The NPS Survey Template Checklist: What to Include Before You Send¶
Everyone knows the core NPS question - "How likely are you to recommend us to a friend or colleague, 0 to 10?" What separates a genuinely useful NPS survey from one that produces a number nobody can act on is everything built around that single question, and it's the part people most often skip when they're in a hurry to send the first wave.
The Core Question¶
A single Rating question, 0 to 10, with clear endpoint labels ("Not at all likely" through "Extremely likely," not bare numbers alone). Keep the scale identical every time you run this survey - switching between a 0-10 and any other range partway through your history of waves makes every later comparison harder than it needs to be.
The Follow-Up That Makes the Score Actionable¶
A single number tells you where someone sits; it tells you nothing about why. Every NPS survey worth running needs an open-ended follow-up - "What's the main reason for your score?" - and it should be genuinely open, not multiple choice, since the whole point is capturing the reasoning in the respondent's own words rather than forcing it into categories you guessed at in advance. If you're planning to code these responses at any real volume, this is exactly the kind of open-ended data our AI-powered text analytics is built to categorize at scale rather than reading one by one.
Segmenting Detractors, Passives, and Promoters Differently¶
The standard NPS bands - detractors (0-6), passives (7-8), promoters (9-10) - mean genuinely different things, and a well-built template treats them differently downstream rather than lumping every response into one undifferentiated pile. Display logic showing a slightly different follow-up prompt depending on which band a respondent falls into (a promoter might get "What do you love most?" while a detractor gets "What would have made this better?") produces more specific, more usable answers than one generic follow-up shown to everyone regardless of their score.
A Hidden Field for Tracking Context¶
If you're sending NPS through more than one channel - email, in-app, a support ticket follow-up - a Hidden Field question capturing which channel a response came from (via a URL parameter on each distribution link) lets you segment results by source later without asking respondents to self-report where they came from. See our guide on passing data into a survey with hidden fields for the setup.
What to Leave Out¶
Resist adding extra rating questions "while you're at it." NPS surveys work best short - the core score, one open-ended follow-up, maybe a light segmenting question if you don't already know it from other data. Every additional question is a small tax on completion rate, and a bloated NPS survey defeats the purpose of a pulse check meant to be answered in under a minute.
Save It Once, Reuse It Every Wave¶
Build this structure once - question, follow-up, branching by score band, tracking field - and save it as a template rather than rebuilding it for every quarterly wave. The value of NPS specifically comes from tracking the same measure consistently over time, and a template is what keeps every wave identical to the one before it. See our guide on saving a survey as a reusable template for how.
FAQ¶
Should the follow-up question be required?
Generally no - forcing an explanation adds friction right at the end of an otherwise quick survey, and a genuinely engaged respondent will usually answer anyway. Making it optional tends to produce a better completion rate with only a modest drop in follow-up response rate.
How often should I run an NPS survey?
There's no universal cadence, but quarterly is common for most B2B relationships - frequent enough to catch a real trend, infrequent enough not to fatigue the same respondents.
Can I use a ready-made NPS template instead of building this from scratch?
Check the template gallery before building your own - a ready-made Classic Survey template covering this structure may already exist there. See Starting a Survey from the Template Gallery.