PostHog Review: Open-Source Product Analytics Suite

PostHog

PostHog's all-in-one product platform with analytics, session replay, feature flags, experiments and a data warehouse, with free monthly allowances.

SQL
PostHog Review: Open-Source Product Analytics Suite

Tool Overview

PostHog is a product platform that bundles many tools product teams usually buy separately: product and web analytics, session replay, feature flags, A/B experiments, surveys, error tracking, a data warehouse with SQL, and AI observability for LLM features. All of them share the same event data and the same user profiles, so you can go from a funnel drop-off to the session recordings of the users who dropped, then test a fix behind a feature flag.

PostHog also has an open-source version under the MIT license for organizations that want to run it themselves, while most users choose the hosted cloud product. For data learners, PostHog is interesting because it is unusually friendly to analysts who like SQL: you can query events directly and join them with data pulled in from other sources, instead of being limited to point-and-click reports.

Key Features

  • Product and web analytics — trends, funnels, retention, paths and web traffic reports from one event stream
  • SQL access and data warehouse — query events with SQL and sync external sources (such as payment or CRM data) to analyze alongside product data
  • Session replay — watch recordings of user sessions linked to events and errors
  • Feature flags and experiments — roll out features gradually and run A/B tests billed together with flag requests
  • Surveys — in-product surveys and feedback collection
  • Error tracking and logs — capture exceptions and application logs next to user behavior
  • AI observability and evaluations — monitor and evaluate LLM features in your product
  • Destinations and workflows — send events and data to other tools and trigger messages
  • PostHog AI — an assistant that works across products, available in the web app, Slack, desktop and through an MCP server

Free Tier & Pricing

PostHog uses per-product, usage-based pricing with a free allowance that resets every month and does not expire. On the free plan, the monthly allowance includes:

  • 1M analytics events
  • 5K session recordings
  • 1M feature flag requests (experiments are billed with flags)
  • 100K exceptions in error tracking
  • 1,500 survey responses
  • 1M synced data warehouse rows
  • 10 GB of logs

The free plan includes unlimited team members, one project and one year of data retention, and usage simply stops at the limits so you cannot be charged by surprise. Pay-as-you-go keeps the same free allowance, then bills for usage above it, with six projects and seven-year data retention. You can set a billing limit per product. Platform add-ons cover SSO, audit logs and permissions, and PostHog offers discounts for startups and non-profits. There are no minimums or annual commitments.

Who Is This Tool For?

PostHog suits engineering-led startups and product teams that want analytics, replay, flags and experiments in one place, and analysts who prefer writing SQL over clicking through report builders. The monthly free allowance makes it a good choice for side projects, portfolio apps and early-stage products.

It is a less natural fit for marketing teams that only want simple traffic dashboards, or for organizations that prefer one polished specialist tool for each job. Because there are many products, non-technical users may find the interface denser than a pure analytics tool.

Setup & Learning Curve

Adding the tracking snippet or an SDK starts data flowing quickly, and the free tier needs no credit card. Basic analytics and replays are easy to use. The deeper skills are designing clean events, setting up flags and experiments correctly, and using SQL against the event tables. If you plan to self-host the open-source version, expect real infrastructure work compared with the cloud product.

Pros and Cons

Pros

  • Many tools in one platform sharing the same data
  • Free monthly allowances on every product, with hard stops so there are no surprise bills
  • SQL access and a built-in data warehouse suit analysts
  • Open-source MIT version available
  • Billing limits can be set per product

Cons

  • Usage-based pricing across many products makes total cost harder to predict at scale
  • The breadth of features means a busier interface and more to learn
  • The free plan is limited to one project and short recording retention
  • Self-hosting the open-source version requires engineering effort

FAQ

Is PostHog free? Yes. Every product has a free monthly allowance, such as 1M analytics events and 5K session recordings, and the free plan needs no card.

What happens if I go over the free limits? On the free plan, extra data is dropped. On pay-as-you-go, usage above the allowance is billed.

Can I use SQL in PostHog? Yes. You can query event data with SQL and join it with synced data warehouse sources.

PostHog vs Mixpanel or Amplitude? All three cover core product analytics. PostHog stands out for bundling flags, replay, error tracking and a warehouse, and for its open-source option.

Install PostHog on a side project and explore one funnel with SQL to see whether its all-in-one approach suits you.

Similar listings in category