PROVIDER - Open-source experience management

Formbricks

Formbricks is a German open-source experience management platform under AGPLv3, covering in-app, link and website surveys, and self-hosting keeps survey data inside your own database.

Form & Survey Germany 2026.11.20
3.8pts

Detailed Notes

Pros 5 items

Released under AGPLv3 and self-hostable with Docker, keeping survey definitions, responses and user attributes inside your own Postgres instance
Ships 3 delivery modes - in-app surveys, link surveys and website surveys - and prices on responses and features rather than on the number of forms
Official SDKs cover JavaScript, React and React Native, so 1 snippet is enough to trigger surveys from user attributes and events
Targeting rules combine attributes, events and URLs, so 1 survey can run multi-page flows and follow-ups while responses aggregate in the dashboard in real time
The cloud runs in an EU region and both self-hosted and cloud deployments share 1 codebase, with public data processing terms

Cons 4 items

Self-hosting means maintaining Postgres, auth secrets and upgrades yourself, while the cloud bills on response tiers that climb steeply at volume
The editor and template ecosystem are younger than commercial SaaS, so some question types and cross-tab analysis still need custom components or a BI export
There is no Chinese interface or documentation, onboarding needs English material, and access to the cloud plus static assets from mainland China is unstable
In-app surveys depend on the client SDK, so blocked or uninitialised scripts mean no survey fires and debugging spans both the frontend and the trigger log

Service Overview

Formbricks is a German open-source experience management platform released under AGPLv3. Its premise is that asking users questions should be a built-in product capability rather than a periodic email campaign. Within the Forms and Surveys category, it stands out by covering in-app surveys, link surveys and website surveys at once, while the whole system can be self-hosted so data never leaves your own database.

The architecture is a client SDK, a hosted editor and your own database. Drop the script into a product and surveys can be triggered by user attributes, events and page URLs; content is edited visually in the dashboard and responses land in Postgres. For SaaS teams this means seeing who dropped off at which step and which plan scores lowest, instead of a pile of anonymous answers.

The open-source licence has a second face. Self-hosting means owning database maintenance, secrets, email delivery and upgrades, while the cloud version meters responses. It fits teams with engineering capacity, sensitivity about data location, and a need for a feedback channel that lives inside the product for years.

Key Strengths

  • Data stays under your control: AGPLv3 and Docker self-hosting keep survey definitions, responses and attributes in your own Postgres, with backups and migrations independent of any vendor.
  • Three delivery modes cover the whole loop: In-app surveys watch behaviour, link surveys reach external users, and website surveys intercept visitors, all managed from one dashboard instead of three tools.
  • Developer-friendly SDKs: Official packages for JavaScript, React and React Native let you trigger surveys by user id and event, with logic and content edited in the dashboard rather than shipped in a release.
  • Targeting and follow-ups compose: Attributes, events and URLs stack into trigger conditions, and responses can chain follow-up questions, which suits churn-reason and feature-satisfaction research that needs two or three turns.
  • Controllable hosting region: The cloud runs in an EU region with public data processing documentation, which is clearer for GDPR contexts than most survey SaaS vendors offer.

Product Ecosystem

Platform capabilities

The dashboard is organised as surveys, triggers and analysis. The editor supports multiple question types, multiple pages, branching and brand styling. Triggers combine attributes, events and URLs with frequency caps so one user is not asked repeatedly. Analysis aggregates responses by survey and timeline, sliced by user attributes.

Analysis is deliberately basic next to commercial SaaS. Cross-tabulation, cohort comparison and export dimensions are narrower than on dedicated research platforms, so heavy statistics usually move to a BI tool after export.

Integrations

Webhooks push responses to your own service or an automation platform, and notification channels keep a backlog visible. User identity travels through the SDK's setUserId and setAttribute calls, which makes mapping to your own user table a prerequisite: without a solid identity map, segmentation and follow-ups drift.

Typical scenarios

Feature feedback and churn research in a SaaS product, in-app NPS and CSAT, visitor intent surveys on a marketing site, waitlists and user interview recruitment are the usual fits. Large-scale external market research is not: quota control and sampling are stronger on dedicated platforms.

Limitations

  • Self-hosting is a long-term commitment: Postgres, key rotation, email delivery and upgrades are yours to run, and without spare operations capacity the hidden cost exceeds a cloud subscription.
  • Advanced features are still catching up: Fewer question types and templates than mature commercial tools, and cross-tab analysis, quotas and advanced sampling require extra work.
  • No Chinese support: Interface, documentation and community discussion are English-first, which lengthens onboarding for teams in China, and access to the cloud from mainland China is inconsistent.
  • In-app delivery depends on the client: Blocked scripts, missing initialisation or a user with JavaScript disabled mean no survey, so debugging spans frontend logs and the trigger history together.

