Surveys

Surveys let you ask users questions directly in your product. Collect qualitative feedback, measure satisfaction with NPS/CSAT, and understand user needs without leaving PostHog.

How surveys work

Surveys appear as popovers while users are in your product:

Define survey questions and appearance
Target specific users or pages
Survey appears as an overlay in your app
Responses are collected in PostHog
Analyze results and identify themes

Surveys use the same PostHog snippet as analytics. No additional installation required.

Survey types

Choose the right survey type for your use case:

Popover (default)
Widget
API

Appears via CSS selector or embedded button:

<!-- Button triggers survey -->
<button data-survey-id="your-survey-id">
  Give feedback
</button>

Best for: Opt-in feedback, contact forms, feature requests

Headless survey for custom implementations:

// Get survey config
const survey = posthog.getActiveMatchingSurveys()[0]

// Show your custom UI
showCustomSurveyUI(survey)

// Submit response
posthog.capture('survey sent', {
  $survey_id: survey.id,
  $survey_response: userAnswer
})

Best for: Custom UI, mobile apps, unique workflows

Question types

Open-ended

Free-form text input for qualitative feedback:

Question: "What's the main reason you signed up?"
Answer: [Text input box]

Use for: Understanding motivations, collecting feature requests, bug reports

Single choice

Select one option from a list:

Question: "How easy was it to get started?"
Options:
- Very easy
- Somewhat easy  
- Neutral
- Somewhat difficult
- Very difficult

Use for: Satisfaction levels, yes/no questions, categorical feedback

Multiple choice

Select multiple options:

Question: "Which features are most important to you?"
Options:
☐ Session replay
☐ Feature flags
☐ Experiments
☐ Analytics

Use for: Feature prioritization, use case discovery, preference mapping

Rating scale

Numeric or emoji rating:

Question: "How likely are you to recommend us?"
Scale: 0 [===|====] 10

Or emoji: 😞 😐 😊 😍

Use for: NPS, CSAT, feature satisfaction, sentiment

Link

Display a link with call-to-action:

Message: "Want to learn more about this feature?"
Button: "Read the guide →"
Link: https://docs.yoursite.com/feature-guide

Use for: Documentation links, booking demos, community invites

Creating a survey

Choose a template or start blank

PostHog provides templates for common surveys:

NPS: Net Promoter Score (0-10 scale)
CSAT: Customer Satisfaction (1-5 scale)
PMF: Product-Market Fit (“How would you feel if you could no longer use…”)
Feedback: Open-ended feedback request

Configure questions

Add 1-3 questions maximum for best completion rates:

// Example: NPS with follow-up
[
  {
    type: 'rating',
    question: 'How likely are you to recommend us?',
    scale: 10,
    display: 'number',
    lowerBoundLabel: 'Not likely',
    upperBoundLabel: 'Very likely'
  },
  {
    type: 'open',
    question: 'What could we improve?',
    optional: true  // Make follow-ups optional
  }
]

Set targeting

Control who sees the survey and when:

URL targeting: Show on specific pages
User properties: Target by plan, signup date, etc.
Feature flags: Link to a feature flag
Wait period: Don’t re-survey users too quickly

Customize appearance

Match your brand:

Background color
Button color and text
Position (top/bottom, left/right)
Custom CSS for advanced styling

Targeting options

Show surveys to the right users at the right time:

URL targeting

Show surveys on specific pages:

// Show on pricing page
url: '/pricing'
urlMatchType: 'icontains'  // Contains, case-insensitive

// Show on all blog posts
url: '/blog/'
urlMatchType: 'icontains'

// Don't show on checkout
url: '/checkout'
urlMatchType: 'is_not'

User property targeting

Target by user attributes:

// Identify users with properties
posthog.identify('user_123', {
  plan: 'pro',
  signup_date: '2024-01-15',
  has_team: true
})

Then target survey to users where:

plan = 'pro'
signup_date < 30 days ago
has_team = true

Wait periods

Avoid survey fatigue:

Seen any survey: Wait 30 days after user sees any survey
Completed this survey: Don’t re-ask the same survey
Session frequency: Only show once per session

Feature flag targeting

Link to a feature flag:

// Link survey to feature flag ID
linked_flag_id: 123

// Only show to specific variant
conditions: {
  linkedFlagVariant: 'test'
}

Survey appears only to users who see the feature flag.

