NPS Survey Form Template
Most NPS programs collect scores and file them. The customer who answered 0 — would never recommend and the customer who answered 10 — evangelist receive the same silence, which tells both of them the question was decorative. This template treats every score as the start of a workflow: the survey branches the moment the number lands, sends each band down a different path, and turns a measurement exercise into a set of planned next actions.
Three Bands, Three Different Conversations
The branch happens immediately after the score question:
- Promoters (9–10) — asked what they value most, then invited toward advocacy: a review, a referral, a case study. Enthusiasm is perishable; capture it while the goodwill is warm
- Passives (7–8) — asked the tipping question: what would move you from a 7 to a 9? Passives are customers with an unspoken grievance they consider too small to mention uninvited. Ask, and they’ll tell you exactly what it is
- Detractors (0–6) — asked what went wrong, with space to explain, and offered a route to a human. A detractor who gets a genuine response within days often becomes a louder advocate than someone who never had a problem
The branching uses simple [yes/no-style conditional logic](/help/features/question-types/yes-no/) on the score — each band sees only its own follow-up, so nobody answers questions that don’t apply to them.
The Permission Layer
Outreach without consent is how feedback programs become annoyance programs:
- After the follow-up, one clear question: may we contact you about this?
- Consenters leave their email through a contact field; everyone else’s score still counts, untouched
- Your team works from a list of people who asked to hear from them — which changes the tone of every conversation that follows
Routing Scores to the People Who Act
A follow-up workflow dies in an inbox nobody watches, so wire it to the people who can move:
- A new-submission alert pings the account owner the moment a detractor lands — recovery is a same-week sport
- Push responses into your CRM by webhook so scores attach to the customer record and the follow-up happens inside the tool your team already lives in
- Tag promoters’ consent through hidden fields so marketing receives a clean, permissioned advocacy list without anyone forwarding spreadsheets
The Rhythm That Compounds
Run this branched survey on your regular relational cadence, and the workflow effects stack: detractors recovered before they churn, passives converted by fixes they named themselves, promoters mobilized into reviews and referrals. Score each cycle with the calculator, and watch not just the number climb but the ratio of action to measurement — which is the only metric that separates an NPS program from an NPS ritual.
Frequently Asked Questions
Should follow-up questions really differ by score band?
Yes — it’s the entire efficiency of the design. A promoter has nothing to complain about and a detractor has no interest in your referral program; asking both the same questions wastes their time and muddies your data. Branching respects each respondent’s actual situation, which is also why branched surveys complete at higher rates.
What's the single best question to ask passives?
The tipping question: what would it take to make this a 9 or 10? Passives sit one fixable grievance away from promotion — slow support response, a missing integration, unclear pricing — and this question names it in their own words. Passives converted through their own stated fix show the strongest retention lift of any NPS segment.
How fast should detractor outreach happen?
Within two or three business days, ideally faster. Detractor feedback has a short shelf life — the irritation that produced a 2 becomes a settled decision to leave within weeks. Speed signals that the score mattered; a recovery call a month late reads as damage control rather than care.
Can we run branched NPS without a CRM?
Yes. The alerts, exports, and permission lists work standalone — a small team can run the entire loop from the response dashboard and a weekly review. The webhook matters when volume grows past what inbox triage can handle, not before.
Does branching affect how the score should be interpreted?
Not at all — the score question stays first, untouched, before any band sees a different path. Every respondent answers the identical question in identical context, which is what keeps the number comparable. The branching only shapes what happens after the measurement, which is exactly where it belongs.