal real change.

AssistCall research teamPublished Reviewed by Faruque Amin, VP of Sales and Marketing7 min read

On this page
  1. What NPS measures and why restaurants use it
  2. How to calculate it, step by step
  3. NPS compared with simpler scores
  4. Sample size: when the number means something
  5. Loyalty in restaurants is about frequency
  6. Segment before you conclude anything
  7. What to watch over time
  8. Acting on detractors and promoters
  9. Pitfalls to avoid
  10. Frequently asked questions
  11. References
Guest Experience and Hospitality: Measuring Guest Loyalty With NPSArticle coverGUEST EXPERIENCE AND HOSPITALITYMeasuring Guest LoyaltyWith NPSAva for RestaurantsInsightsGuest Experience and Hospitality: Measuring Guest Loyalty With NPSArticle coverGUEST EXPERIENCE AND HOSPITALITYMeasuring GuestLoyalty With NPSAva for Restaurants

Key takeaways

  • NPS asks how likely a guest is to recommend you on a 0 to 10 scale and subtracts the share of detractors (0 to 6) from the share of promoters (9 and 10)[1].
  • Below about 100 responses a month, one or two guests swing the score; read the trend over a rolling quarter, not week to week.
  • Pair NPS with a return-visit question and real visit data, because loyalty in restaurants shows up as frequency.
  • Segment by daypart, channel and new versus returning guests; the overall number hides the story.

What NPS measures and why restaurants use it

Net Promoter Score is a single-question measure of loyalty. The guest answers 'How likely are you to recommend us to a friend or colleague?' on a zero-to-ten scale. Those who answer 9 or 10 are promoters, 7 or 8 are passives, and 0 to 6 are detractors. The score is the percentage of promoters minus the percentage of detractors; passives are left out of the formula[1]. The result can run from minus 100 to plus 100.

For restaurants, NPS is attractive because recommendation is how many guests find a place, and because the same question can be asked after dine-in, takeout and events. It is not a measure of how good the food was on one night. It is a measure of whether the guest would put their own reputation on the line by sending a friend to you.

How to calculate it, step by step

  1. Ask the question the same way every time

    Use the standard wording and a 0 to 10 scale. Changing the wording or the scale breaks comparisons over time.

  2. Add one follow-up

    'What is the main reason for your score?' as free text. The number tells you where you stand; the reason tells you why.

  3. Count each group

    Sort responses into promoters (9 and 10), passives (7 and 8) and detractors (0 to 6).

  4. Convert to percentages

    Divide each group by total responses, including passives.

  5. Subtract

    Promoter percentage minus detractor percentage. Report it as a whole number, not a percent.

  6. Record the response count

    Always show how many responses sit behind the score. A score from 40 guests and one from 400 are not equally reliable.

Example (fictional): in October a bistro receives 200 responses. 110 are promoters (55 percent), 50 are passives and 40 are detractors (20 percent). NPS is 55 minus 20, which equals 35.

NPS compared with simpler scores

Common guest loyalty and satisfaction measures.
MeasureQuestionGood forWeak at
NPSHow likely are you to recommend us? (0 to 10)Tracking loyalty over months; comparing locationsExplaining a single bad night; small samples
Satisfaction scoreHow satisfied were you with today's visit? (1 to 5)Spotting problems on a given shiftPredicting whether guests come back
Return intentHow likely are you to come back in the next month?Linking feedback to visit frequencyGuests are optimistic about their own plans
Public review ratingStars on Google or YelpWhat prospective guests seeSkewed toward very happy and very unhappy guests
Visit frequencyMeasured, not asked: visits per guest from POS, reservation or loyalty dataThe real outcome of loyaltyOnly works where you can identify guests

A small restaurant does not need all of these. A practical set is a satisfaction score for daily operations, NPS reviewed monthly or quarterly, and visit frequency wherever your reservation or loyalty data allows it. Our guest feedback survey guide shows how to fit the questions into one short survey.

Sample size: when the number means something

NPS moves in big jumps when the response count is small. With 50 responses, each guest is 2 percentage points of the total, so one promoter turning into a detractor moves the score by 4 points. With 500 responses, the same change moves it by less than one point.

  • Under 50 responses in a period: read the comments, not the score.
  • Between 50 and 150: report a rolling three-month score and treat changes under ten points as noise.
  • Above 150: monthly scores are usable, and changes of five points or more are worth a look.
  • Always compare like with like: the same channel, the same question placement and the same timing.

Do not set server bonuses on NPS. Individual sample sizes are tiny, and the score depends on the kitchen, the host and the weather as much as on the server.

Loyalty in restaurants is about frequency

A promoter who visits twice a year is worth less than a passive who comes in every week. That is why NPS should sit beside visit data. Loyalty programs are one source of it: 52 percent of consumers participate in a restaurant loyalty or rewards program, and 96 percent of those participants appreciate them[2].

Rewards programs influence choices differently by segment. Being in a program influences dining choices for 61 percent of delivery customers, 54 percent of quick-service patrons and 41 percent of table-service diners[3]. Off-premises guests are even more program-driven: 65 percent of drive-thru users and over 60 percent of takeout and delivery users say membership affects where they order[4]. Full-service restaurants rely more on experience and recognition, which is exactly what NPS comments tend to describe.

