KeenVerdict

Review · Marketing & SEO

PostHog review

A product platform whose experiments run on its feature flags, alongside analytics, session replay and surveys.

By the KeenVerdict research desk · Updated November 9, 2026

Features (40%)
8.1
Ease of use (30%)
7.8
Value (30%)
9.2

PostHog treats experiments as part of a wider product toolkit. Each experiment creates a feature flag, so experiments are billed with flags, and the free tier includes 1 million flag requests every month along with 1 million analytics events and 5,000 session recordings. PostHog says 97% of companies use it for free, and paid usage is pay as you go with billing limits. Test types include A/B, A/B/n, holdout, fake door and redirect tests, with Bayesian and frequentist engines, funnel and ratio metrics, a shared metrics library and session replays for each variant, which helps explain why a variant won. Experiments can be created from code editors through the PostHog MCP server. For CRO teams the weak spot is the marketer workflow: no-code website experiments are listed as beta, so most tests need some code. Teams that already use PostHog for analytics get experiments with almost no extra setup.

Who it is for

Startups and product teams that want A/B tests next to analytics and session replay on a free tier.

Price

Free every month up to 1M feature flag requests, which also cover experiments; pay as you go above the free tier.

97% of companies use PostHog for free.
Quoted from the vendor pricing page on November 9, 2026.

Key features

  • A/B, A/B/n, holdout, fake door and redirect tests
  • Bayesian and frequentist engines with primary and secondary metrics
  • Funnel, count, value, ratio and retention metrics
  • Session replays for each variant
  • No-code experiments in beta and an MCP server for AI editors

Strengths and weaknesses

  • Strength: Free monthly allowance with no credit card required
  • Strength: Experiments, analytics and replays share one setup
  • Strength: Billing limits on most products to avoid surprise bills
  • Weakness: No-code website experiments are still in beta
  • Weakness: Usage-based pricing makes budgets harder to predict
  • Weakness: Built for engineers more than for marketing teams

Key facts

  • Free tier: 1M feature flag requests per month, experiments billed with flags
  • Seats: Unlimited team members
  • License: MIT licensed open source version available

Where it ranks

PostHog is No. 6 in our Best A/B Testing Tools verdict.

Alternatives we scored

Visit posthog.com · Questions or corrections: editorial@keenverdict.com