NPS survey tool for SaaS
Ask at the perfect moment. Run NPS, star ratings, multiple-choice, or open-text questions right inside your product — then read every response in one place.
Every question type
NPS, star rating, multiple-choice, and open text.
In the moment
Trigger surveys exactly where the experience is fresh.
Read the results
Responses are summarized — averages, NPS, and tallies — and exportable.
Live in three steps.
Pick a type
NPS, rating, choice, or text.
Target the moment
Show it to the right users at the right time.
Review responses
See the summary and export the raw data.
An NPS pulse that fires at the right moment.
NPS results are mostly a function of timing. Asking at login is lazy and skews negative; asking right after delivered value is honest. Here is the build.
Pick the moment
Trigger the survey on a custom event — say report_exported, the moment your product just did its job. Delay it 3 seconds so it never collides with the action itself.
Keep the ask minimal
Step one: the 0–10 NPS scale with end labels ("Not likely" / "Extremely likely"). Step two: one open-text follow-up — "What is the main reason for your score?" Nothing else.
Choose who and how often
Audience: identified users active at least 30 days (a saved segment on a signup-date trait). Frequency: once — a quarterly cadence means republishing quarterly, not nagging weekly.
Read it where it lands
Responses shows the average, the full 0–10 distribution, and every verbatim tied to the user who wrote it — exportable to CSV when finance asks.
What to expect: Response rates for well-timed in-product NPS run far above email surveys, and the verbatims arrive attached to real accounts — so a detractor comment is a follow-up, not an anonymous complaint.
When not to use surveys & nps.
No pattern fits every job. Here’s where we’d point you at something else — sometimes ours, sometimes not.
Right after signup
Users cannot rate what they have not experienced. NPS before value is measured noise — gate it on tenure or on the event that represents delivered value.
Long research questionnaires
In-product surveys win by being tiny. Ten-question studies belong in a dedicated research tool over email — inside the app they tank both completion and goodwill.
When you will not act on it
Collecting scores you never read teaches users that feedback disappears into a void. If nobody owns the follow-up on detractors, fix that before adding the survey.
Interrupting failure states
Never fire a satisfaction survey on an error page or after a support escalation. You will measure the incident, not the product — and salt the wound doing it.
What question types are supported?
NPS (0–10 with end labels), star and numeric rating scales with configurable min/max, multiple choice, and open text — plus full form fields (text, email, number, textarea, select) when you need structured input rather than a score.
Where do responses go?
The Responses tab — project-wide and per guide. Scores show averages plus the full distribution (read the distribution: an NPS "7 average" often hides a promoter/detractor split), choices show ranked tallies, and everything exports to CSV.
Can I see who answered?
Identified users appear with name or email and their traits; anonymous users appear by ID. That link to the account is what makes in-product feedback actionable — a detractor is a person you can reach, not a data point.
How do I stop surveys from annoying people?
Frequency "once" per survey, plus project-wide rate limits (max guides per session and per day, with a cooldown) that surveys respect like any guide. One question at the right moment beats ten at the wrong one.
In-app NPS or email NPS?
In-app response rates are typically several times email's, and answers arrive in context, seconds after the experience being rated. Email still wins for churned or dormant users — the people who are not inside your product to be asked.
See it on your own app.
Ship your first guide in minutes. Free to start — no credit card.