candor

Templates / NPS survey

Customers · Adaptive interview template

NPS survey template

Net Promoter Score has a design flaw that everyone accepts: it reduces a relationship to one number between 0 and 10. That number is comparable and easy to chart, and it tells you almost nothing about what to do next. A room full of 7s and a room full of 9s can have identical stories behind them, and NPS erases the difference.

This template runs NPS properly by treating the score as the opening question, not the survey. A respondent picks their number, and the AI interviewer immediately does what the followup box was always meant to do: it asks a promoter what they would tell a friend, asks a passive what is missing from a 10, and asks a detractor what specifically went wrong. The score still gives you the benchmark; the interview gives you the reason the benchmark moved.

When to use it

  • Relationship NPS, sent on a schedule to gauge overall sentiment over time
  • Transactional NPS, fired after a support ticket, purchase, or renewal
  • When your existing NPS number is flat and nobody can explain why
  • When leadership wants the benchmark but the team needs the reasons behind it

Running it well

  • The score is a routing signal, not the finding. Its job is to tell the interviewer whether to ask about advocacy, gaps, or damage; the transcripts are where the decisions come from.
  • Read passives before promoters. A room of 7s and 8s is the cohort closest to moving in either direction, and their follow-ups are the shortest path to a higher number.
  • Do not force the comment. Skip is always allowed, and the people who choose to explain a low score are giving you your highest-value transcripts.
  • Keep the question wording standard if you want to benchmark. The follow-ups add depth without breaking comparability with your historical NPS.

The questions, and why they're shaped this way

These are the anchor questions. You can edit, reorder, or add to them; the AI interviewer handles the depth between them.

NPS 0 to 10

How likely are you to recommend us to a friend or colleague?

The standard 0 to 10 question gives you the benchmark and, more importantly, tells the interviewer which of three conversations to open. The score is the routing key.

Open text

What's the main reason for your score?

This is the classic NPS follow-up, but here it is a live opening rather than a dead-end text box. Whatever the respondent types, the AI reacts to it and keeps going toward the specific reason.

What the interviewer asks next

Follow-ups aren't scripted. The AI reads each answer and probes what a researcher would. Depending on what a respondent says, it might ask:

"You gave a 10. If you were recommending us to a colleague tomorrow, what exactly would you say to them?"

"You landed on a 7. What is the one thing that keeps it from being a 9 or a 10?"

"You gave a 3. What specifically happened, and was it a one-time thing or a pattern?"

Live demo

Play the nps survey interview

This is the real interviewer, not a recording. Answer as yourself and watch the follow-ups adapt to what you write.

Prefer a full window? Open the demo.

Common questions

Is this still a real NPS score I can benchmark?
Yes. The scoring question is the standard 0 to 10 recommend question, so promoters, passives, and detractors are calculated the usual way. The follow-ups add the reasoning on top without changing the number.
How is this different from NPS with a comment box?
A comment box gets one line and then goes quiet. Here the AI reads that line and asks the next question a researcher would, so a detractor's "support was slow" becomes the specific ticket, the specific wait, and whether it is a pattern.
Should I run relationship or transactional NPS?
Both fit this template. For relationship NPS send on a schedule; for transactional NPS fire it after an event like a ticket or renewal and the follow-ups will anchor to that experience.
What do the results look like?
The NPS breakdown by segment, themes with supporting quotes and counts, full transcripts, and a plain-language question box for asking things like "what did passives say was missing from a 10?"

Run the nps survey with your own audience

Join the waitlist