Loading tool details…
Complete pricing guide for PostHog. Compare all plans, analyze costs, and find the perfect tier for your needs.
Not sure if free is enough? See our Free vs Paid comparison →
Still deciding? Read our full verdict on whether PostHog is worth it →
Pricing sourced from PostHog · Confirm current pricing and terms with the vendor.
A check mark means listed in that plan’s record. A dash means not listed, rather than confirmed unavailable. Verify current vendor terms.
Detailed feature comparison coming soon. Visit PostHog's website for complete plan details.
View Full Features →PostHog bundles product analytics with session replay, feature flags, A/B experiments, surveys, error tracking, and LLM analytics — all sharing the same event stream and identity graph. Mixpanel and Amplitude focus almost exclusively on analytics; you'd pair them with LaunchDarkly, FullStory, and Sentry to get the same surface. PostHog is also MIT-licensed with a self-hostable version, which neither Mixpanel nor Amplitude offers. The tradeoff is that Mixpanel and Amplitude have deeper cohort and journey analysis for enterprise usage, while PostHog trades some analytics depth for breadth and a lower overall bill.
Yes. The free tier includes 1M events, 5,000 session replays, 1M feature flag requests, and 100k errors per month — enough to run a small-to-mid production SaaS without paying. There's no seat cap, so your whole team can log in. You start paying only when you exceed a specific product's free allotment, and each product meters separately. Most early-stage teams live on the free tier for their first year.
PostHog's LLM analytics wraps OpenAI, Anthropic, and other providers via a small SDK layer and logs traces, model, tokens, cost, and latency alongside product events. It's not as deep as Langfuse or LangSmith for prompt-management or evaluation workflows, but it's the easiest way to have LLM metrics in the same dashboards as user-facing product metrics. Many teams use PostHog for product+LLM cost visibility and pair it with a dedicated eval tool for prompt iteration.
Self-hosted PostHog is MIT-licensed and includes all product features, but you're responsible for Kubernetes, ClickHouse, Postgres, Redis, Kafka, and upgrades. Cloud handles all of that plus scaling for a per-event fee. Teams pick self-hosted for data-residency or compliance reasons; most teams save engineering time by starting on Cloud (US or EU) and only migrate off if their event bill exceeds their SRE cost.
Yes — PostHog ships an official Model Context Protocol server so assistants like Claude and Cursor can query analytics, inspect feature flags, and pull session replays as tools. Combined with Max AI (PostHog's in-app assistant), this lets AI-assisted workflows read product data and answer questions grounded in real analytics rather than guesses.
AI builders and operators use PostHog to streamline their workflow.
Try PostHog Now →Mixpanel: Advanced product analytics platform to analyze user behavior, optimize conversion funnels, and improve retention with event-based tracking.
Compare Pricing →Product analytics platform that combines natural language AI queries with behavioral cohort analysis, enabling teams to ask complex questions in plain English while building precise user segments based on actual behavior patterns.
Compare Pricing →Auto-capture product analytics platform (now part of Contentsquare) that retroactively tracks all user interactions without manual instrumentation for instant behavioral insights.
Compare Pricing →