Use Cases

  • SaaS feedback and churn research (★★★★★): Behaviour-triggered surveys bind answers to user ids, so findings point at specific accounts and plans
  • In-app NPS and CSAT (★★★★☆): Short surveys embedded in the product with controlled frequency do not turn into popup noise
  • Teams with data sovereignty needs (★★★★☆): Self-hosting keeps data inside the network, which suits finance, healthcare and public sector contexts
  • Visitor intent research on a website (★★★★): Website surveys intercept before exit and capture why visitors left
  • Large external market research (★★☆): Sampling, quotas and advanced statistics are limited, and dedicated research platforms fit better

Pricing

Item Price Notes
Self-hosted community edition $0 Full AGPLv3 source, server, database and operations costs are yours
Cloud free tier $0 Monthly response cap, upgrade needed beyond it, limits follow the official site
Cloud paid tiers Monthly subscription, response-based Higher response caps, white-label and collaboration, unit price steps with volume
Enterprise plan Sales quote SSO, dedicated support and custom terms, usually annual

Prices and response caps follow the official pricing page. Budget along two paths: self-hosting as server, database and operations labour converted to an annual cost, or cloud as monthly responses times the matching tier. With modest volume and no spare operations capacity, cloud is usually cheaper.

FAQ

Does self-hosting match the cloud feature set?

Core survey, trigger and collection features match, with differences in hosting, billing and enterprise add-ons such as SSO and dedicated support. Self-hosting means maintaining the database and upgrades yourself, while the cloud absorbs that work and meters responses.

Will in-app surveys annoy users?

Trigger conditions and frequency caps control that. Trigger on events rather than dwell time and set a minimum interval per user; asking too often erodes both response rates and product sentiment at the same time.

What does compliance require?

With self-hosting, data location is your decision, and the privacy policy must state what is collected and how long it is retained. On the cloud, confirm the hosting region and data processing agreement. Either way, avoid passing unnecessary sensitive fields through SDK attributes.

How is this different from Typeform or SurveyMonkey?

Those are finished external survey products, strong on question types and statistics. Formbricks is strong on in-app delivery, identity binding and self-hosting. Use it when the feedback channel lives inside your own product; use a finished SaaS when you just need to send a survey out once.

Code Examples

Initialise the SDK in an app and identify the user

// Initialise at app start, then attach identity after login; both ids come from the dashboard
(function () {
  var s = document.createElement("script");
  s.src = "https://app.formbricks.com/js/formbricks.umd.cjs";  // swap for your own domain when self-hosting
  s.async = true;
  s.onload = function () {
    window.formbricks.init({
      environmentId: "$FORMBRICKS_ENVIRONMENT_ID",  // injected by the build environment, never hardcoded
      apiHost: "$FORMBRICKS_API_HOST",
    });
  };
  document.head.appendChild(s);
})();

// Attach identity after a successful login so surveys map to a real user
function identify(user) {
  window.formbricks.setUserId(user.id);
  window.formbricks.setAttribute("plan", user.plan);
  window.formbricks.setAttribute("signupDays", user.signupDays);
}

Do not call setUserId before initialisation. Calls made while the script is still loading fail silently, and the dashboard ends up full of anonymous responses.

A minimal self-hosted compose

# Illustrative minimum, check image tags, variable names and init steps in the official repo compose file
services:
  postgres:
    image: postgres:15-alpine
    environment:
      POSTGRES_USER: formbricks
      POSTGRES_PASSWORD: ${POSTGRES_PASSWORD}
      POSTGRES_DB: formbricks
    volumes:
      - pgdata:/var/lib/postgresql/data

  formbricks:
    image: ghcr.io/formbricks/formbricks:latest
    depends_on:
      - postgres
    environment:
      WEBAPP_URL: ${WEBAPP_URL}
      DATABASE_URL: postgresql://formbricks:${POSTGRES_PASSWORD}@postgres:5432/formbricks
      NEXTAUTH_SECRET: ${NEXTAUTH_SECRET}
      ENCRYPTION_KEY: ${ENCRYPTION_KEY}
    ports:
      - "3000:3000"

volumes:
  pgdata:

Inject secrets and the encryption key from .env rather than the compose file; rotating ENCRYPTION_KEY makes already-encrypted sensitive fields unreadable.

Community & Resources

Official channels

Learning resources and tools

Visit Formbricks Website → https://formbricks.com/
Visit Website ↗