PostHog
FreemiumThe only all-in-one platform built for engineers.
Published 29 May 2026 · Last updated 27 September 2026
Scores
Popularity4/5
PostHog has over 34,000 GitHub stars and a rapidly growing community of engineering-led teams. It is frequently cited in developer communities (Hacker News, Reddit) as the de-facto open-source alternative to Amplitude and Mixpanel. Strong Y Combinator backing and active open-source contributions signal broad adoption, though enterprise penetration is still growing relative to legacy analytics vendors.
Learning Curve3/5
PostHog's quickstart and autocapture make initial setup straightforward for developers, and prebuilt dashboards reduce the barrier to first insights. However, unlocking advanced analytics requires learning HogQL (a SQL dialect), and the breadth of features — analytics, flags, replays, experiments, pipelines — creates a steep ramp for teams that want to use the platform fully. Non-technical product managers often find the interface complex and require hand-holding from engineers.
Flexibility5/5
PostHog exposes SQL-based querying, unlimited custom events, extensive SDK coverage, 60+ data pipeline integrations, a public API, and full self-hosting support. Teams can build custom dashboards, define complex cohorts, and join PostHog data with external sources in the built-in data warehouse. The platform functions as a composable data stack rather than a fixed-schema SaaS tool.
Performance4/5
Cloud deployments handle millions of events per month with fast query response times and reliable uptime. Real-time alerting and dashboard refresh rates are responsive for typical production workloads. Self-hosted deployments are recommended only up to ~100K events/month on hobby setups, beyond which cloud is more appropriate. Some users report occasional query slowdowns on very large datasets.
Portability5/5
Released under the MIT license with a fully self-hostable architecture, PostHog gives teams complete ownership of their event data. Migration paths exist for moving between cloud and self-hosted deployments, and data can be exported via the warehouse or pipeline integrations. No proprietary data formats or API lock-in — teams can switch tooling without losing historical data.
About PostHog
PostHog is a comprehensive open-source platform designed for engineering teams to understand, test, and improve their products. It unifies product analytics, web analytics, session replay, feature flags, A/B experimentation, surveys, error tracking, and a customer data pipeline into a single integrated stack — eliminating the need to wire together multiple specialized tools.
The platform is accessible through two primary deployment models. PostHog Cloud is a fully managed SaaS offering hosted on US or EU regions, suitable for teams that want to get started quickly without infrastructure overhead. For organizations with data sovereignty or compliance requirements, PostHog can be self-hosted on Docker, Kubernetes, or bare metal using the open-source codebase licensed under MIT.
PostHog is built around an event-driven data model. Teams instrument their applications with SDKs for web, mobile, and server-side environments, then query behavior using the platform's HogQL interface — a custom SQL dialect — or through prebuilt dashboards. Autocapture reduces manual instrumentation by automatically recording common user interactions.
Core use cases include understanding user behavior at the funnel and cohort level, diagnosing UX issues through session recordings, safely rolling out features to subsets of users via feature flags, and running statistically rigorous experiments. The integrated data warehouse allows joining PostHog event data with external sources like Stripe, Salesforce, or custom data pipelines for richer analysis.
PostHog is particularly popular with engineering-led product teams and technical founders who want a single source of truth about user behavior with full data ownership and flexible infrastructure options.
Key Features
- Product analytics with autocapture and custom event tracking
- Session replay with network monitoring and rage-click detection
- Feature flags for safe rollouts with cohort targeting
- A/B experimentation with Bayesian and frequentist statistics
- Error tracking with real-time alerts
- SQL-based querying via HogQL
- Customer data pipelines with 60+ integrations
- Self-hosting option with full data ownership
Pros
- All-in-one platform eliminates the need to integrate multiple tools
- Generous free tier covers most small-to-mid teams
- Open-source and self-hostable for full data control and compliance
- Developer-friendly with autocapture, SDKs, and SQL access
- Active product team with fast release cadence and responsive support
- Session replay is highly effective for diagnosing UX issues
Cons
- Extensive feature set can feel overwhelming for non-technical users
- Advanced analytics require SQL (HogQL) knowledge
- Self-hosting at scale demands significant infrastructure expertise
- Event-based pricing can become unpredictable at high volumes
- Dashboard navigation is complex with many overlapping features
- Documentation has gaps that create friction during implementation
PostHog Pricing
Freemium- · 1 project, 1-year data retention
- · Unlimited team members
- · 1M analytics events/month
- · 5K session recordings/month
- · No fixed monthly fee — pay only for usage above free tier limits
- · 6 projects, 7-year data retention
- · Analytics from $0.00005/event
- · Session replay from $0.005/recording
- · Custom pricing (approx. $2,000/month minimum)
- · HIPAA BAA and SSO enforcement
- · SAML and RBAC
- · Priority support and dedicated account management
Tech Stacks with PostHog
MERN Stack
ProjectMongoDB, Express, React, Node.js: the classic JavaScript full-stack. A well-trodden path with massive community support and job-market demand.
Databases
Authentication
Hosting
Frontend
Programming
Supabase + Next.js
ProjectThe fastest way to launch a modern web app. Next.js handles the frontend and API routes; Supabase provides Postgres, auth, realtime, and storage out of the box.
Python Web (FastAPI + React)
ProjectA clean separation of concerns: React on the frontend, FastAPI serving a typed REST API, and PostgreSQL for persistent storage (MySQL, MariaDB, and serverless Postgres hosts are also available). Docker keeps environments consistent.
Tools Related to PostHog
Integrates with PostHog(3)
PostHog publishes official Bubble docs and an official PostHog Analytics plugin in Bubble's plugin store — install it, add the API key, and events and session replays start flowing.
PostHog has an official Stripe data pipeline integration that imports revenue events, subscriptions, and payment data into PostHog, allowing product teams to correlate user behavior with revenue metrics in a single dashboard.
PostHog integrates with Sentry so that session replay recordings are linked directly inside Sentry error reports — engineers can jump from a Sentry exception to the exact PostHog session replay where the error occurred.
Alternatives to PostHog(4)
PostHog and Umami both offer web analytics and event tracking, but PostHog is a full product analytics platform with feature flags, session replay, A/B testing, and error tracking, while Umami focuses on simple privacy-first traffic reporting.
PostHog and Plausible Analytics address overlapping web analytics use cases but differ significantly in scope — Plausible is intentionally minimal and privacy-first, while PostHog is an all-in-one product platform with session replay, feature flags, and experimentation.
Google Analytics is the free, closed-source incumbent that PostHog competes with for web/product analytics — teams pick PostHog for self-hosting, SQL-level access, and an all-in-one product suite, and Google Analytics for its Ads/Search Console integration and market-standard reporting.
PostHog offers full product analytics (flags, replay, experimentation) that Vercel Analytics doesn't attempt; teams that only need basic page-view and event tracking on a Vercel-hosted project can skip PostHog's extra setup entirely.
Vendor
PostHog
Website →Tags
Details
- Maintained
- Yes
- Monitoring
- Product Analytics, Session Replay, Error tracking, Feature Flags, A B Testing
- Hosting
- Cloud & Self-hosted
- Open source
- Yes
- Alerting
- Yes
- GitHub stars
- 39.9k
- Stars updated
- 2026-09-23