PostHog bundles product analytics, session replay, feature flags, experiments, error tracking, surveys, data warehouse, pipelines, LLM analytics, logs, workflows, and AI assistants behind one project model. Teams adopt it to reduce vendor sprawl: the same events power dashboards, replay, and flag targeting.
PostHog’s pricing page (posthog.com/pricing, March 2026) emphasizes monthly free allotments per product on all plans, then pay-as-you-go beyond those caps with optional billing limits per product. Representative free tiers include 1M product analytics events, 5k session recordings, 1M feature flag requests, 1.5k survey responses, 100k error-tracking exceptions, 50 GB logs ingested, 10k workflow messages per channel, 1M managed warehouse rows (plus free historical sync positioning), 10k pipeline trigger events, 100k LLM analytics events, and 2k PostHog AI credits—always confirm the live table.
After free tiers, public usage rates include product analytics from about $0.00005 per event (with volume steps down into the hundred-million range), session replay from about $0.005 per recording on the first paid tranche, feature flag requests from about $0.0001 per request after 1M free, logs from $0.25 per GB after 50 GB free (then $0.15/GB at higher ingest), and error tracking from about $0.00037 per exception after 100k free. Platform packages (Boost, Scale, Enterprise add-ons) add RBAC, SSO enforcement, and support tiers at $250, $750, and $2,000/month list prices on the calculator section.
Self-hosting remains MIT open source if cloud billing is a blocker—ops cost shifts in-house. Compare Amplitude or Mixpanel for analytics depth without flags; compare LaunchDarkly for standalone enterprise feature management; compare Sentry if errors-only workflow is the priority.