Survey type guide

Survey Types: 6 Kinds of Surveys and When to Use Each

The type you choose determines the question, the scale, and the moment you send it.

Different types of surveys measure different things. Using the wrong kind gives you a real number that answers the wrong question. This guide covers the 6 main survey types used by SaaS teams - what each measures, what question to ask, when to send it, and what the result actually tells you.

No credit card required

6 survey types at a glance

Each kind of survey answers a distinct question. Match the type to the question - not to habit.

TypeMeasuresScaleCadence

NPS

Net Promoter Score

Overall loyalty and likelihood to recommend0–10 numeric scaleQuarterly (relationship survey). Also triggered at 30 days post-onboarding and 60 days before renewal.

CSAT

Customer Satisfaction Score

Satisfaction with a specific interaction or touchpoint1–5 scale. CSAT score = % of respondents rating 4 or 5.Triggered immediately after: support ticket resolution, onboarding completion, first use of a key feature.

CES

Customer Effort Score

How much effort a task required - friction and difficulty1–7 agreement scale. CES score = average across all responses.Triggered after: setup wizard, support ticket, checkout, any multi-step task where friction is a known risk.

PMF

Product-Market Fit Survey

Whether your product is essential to users3 options: Very disappointed / Somewhat disappointed / Not disappointedOnce you have 40+ active users. Repeat after major product changes, pivots, or new segment entry.

Exit

Exit / Churn Survey

Why users cancel - the specific reason, not a satisfaction scoreMultiple choice (pricing / missing feature / switching to X / no longer needed) + optional open-ended follow-upTriggered at the moment of cancellation, before the account is closed. Not sent in a follow-up email days later.

UX

UX Research Survey

Friction and confusion in a specific flow or featureOpen-ended, qualitative. Sometimes paired with a single 1–5 rating for the experience.After completing a specific task for the first time. Or sent to users who started a flow but did not complete it.

Each survey type in depth

Question, scale, trigger, output, and benchmark for each of the 6 kinds.

NPS

Net Promoter Score

Measures: Overall loyalty and likelihood to recommend

Standard question

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

Scale

0–10 numeric scale

When to send

Quarterly (relationship survey). Also triggered at 30 days post-onboarding and 60 days before renewal.

Benchmark

B2B SaaS median: 31–40. Above 50 is excellent.

What the result tells you: Promoters (9–10) are your referral engine. Detractors (0–6) are churn risk. The gap between segments tells you where to act.

Complete NPS guide →

CSAT

Customer Satisfaction Score

Measures: Satisfaction with a specific interaction or touchpoint

Standard question

"How satisfied were you with your [support / onboarding / feature] experience?"

Scale

1–5 scale. CSAT score = % of respondents rating 4 or 5.

When to send

Triggered immediately after: support ticket resolution, onboarding completion, first use of a key feature.

Benchmark

75–85% is good for SaaS. Below 70% warrants immediate investigation.

What the result tells you: Identifies which specific touchpoints are failing. A low CSAT after support signals a process problem, not a product problem.

Complete CSAT guide →

CES

Customer Effort Score

Measures: How much effort a task required - friction and difficulty

Standard question

"The company made it easy to handle my issue." (Strongly disagree → Strongly agree)

Scale

1–7 agreement scale. CES score = average across all responses.

When to send

Triggered after: setup wizard, support ticket, checkout, any multi-step task where friction is a known risk.

Benchmark

5.5+ is good. Below 5.0 is a red flag for that interaction.

What the result tells you: High effort predicts churn better than low satisfaction. A CES below 5.0 identifies friction that will drive cancellations.

Complete CES guide →

PMF

Product-Market Fit Survey

Measures: Whether your product is essential to users

Standard question

"How would you feel if you could no longer use this product?"

Scale

3 options: Very disappointed / Somewhat disappointed / Not disappointed

When to send

Once you have 40+ active users. Repeat after major product changes, pivots, or new segment entry.

Benchmark

40% very disappointed is the threshold. Superhuman reached 58%.

What the result tells you: 40%+ "very disappointed" = product-market fit. Below 40% = keep iterating. Segment by user type to find where PMF exists.

Complete PMF guide →

Exit

Exit / Churn Survey

Measures: Why users cancel - the specific reason, not a satisfaction score

Standard question

"What is the main reason you decided to cancel today?"

Scale

Multiple choice (pricing / missing feature / switching to X / no longer needed) + optional open-ended follow-up

When to send

Triggered at the moment of cancellation, before the account is closed. Not sent in a follow-up email days later.

Benchmark

No standard benchmark. Track the top reason over time - if it changes, your product or pricing changed.

What the result tells you: The single highest-signal data source for product and pricing decisions. Identifies pattern cancellation reasons by segment.

Complete Exit guide →

UX

UX Research Survey

Measures: Friction and confusion in a specific flow or feature

Standard question

"What, if anything, felt confusing or unclear when you [completed X]?"

Scale

Open-ended, qualitative. Sometimes paired with a single 1–5 rating for the experience.

When to send

After completing a specific task for the first time. Or sent to users who started a flow but did not complete it.

Benchmark

No benchmark. Track themes across 20–30 responses - patterns emerge quickly.

What the result tells you: Specific language users use to describe confusion - invaluable for rewriting UI copy. Reveals friction invisible in analytics.

Complete UX guide →

How to choose the right survey type

Start with the question you need to answer - not the survey type you are most familiar with.

Do users love us enough to recommend us?

Run quarterly. Score below 20 = urgent retention problem.

Use NPS

Was this specific interaction satisfying?

Trigger immediately after support, onboarding, or feature use.

Use CSAT

Was this task easy or was it frustrating?

Trigger after any multi-step task. High effort predicts churn.

Use CES

Do we have product-market fit?

Run at 40+ active users. 40%+ very disappointed = PMF.

Use PMF survey

Why are users cancelling?

Trigger at cancellation. The highest-signal data you can collect.

Use Exit survey

Where is this specific flow confusing?

Trigger after task completion or after an abandoned flow.

Use UX research survey

Running multiple survey types: frequency rules

The goal is the right survey at the right moment - not comprehensive coverage of every user in a single week.

One relationship survey per quarter

NPS is a quarterly signal - sending it more frequently produces noise, not insight. If NPS drops quarter over quarter, investigate with CSAT and CES at specific touchpoints rather than re-running NPS.

Trigger transactional surveys by event

CSAT and CES should fire automatically after the relevant event occurs - not on a schedule. A user who contacts support three times in a month should receive CSAT three times, at each resolution. That is not over-surveying - it is the right signal at the right moment.

Cap to one survey per 30 days per user

Apply a frequency cap so the same user cannot receive more than one survey per 30-day window across all types. This prevents a single user from getting NPS on Monday and CSAT on Wednesday. The cap protects response rates without reducing signal quality.

Frequently asked questions

Get started today

Run NPS, CSAT, CES, and PMF surveys - all from one place

Every survey type in one tool. In-product triggers, every response linked to a real user, segmented from the first result. Free to start.

Start Free

No credit card required