Common survey patterns

NPS survey

Question 1: “How likely are you to recommend us?” (0-10 rating)Question 2: “What’s the main reason for your score?” (open text, optional)Targeting: All users 30+ days after signup

CSAT survey

Question 1: “How satisfied are you with [feature]?” (1-5 rating or emoji)Targeting: Users on specific page, 10 seconds after page load

PMF survey

Question: “How would you feel if you could no longer use [product]?”Options:

Very disappointed
Somewhat disappointed
Not disappointed

Targeting: Active users (used product 3+ times)

Feature feedback

Question 1: “Did this feature solve your problem?” (Yes/No)Question 2: “What would make it better?” (open text)Targeting: Users who used the feature, via feature flag

Analyzing responses

View and analyze survey results:

Response overview
Individual responses
Export data

See aggregate results:

Total responses and response rate
Average rating (for rating questions)
Response distribution (for choice questions)
Responses over time chart

Export for deeper analysis:

// Export via API
const responses = await posthog.api.get(
  `/api/projects/${projectId}/surveys/${surveyId}/responses`
)

// Or download CSV from UI

NPS scoring

PostHog automatically calculates NPS:

Promoters (9-10): Loyal enthusiasts
Passives (7-8): Satisfied but unenthusiastic
Detractors (0-6): Unhappy customers

NPS Score = % Promoters - % Detractors Example: 60% promoters, 30% passives, 10% detractors = NPS of 50

Good NPS varies by industry. SaaS typically sees 30-40. Above 50 is excellent.

Survey best practices

Keep surveys short (1-3 questions max)

Users are trying to use your product, not take a survey. One focused question gets better completion than five scattered questions.

Make follow-ups optional

First question required, rest optional. You’ll get more responses and those who answer follow-ups are more engaged.

Time surveys appropriately

After activation: Wait until users experience core value
After feature use: Ask about specific features right after use
Before churn: Survey users showing churn signals

Don’t survey on first visit or during checkout.

Combine with session replay

Watch replays of users who gave negative feedback. See what frustrated them before they answered.

Close the loop

Respond to user feedback, especially detractors. Show users you’re listening and taking action.

Test survey appearance

Preview surveys before launching. Check mobile responsiveness, color contrast, and positioning.

Integration with analytics

Connect surveys to product data:

Filter insights by survey response (e.g., behavior of promoters vs detractors)
Create cohorts from survey responses
Track survey completion rate as a metric
Correlate NPS scores with feature usage

// Create cohort of promoters (NPS 9-10)
const promoters = {
  name: 'NPS Promoters',
  filters: {
    properties: [{
      key: '$survey_response',
      value: [9, 10],
      operator: 'in'
    }]
  }
}

// Compare promoter behavior to detractors in analytics

Advanced: Programmatic surveys

Show surveys programmatically:

// Get available surveys for this user
const surveys = posthog.getActiveMatchingSurveys()

// Show specific survey
const targetSurvey = surveys.find(s => s.name === 'Onboarding feedback')
if (targetSurvey) {
  // Trigger survey display
  posthog.capture('survey shown', {
    $survey_id: targetSurvey.id
  })
}

// Or dismiss programmatically
posthog.capture('survey dismissed', {
  $survey_id: targetSurvey.id
})

Survey templates

PostHog provides pre-built templates:

NPS: Net Promoter Score with follow-up
PMF: Product-Market Fit assessment
CSAT: Customer Satisfaction rating
Feature feedback: Gather thoughts on specific features
Churn prevention: Understand why users might leave
Onboarding: Check if setup was smooth

Get Started

Core Products

Data & Infrastructure

Integration

Configuration

How surveys work

Survey types

Question types

Creating a survey

Targeting options

Common survey patterns

NPS survey

CSAT survey

PMF survey

Feature feedback

Analyzing responses

NPS scoring

Survey best practices

Integration with analytics

Advanced: Programmatic surveys

Survey templates

Build docs developers (and LLMs) love

Get Started

Core Products

Data & Infrastructure

Integration

Configuration

​How surveys work

​Survey types

​Question types

​Creating a survey

​Targeting options

​Common survey patterns

NPS survey

CSAT survey

PMF survey

Feature feedback

​Analyzing responses

​NPS scoring

​Survey best practices

​Integration with analytics

​Advanced: Programmatic surveys

​Survey templates

Build docs developers (and LLMs) love

How surveys work

Survey types

Question types

Creating a survey

Targeting options

Common survey patterns

Analyzing responses

NPS scoring

Survey best practices

Integration with analytics

Advanced: Programmatic surveys

Survey templates