NPS · CSAT · CES · PMF · Qualitative · Behavioral
Types of Customer Feedback
The 6 types of customer feedback every SaaS team should know - what each one measures, when to use it, and what it misses.
No credit card required
The 6 types of customer feedback
Each type answers a different question. Knowing which to use - and when - is more valuable than running all of them at once.
NPS - Net Promoter Score
Core question:
"How likely are you to recommend us to a friend or colleague?" (0–10)
Measures
Long-term loyalty and word-of-mouth potential
When to use
Every 90 days as a relationship survey, or triggered at 30 days post-onboarding and before renewal
What it tells you
Whether your users would recommend you - the best leading indicator of organic growth
Scale
0–10 · Detractors (0–6), Passives (7–8), Promoters (9–10)
Blind spot
Why they gave that score without a follow-up open-ended question
CSAT - Customer Satisfaction Score
Core question:
"How satisfied were you with [interaction]?" (1–5)
Measures
Satisfaction with a specific touchpoint - feature, support interaction, onboarding step
When to use
Within minutes of a specific interaction - support resolution, feature completion, onboarding milestone
What it tells you
Whether a specific touchpoint is meeting expectations - much more precise than overall satisfaction
Scale
1–5 · % who rate 4–5 = your CSAT score
Blind spot
Long-term relationship health - a user can be satisfied with support but still churn
CES - Customer Effort Score
Core question:
"How easy was it to [complete task]?" (1–7)
Measures
Friction in specific workflows - the biggest driver of churn you can actually fix
When to use
After high-effort interactions: onboarding setup, support ticket resolution, import/export tasks, billing changes
What it tells you
Where users are working too hard - friction that accumulates into churn before users tell you
Scale
1–7 · Gartner research shows CES is the strongest predictor of churn among the three core scores
Blind spot
Whether users find the product valuable - a frictionless experience of the wrong thing still churns
PMF - Product-Market Fit
Core question:
"How would you feel if you could no longer use this product?" (Very / Somewhat / Not disappointed)
Measures
How essential your product is to users - the leading indicator of retention and organic growth
When to use
After 30+ days of active use (minimum 40 responses needed), and after major product changes
What it tells you
Whether you've found product-market fit - the most important question for early-stage teams
Scale
3 choices · 40%+ 'Very disappointed' = product-market fit. Below 40%: iterate before scaling.
Blind spot
Why users feel that way - always pair with 'What would you use instead?' and 'What do you love most?'
Open-Ended Qualitative
Core question:
"What's the one thing we could do to make this product better for you?"
Measures
The why behind quantitative scores - motivations, friction, desires, use cases
When to use
As a follow-up to any quantitative survey, in user interviews, at onboarding, and at cancellation
What it tells you
What your users are actually experiencing - often reveals problems you didn't know to ask about
Scale
No scale - qualitative text. Group into themes: feature requests, usability issues, pricing, praise, churn signals
Blind spot
Prevalence - you need 20+ responses to know if a theme is widespread or just one vocal user
Behavioral Signals
Core question:
What users do - not what they say
Measures
Actual usage patterns: feature adoption, login frequency, session length, drop-off points, export behavior
When to use
Continuously - behavioral data is always flowing from your product. Review it alongside survey data.
What it tells you
What users actually value (vs. what they say they value) - adoption data is more honest than stated preferences
Scale
No scale - usage metrics from your product analytics (Mixpanel, Amplitude, PostHog)
Blind spot
The why - a user who stops using a feature could be satisfied (found a workaround) or frustrated (gave up)
Feedback collection methods
The channel matters as much as the question type. In-product surveys consistently outperform email on response rate and data quality.
In-product (in-app)
Strengths:
- ✓ Highest response rate - reaches users in context
- ✓ Tied to specific user actions and attributes
- ✓ Can segment by plan, role, tenure
Limitations:
- – Interrupts the product experience if overused
- – Only reaches logged-in users
Email surveys
Strengths:
- ✓ Reaches users outside the product
- ✓ Good for churn exit surveys
- ✓ Can include longer survey formats
Limitations:
- – Lower response rate than in-product
- – No behavioral context at time of completion
Website widget / feedback button
Strengths:
- ✓ Captures feedback from anonymous visitors
- ✓ Always-on - no scheduling needed
- ✓ Good for pricing page and docs friction
Limitations:
- – No user identity - can't segment by attributes
- – Often skewed toward frustrated users who seek out feedback buttons
Link surveys
Strengths:
- ✓ Works outside your product - Slack, email, social
- ✓ Good for non-customer research
- ✓ No installation required
Limitations:
- – No identity linking - anonymous by default
- – Self-selection bias - who clicks the link shapes the results
User interviews
Strengths:
- ✓ Deepest qualitative insights available
- ✓ Can follow up on any answer in real time
- ✓ Builds customer relationships
Limitations:
- – Time-intensive - 30–60 min per session
- – Small sample size - not representative
- – Requires scheduling and recruitment effort
Which type to use for your goal
Pick the feedback type by goal - not by which one you've heard of most.
| Your goal | Recommended type | Frequency | Notes |
|---|---|---|---|
| Track loyalty over time | NPS | Quarterly | Your primary relationship health metric |
| Improve onboarding | CSAT + CES | After onboarding completes | CSAT for overall satisfaction, CES for ease |
| Reduce support churn | CSAT | Within 5 min of ticket resolution | Support CSAT is a direct churn predictor |
| Validate product-market fit | PMF | At 40+ active users, then quarterly | 40%+ 'very disappointed' = PMF |
| Understand why users leave | Exit survey (open-ended) | At cancellation trigger | Most critical feedback - highest urgency to act |
| Find feature usability issues | CES + open-ended | After first use of a feature | CES finds friction; open-ended names it |
| Discover unknown problems | Open-ended qualitative | User interviews monthly | Can't find unknown unknowns with surveys alone |
Frequently asked questions
Run all 4 survey types from one platform
NPS, CSAT, CES, and PMF surveys - in-product or via email. Every response linked to the real user so you can segment by plan, role, or any custom attribute.
Start FreeNo credit card required