PostHog
FreemiumThe only all-in-one platform built for engineers
Scores
About
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
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
Possible Stacks
SaaS Starter + PostHog
ProjectA production-ready SaaS foundation with product analytics built in from day one. Next.js and Supabase handle the app and auth layer; PostHog adds product analytics, feature flags, session replay, and A/B testing — replacing Amplitude, LaunchDarkly, and FullStory with a single open-source platform.
PostHog Self-Hosted
InfrastructureSelf-host PostHog for full-stack product analytics with complete data ownership. PostgreSQL handles transactional storage for user and event metadata; ClickHouse powers fast analytical queries over large event volumes. Kafka buffers the event ingestion pipeline between capture and ClickHouse; Redis provides the in-memory caching layer. Docker Compose orchestrates the full multi-service deployment.
Related Tools
Works well with (7)
Integrates with (2)
Alternative to (2)
Learning Resources
No resources yet — check back soon.
