PostHog
Analytics · Freemium
Overview
A product analytics platform that shows how people use an app or website. It was founded in 2020 and is open source. Its pitch is that one tool covers jobs teams often buy separately: product analytics, session recordings, feature flags, A/B experiments, surveys, error tracking, and a built in data warehouse. Teams can use the hosted cloud version or run it on their own servers. It is aimed mostly at engineers and product teams at startups and growing companies. The company publishes an unusually detailed public handbook about how it works.
What people use it for
Product teams use PostHog to track which features get used, build funnels, and measure retention and drop off. They watch session recordings to see where users get stuck. They roll out features to a subset of users with feature flags before a full launch. They run A/B experiments and measure the effect on a chosen metric. They collect user feedback with in product surveys. They catch and group front end errors with error tracking. They also query event data directly with SQL and join it to data pulled in from other sources through the warehouse feature.
Key capabilities
PostHog combines product analytics, session replay, feature flags, experiments, surveys, error tracking, and a SQL data warehouse in one platform. Analytics covers events, funnels, retention, cohorts, and paths. Session replay records web and mobile sessions with masking options. Feature flags support percentage rollouts and targeting rules, and experiments build on top of them. The product is open source and the full code is on GitHub. It can be self hosted, though the open source deployment is limited to one project and leaves out features such as experimentation and advanced permissions. Pricing is usage based per product with a monthly free allowance for each: the free tier includes one million events, 5,000 recordings, one million feature flag requests, and more, and it resets monthly. The company states most customers stay on the free tier.
Limitations
PostHog now steers self hosting toward small, hobby scale setups. The open source version allows only one project and drops experimentation, advanced paths, and advanced permissions. Self hosted users get no commercial support and miss features that ship continuously to the cloud. Reviewers report the setup has silent failures the docs do not cover. On the product side, common complaints are navigation and UI rough edges, privacy controls in replay, and a learning curve for less technical team members. Each bundled tool is a little less polished than the specialist product it replaces.
Insight
PostHog is a good deal when a team wants analytics, recordings, flags, surveys, and experiments and does not want five separate bills and five sets of tracking code. The trade is that no single piece is best in class. The analytics is solid but less refined than Amplitude, the replay less polished than a dedicated tool, and so on. For most teams that is an acceptable trade, especially early. Self hosting deserves a warning: PostHog has quietly moved away from supporting large self hosted deployments, so the open source route is now really for small internal or hobby use, not a full production replacement for the cloud. The usage based free tier is genuinely generous and covers a lot of startups. Best for engineering led teams that want one reasonably good tool covering broad ground and are happy on the hosted version.
Pricing
Freemium
Website
Last checked 2026-08-30