How to measure NPS properly: ask the 0-to-10 rating plus an open-ended why, fix the method, and read the reason on arrival.
You measure NPS by asking one rating question — how likely are you to recommend us, on a 0-to-10 scale — plus one open-ended follow-up asking why, then calculating the score as the percentage of promoters (9–10) minus the percentage of detractors (0–6). Measuring it well means capturing the reason, not just the rating, and keeping both on the person over time. The rating is the metric; the reason is what makes it worth measuring.
Most teams get the calculation right and the measurement wrong. They send the rating question, tally the score, and drop the “why” comment into a column no one reads. So they measure a number every month and learn nothing from it, because the half of the survey that explains the score is discarded at collection.
Key takeaways
The Net Promoter question is deceptively small: one number from zero to ten. Its power was never the number alone but the pairing with a “why,” because the number ranks customers and the reason tells you what to do about each rank. A program that measures only the rating has instrumented the thermometer and thrown away the chart, then wonders why the score never turns into action.
The discard happens at collection. A form-centric tool treats the rating as the datum and the comment as optional text, storing them apart with no mechanism to read the comment or keep it on the respondent. Sopact calls the alternative the Verbatim Thread: the reason captured and read against your codebook on arrival, themed and cited, and kept on the same persistent contact record as the rating and every later wave. Measurement then yields a ranked set of drivers by default, the model behind Sopact’s feedback tools.
Measuring NPS moved through three eras. First, an email survey and a spreadsheet where the score was calculated by hand and the reasons skimmed. Then dashboards — Delighted, AskNicely, SurveyMonkey and peers — automated the send, the calculation, and the trend, and offered a word cloud for the comments. The current era captures the reason clean at the source and reads it into drivers on arrival, so measuring the score and understanding it are one action.
The one test that separates the eras: ask whether your measurement captures the reason as structured, readable data tied to the rating and the respondent — or as loose text in a column. If the “why” is loose text, you are measuring a number and archiving its explanation. Measurement that keeps the reason readable is the difference between a metric and an instrument for improvement.
NPS is sensitive to how it is collected. A relationship survey asked annually measures overall sentiment; a transactional survey asked after a specific interaction measures that interaction. Send timing, channel, sampling frame, and exact wording all move the score, so two measurements taken differently are not comparable even for the same customers. Measuring NPS well means deciding which of these you want and then holding it fixed.
The discipline that makes a trend trustworthy is fixing the method and validating at the source: one clear question, a consistent trigger, and a unique link per respondent so responses land clean on the right record. That clean-at-the-source capture is what keeps wave two comparable to wave one and lets you follow the same respondent, the same foundation continuous feedback and honest benchmarking both rest on.
Ask the standard 0-to-10 recommend question, follow it immediately with an open-ended why, fix the timing and audience, assign each respondent a persistent identity, and read the reasons on arrival — so the score arrives already explained. The open-ended follow-up is not optional; it is the part that converts a rating into a driver. Fixing timing and identity is what lets you trend the score and follow people across waves.
The output of a well-measured NPS is not just a number but a number with its ranked reasons attached, per segment, traceable to the comments. Because Sopact captures the reason clean at the source and keeps it on the Verbatim Thread, every measurement produces that immediately, and the score you report to leadership already carries the drivers your team needs — the questions that make this work are on NPS survey questions.
Measuring only the rating produces a number; measuring the rating and the reason produces an instrument you can act on. The difference is decided at collection, not analysis.
| The question | Rating only | Rating + reason (Verbatim Thread) |
|---|---|---|
| What is captured? | The 0–10 score | The score and the read, structured why |
| Can it explain the number? | No: the reason is discarded | Yes: drivers ranked and cited on arrival |
| Does it follow the person? | No: each wave is its own pool | Yes: one persistent record across waves |
| When do you learn? | At the next analysis, if ever | The day the response lands |
The reading side is NPS analysis; the question wording that makes measurement clean is NPS survey questions.
An NPS number is a lagging summary; the reason a customer gave it is in the comment, and only if someone reads the comment while it still matters. That is the premise of the Loop, Sopact’s method for continuous intelligence: collect clean at the source, analyze the moment a response arrives, improve while there is still time to act. On a feedback program that means a detractor’s reason is themed the day it lands and routed to someone who can close the loop, not filed for a quarterly readout.
The Loop is also what makes an NPS finding defensible: every driver and theme traces to the exact comment it came from, the standard detailed in Loop traceability, so “price is the top detractor driver” is backed by the sentences, not a hunch.
One method, three moves that never stop
Then the cycle runs again, a little sharper each time. Read the method: the Loop methodology →
The fastest way to fix a measurement is to run your own responses through a proper read. Export your ratings and why-comments, then paste the prompts below into Sopact Sense’s Assistant, or reason through them with your team. The arrow above each links the Academy walkthrough with the expected output and tips.
Academy walkthrough → Analyze open-ended survey responses
Here are our NPS comments with each respondent's score: [ATTACH]. Theme every comment against our codebook, cite the sentence behind each theme, and show the top five themes among detractors, passives, and promoters separately.
Academy walkthrough → Analyze sentiment and its drivers
For these NPS verbatims: [ATTACH], score sentiment per comment, then show the specific drivers behind the detractor comments — the recurring reasons, each with a quoted example — so I know what to fix, not just the score.
Academy walkthrough → Connect quant and qual data
Here are our NPS scores and open-ended comments on the same respondent IDs: [ATTACH]. Show which themes explain the detractor scores, quote a comment for each, and list the specific respondents worth a personal follow-up.
Academy walkthrough → Analyze NPS across waves
Here are two waves of NPS from the same respondents on the same IDs: [ATTACH]. Track each respondent's movement between waves, show who moved from detractor to promoter and who slipped, and surface the comments that explain the biggest moves.
Each walkthrough is short and practical: what to do, the prompt to run, the output to expect, and the tips that keep it reliable.
Watch: reading feedback on arrival and tying every comment to the person who gave it.
Ask the 0-to-10 recommend question plus an open-ended why, then calculate the percentage of promoters (9–10) minus the percentage of detractors (0–6). Measuring it well means capturing and reading the reason, not just the rating. Sopact keeps both on the Verbatim Thread so the score arrives with its drivers attached.
Percentage of promoters minus percentage of detractors, giving a number between minus 100 and plus 100; passives (7–8) count toward the base but not the score. Sopact keeps the calculation tied to the comments, so the number is never reported without the reasons behind it available.
Yes. The open-ended why is the part that turns a rating into a driver you can act on. Measuring only the number instruments the thermometer and discards the diagnosis. Sopact captures the reason clean at the source and reads it on arrival, so the follow-up is analyzed, not archived.
Yes, significantly. A relationship survey and a transactional one, different send times, channels, and wording all move the score, so measurements taken differently are not comparable. Measuring NPS well means choosing the method you want and holding it fixed. Sopact validates capture at the source so the trend is honest.
Often enough to catch movement while you can act, which for most programs means a continuous or rolling measurement rather than an annual one. Sopact reads each response on arrival, so continuous measurement produces continuous drivers rather than a once-a-year reading marathon.
Fix the question wording, audience, timing, and scale, and keep each respondent identifiable across waves on a persistent record. Then a change in the score is a change in the experience. Sopact’s Verbatim Thread keeps the same respondents connected so wave two is comparable to wave one.
One clear rating question, an immediate open-ended why, fixed timing and audience, a persistent identity per respondent, and reasons read on arrival. That produces a score already explained by ranked drivers. Sopact builds all of it into collection, so measurement and understanding are the same step.
Next: read what you measured in NPS analysis, or write questions that capture clean in NPS survey questions.