Share of guests whose dining choices are influenced by being in a rewards programNational Restaurant Association, 2025
Share of guests whose dining choices are influenced by being in a rewards programColumn chart. Delivery customers: 61%; Quick-service patrons: 54%; Table-service diners: 41%.0%20%40%60%80%Delivery customers: 61%61%Delivery customersQuick-service patrons: 54%54%Quick-service patronsTable-service diners: 41%41%Table-service dinersShare of guests whose dining choices are influenced by being in a rewards programColumn chart. Delivery customers: 61%; Quick-service patrons: 54%; Table-service diners: 41%.0%20%40%60%80%Delivery customers: 61%61%DeliverycustomersQuick-service patrons: 54%54%Quick-serv-ice patronsTable-service diners: 41%41%Table-serv-ice diners

Source: National Restaurant Association, 2025[3]

Show the data
ItemValue
Delivery customers61%
Quick-service patrons54%
Table-service diners41%

Segment before you conclude anything

  • By channel: dine-in, takeout, delivery and events usually score very differently. A falling overall NPS may just mean takeout grew as a share of responses.
  • By daypart: brunch and dinner often have different crews, menus and guest expectations.
  • New versus returning guests: first-timers who are detractors point to expectation gaps, often in price, portion or wait. Regulars who turn into detractors point to something that changed.
  • By location: compare locations on the same question, timing and channel only.
  • By comment theme: tag the reason text and calculate NPS for each theme to see which issues drag the score most.

What to watch over time

  1. The rolling quarterly score and its response count.
  2. The detractor share on its own: reducing detractors is usually faster than creating promoters.
  3. The top three detractor reasons and whether they change after you fix something.
  4. Promoter comments that name staff or dishes, which tell you what to protect when you change the menu or the team.
  5. Visit frequency among respondents, where you can match it, to check that promoters really do come back more often.

Example (fictional): a cafe's quarterly NPS falls from 42 to 31 over the summer. Segmenting shows dine-in is stable but mobile-order pickup has dropped sharply, with comments about orders not being ready on time. The fix is a pickup shelf and staggered ready times, not a hospitality retraining for the whole team.

Acting on detractors and promoters

A score that nobody acts on is just a number in a slide. Build two simple loops, one for each end of the scale.

  1. Detractor loop: any score of 0 to 6 with contact permission goes to the manager on duty the next day. The manager calls, listens, applies the normal recovery policy and logs the reason. The aim is to understand, not to argue the guest up the scale.
  2. Detractor review: once a month, read every detractor comment in one sitting and group them. The top group becomes next month's fix, with an owner and a measure.
  3. Promoter loop: promoters have told you they would recommend you. Thank them, and include them in the same honest, general review request every guest receives. Never ask only promoters for public reviews; that is gating.
  4. Passive loop: passives often mention one small irritant. Their comments are the cheapest improvements on the list, such as a slow check, a cold room or a missing menu item.
  5. Team loop: share the monthly score, the main reasons and any staff mentioned by name at the next all-hands or pre-shift.

Keep the loops small enough to run every month. A general manager who calls three detractors a week learns more about the restaurant than any dashboard can show.

Pitfalls to avoid

  • Asking only happy guests, which inflates the score and makes it useless for decisions.
  • Changing the question, the scale or the timing and comparing the new score with the old.
  • Chasing an industry benchmark you found online. Published benchmarks rarely match your segment, survey timing or channel mix. Your own trend is the benchmark.
  • Treating passives as fine. They are satisfied but open to switching when a new place opens nearby.
  • Reporting the score without the comments. The number starts the conversation; the reasons tell you what to do.
  • Sending the NPS question after a complaint the manager already resolved, without noting it, which distorts the detractor count.
  • Surveying the same regulars every month, so a small group of loyal guests dominates the score and new guests barely register.

Frequently asked questions

How is restaurant NPS calculated?

Ask guests how likely they are to recommend you on a 0 to 10 scale. Subtract the percentage who answer 0 to 6 from the percentage who answer 9 or 10. Passives, who answer 7 or 8, are not counted in the formula but are included in the total.

What is a good NPS for a restaurant?

Compare against your own trend rather than an outside benchmark, because survey timing, channel and segment change the score a lot. A rising score with a falling detractor share is the sign of real progress.

How many responses do we need for a reliable NPS?

Below about 50 responses in a period, read the comments rather than the score. Between 50 and 150, use a rolling three-month score. Above 150, monthly scores become usable.

Is NPS better than a satisfaction score?

They do different jobs. A satisfaction score helps spot problems on a shift; NPS tracks loyalty over months. Many restaurants use both in one short survey.

References

  1. Bain & Company (Net Promoter System) (2026). Measuring Your Net Promoter Score (opens in a new tab) ↑

  2. National Restaurant Association (2024). Where operators plan to invest in tech (opens in a new tab) ↑

  3. National Restaurant Association (2025). Restaurant Industry Poised for Growth in 2025 (2025 State of the Restaurant Industry press release) (opens in a new tab) ↑

  4. National Restaurant Association (2025). From Trend to Transformation: Off-Premises Dining Now Essential for Restaurant Consumers, Operators (opens in a new tab) ↑

Book a demo

See Ava handle your calls.

Book a short demo. We’ll walk through your calls and workflow, and agree what Ava should take care of.

Book a demo