PostHog

All-in-one product analytics with session replay, flags, and experiments

Freemium Some setup needed Web · API

About

Track events, watch replays, and ship flags and experiments from one stack instead of stitching five tools. Engineering-led product teams use it to analyze usage, debug issues, and roll out features while keeping data in a single place. Usage-based pricing with generous free tiers makes it easy to start and scale without a sales call.

Editor's Take

Best suited for engineering-led product teams that want an all-in-one stack (analytics, replays, flags, experiments) and either need self-host control or prefer generous free tiers before scaling. Expect some setup and operational work—PostHog is powerful when teams can invest engineering time.

Key Features

Run `npx @posthog/wizard` → install and configure PostHog from a single prompt
Capture product events → build usage analyses (funnels, cohorts, retention) in one place
Turn on a feature flag for a cohort → gradual rollouts and A/B tests with measured impact
Open a user’s session recording → pinpoint UX issues tied to specific events
Sync product data to the managed warehouse → query millions of rows with pay-per-row pricing

Use Cases

A product engineer rolling out a new checkout flow behind flags and validating results with an experiment
A growth PM investigating a drop in activation by watching 20 session replays and checking cohort retention
A data analyst centralizing web and app events into the managed warehouse to answer ad-hoc product questions

Try It Like This

1
Quick local install with wizard
Run npx @posthog/wizard → follow prompts to install a local PostHog instance and instrument a dev site with one line of JS → open the dashboard and confirm events appear in real time.
2
Roll out a checkout change behind flags
Create a feature flag for the new checkout component → target a 10% cohort and enable gradual rollout with metrics to measure success → run an experiment and compare cohort conversion & retention in PostHog.
3
Debug drop in activation with session replay
Build an activation cohort and filter users who failed the flow → open several session recordings linked to those events → correlate UX problems seen in replays with funnel drops and make a fix.
4
Sync product events to a warehouse
Enable the managed warehouse sync for your project → let PostHog export event rows to the warehouse on pay-per-row terms → run SQL queries on millions of rows to answer ad-hoc product questions.
5
Self-host and customize pipelines
Deploy PostHog self-host using the recommended commands or Docker setup → configure custom event capture and plugin code for data enrichment → monitor resource usage and scale components as traffic grows.

Pros & Cons

Pros

Open-source with a self-host option, letting teams keep data on their own infrastructure or run the managed cloud.
Bundles analytics, session replay, feature flags, experiments, heatmaps and more in one product so teams don’t need separate Mixpanel/Hotjar/LaunchDarkly stacks.
Developer-friendly setup including npx @posthog/wizard and built-in warehouse sync to export event rows for SQL analysis.

Cons

Has a steeper learning curve and requires engineering effort to self-host, instrument events correctly, and maintain at scale.
Usage-based (pay-per-row/usage) pricing and high event volume can make costs harder to predict as you scale.
Self-hosting adds operational overhead (infrastructure, scaling, backups) compared with fully managed SaaS alternatives.

Getting Started

1 Create an account at app.posthog.com or paste `npx @posthog/wizard` into your terminal
2 Add the snippet/SDK to your app and enable the products you need (analytics, replay, flags)
3 Send your first events and open a dashboard or replay to verify data within minutes

Pricing

Product	Free tier	Pricing
Product Analytics	1 million events/mo	$0.00005/event
Session Replay	5,000 recordings/mo	$0.005/recording
Feature Flags	1 million requests/mo	$0.0001/request
Managed warehouse	1 million rows/mo	$0.000015/row