NPS Survey Questions: Templates That Get Answers You Can Use

An NPS score without follow-up answers is a number you can worry about but not act on. A 32 became a 28 — now what?

The score question itself is standardized, and you shouldn't touch it. All the room for craft is in the one or two questions that come after — and in when you ask. This post gives you copy-ready follow-ups for each score segment and the timing rules that decide whether anyone answers at all.

The main question: don't get creative

The canonical wording:

How likely are you to recommend [product] to a friend or colleague? (0 = not at all likely, 10 = extremely likely)

Resist the urge to reword it. "How much do you love our product?" and "Would you tell your team about us?" feel fresher, but comparability is the entire point of NPS — against your own past scores and against benchmarks. Change the wording and you're running a custom survey with an 11-point scale, which is fine, but it isn't NPS.

Two rules that do matter:

• Keep the 0–10 scale with labeled endpoints. Users anchor on the labels, not the numbers.
• Substitute the product name, nothing else. "Recommend SeggWat" — not "recommend our feedback platform solution."

The follow-up is where the value lives

One open question after the score converts a metric into a roadmap input. The mistake is asking everyone the same generic "Why?" — a detractor and a promoter are in completely different conversations with you.

Segment the follow-up by score:

Detractors (0–6):

What's the main thing we'd need to fix for that score to go up?

This works better than "What went wrong?" because it asks for one thing and points forward. Angry users write essays when asked what's wrong; asked what to fix, they prioritize for you.

Passives (7–8):

What's missing for that to be a 9 or 10?

Passives are your highest-information segment. They like the product, they've stayed, and they can usually name the exact gap keeping them from enthusiasm. This is the closest thing NPS has to a feature-request generator.

Promoters (9–10):

What's the main thing we should never change?

Most teams ask promoters nothing, which wastes the segment. Their answers tell you what your product's actual core value is — frequently not the thing on your landing page. That's copywriting material and a do-not-break list in one.

Keep it to one open question per response. Every additional question costs completion rate, and a half-finished survey with three thoughtful prompts loses to a finished one with a single good prompt.

Optional: one closing question

If you want a second data point, make it closed and cheap to answer:

Which best describes your role? (Founder / Engineer / Product / Other)

Segmenting NPS by role or plan is often more revealing than the aggregate score — a product beloved by founders and tolerated by their teams shows up here first.

Timing questions are survey questions

The same questions fire at different moments and produce different data. Three rules:

• Wait for an opinion to exist. NPS during onboarding measures your signup flow, not loyalty. Gate it on account age or a meaningful usage threshold — for example, only on paths behind a login, after the user has hit their key action a few times.
• Ask in-app, not by email. Email NPS response rates are typically single-digit and skew toward the extremes. An in-product survey catches users while the product is literally in front of them.
• Enforce a cooldown. Nobody should see NPS twice a quarter, and answering (or dismissing) it should suppress every other survey for a while. One over-surveyed week costs you months of goodwill.

In SeggWat's survey widget, these are trigger settings rather than engineering work: URL-pattern targeting, time-on-page delays, custom-event triggers like SeggwatSurvey.trigger("report_exported"), and a global cooldown that spaces surveys automatically.

Copy-ready template

The full set, ready to paste:

1. How likely are you to recommend [product] to a friend or colleague? (0–10)
2. If 0–6: What's the main thing we'd need to fix for that score to go up? (open text)
3. If 7–8: What's missing for that to be a 9 or 10? (open text)
4. If 9–10: What's the main thing we should never change? (open text)
5. (Optional) Which best describes your role? (choice)

Trigger: in-app, users active ≥ 30 days, 90-day cooldown, suppressed on checkout and settings pages.

SeggWat ships this as a ready-made NPS template — pick it when creating a survey, adjust the follow-ups, set it live. The 0–10 button row, score segmentation, and response dashboard are already wired. Click through the whole flow in 8 screenshots if you want to see it before signing up.

Try SeggWat free and have the first response before your next deploy.