PostHog Review: The Best Free Analytics Stack?

Score context

PostHog scores 8.8 — the highest in this batch — because it replaces an entire stack of paid tools. Without PostHog, you'd be paying for Mixpanel (analytics) + FullStory (session replay) + LaunchDarkly (feature flags) + Optimizely (A/B testing) separately. That's hundreds of dollars per month. PostHog gives you all of it with a free tier that covers 1M events, 5K recordings, and unlimited feature flags. It lost points because the breadth of features creates a real learning curve, and session replay costs can spike if you record everything without filters.

Pricing Breakdown

PostHog's pricing is famously founder-friendly. The Free tier includes 1M events/month, 5K session recordings/month, unlimited feature flags, unlimited A/B tests, and unlimited surveys. This is genuinely enough for most solo products for 12+ months. No credit card required to start, and the free tier never expires—it's not a trial, it's a permanent offering.

The Pro tier ($450/month) adds analytics features like cohort queries, advanced filtering, and higher event limits. It includes 10K session recordings and the same unlimited feature flags and experiments. Beyond this, pricing becomes event-based: additional events cost $0.00001 per event (about $10 per 1M events), session recordings cost $0.005 per additional recording, and you pay for data retention. Pricing was verified on April 9, 2026.

The real value is in the free tier. Most solo products' traffic doesn't exceed 1M events monthly until they're handling meaningful scale. Session replay can become costly if you record every user session indiscriminately—each recording is captured, adding up quickly. The solution: set recording sampling rules and replay only specific user segments or pages where you're debugging issues.

Feature Deep Dive

Event Analytics & Autocapture: PostHog captures user events automatically without code—button clicks, page views, form submissions. You don't write instrumentation code; the SDK logs events automatically. You then query these events to answer questions like "How many users completed the signup flow?" or "What's the drop-off rate at each step?" Autocapture is powerful because you get data retroactively—you can ask questions about your product before you planned to measure them. Edge case: autocapture logs too much data. A form with live validation triggers events on every keystroke, bloating your quota. The solution: configure autocapture to capture only meaningful interactions.

Session Replay: Watch the exact screen recording of what a user did. This is invaluable for debugging confusing behaviors—you see the user's cursor position, which form fields they struggled with, where they got stuck. Common use case: a user reports "the checkout button doesn't work" and you watch the replay to see they're clicking the wrong element. Edge case: privacy. Recording every user's screen captures form inputs, passwords (if not masked), and potentially sensitive information. PostHog provides masking tools, but misconfiguring them leaks data. Always review your recording settings.

Feature Flags & Rollouts: Feature flags let you enable features for subsets of users. Common use case: you release a new design and want to see how 10% of users respond before rolling it out fully. You toggle a flag in PostHog without redeploying. PostHog makes this easy with targeting rules—enable the feature for all users in New York, or users who've completed checkout at least once. Edge case: flag overload. If you have dozens of flags active simultaneously, reasoning about feature state gets complicated. Good naming and cleanup practices are essential.

A/B Testing & Experiments: PostHog runs A/B tests natively, calculating statistical significance automatically. Describe your hypothesis, set up a test variant, and PostHog tracks conversion rates, tracks changes in metrics, and tells you when results are statistically significant. Common use case: testing different pricing pages or signup flows. Edge case: most SaaS founders don't run rigorous experiments—the sample size is too small or test duration too short for statistical significance. PostHog warns you when this happens, but it's easy to misinterpret inconclusive results.

Getting Started & Setup

Getting PostHog live takes 30 minutes. Visit posthog.com, create a project, copy the snippet into your HTML <head>, and events start flowing. The JavaScript SDK (also available for Python, Go, Node.js, etc.) is straightforward. Within an hour, you have a basic dashboard showing page views, user counts, and conversion funnels.

The first-value moment is immediate—you see user behavior data without writing instrumentation code. The learning curve is steep because PostHog offers so many features. The Insights section (analytics queries) uses a UI that's powerful but not obvious. The Experiments section has good defaults but feels buried. The documentation is extensive and improving, but the breadth of features means you'll spend time exploring to find features relevant to your use case.

Documentation is solid. Posthog has tutorials, example queries, and community slack. The dashboard is self-exploratory—you can usually figure out how to ask a question through the UI. For advanced use cases (cohort analysis, complex segmentation), the documentation provides working examples.

Real Usage Experience

After 3–6 months of PostHog, you've built habits around checking metrics. A/B tests become routine. You catch product issues before users complain—you see that users are hitting a particular error and investigate before they file support tickets. The feature flag workflow changes how you deploy—instead of big releases, you feature-flag new code and roll out gradually.

The surprises surface with session recording costs. If you record every session without sampling, your costs climb unpredictably. A traffic spike can suddenly cost more than expected. The solution: set up recording filters to sample 10% of traffic or record only specific pages. Also, session replay data retention costs money—older recordings are less useful but still cost to store.

The delightful discoveries: PostHog's data is yours. You can export event data, query it in a data warehouse, and build custom analyses. You can self-host PostHog if you want full data control. The team is very builder-friendly—they're scrappy, responsive, and actually use their own product. Updates roll out frequently, and the product direction reflects what early-stage teams need.

Expanded FAQ

Is PostHog worth it for solo founders? Yes, without any qualification. The free tier is so generous you should install it day one. You're leaving money on the table if you're not using it. 1M events/month is an enormous quota for a bootstrapped product.

What's the difference between PostHog and Google Analytics? Analytics measures pageviews and traffic sources. PostHog captures user actions—every button click, form submission, API call. GA is for "how many visitors came to my site." PostHog is for "what did users do after landing." GA is good for SEO insights. PostHog is good for product insights.

Can I use PostHog without accepting legal complexity? Yes. PostHog provides built-in privacy features—you can mask form inputs, exclude events from storage, and host on EU servers. You still need a privacy policy mentioning analytics, but PostHog makes compliance easier than most alternatives.

How do I avoid large session replay bills? Set recording sampling to 10–20% of traffic instead of 100%. Use segment filters to record only specific pages or user cohorts. Disable recording on low-priority pages. Delete old recordings past 30 days. Most solo founders spend less than $20/month on replay with proper configuration.

Strengths and tradeoffs

Who should use PostHog

Any technical founder who wants product analytics, session replay, and feature flags without stitching together multiple services. PostHog is particularly strong if you care about data ownership — you can self-host it, and even on the cloud version, you own your data. The autocapture feature means you get basic event tracking without writing any instrumentation code, which is a fast way to start.

Skip PostHog if you're a non-technical founder who just wants a traffic dashboard. Plausible gives you simple pageview analytics for $9/mo with zero setup. Also skip it if you only need one of its features — if all you want is feature flags, a simple config file or Vercel's built-in flags might be enough.

Alternatives worth considering

Mixpanel

Focused behavioral analytics with strong funnel/retention views.

pricing: Free + usage

Amplitude

Enterprise-grade product analytics and growth features.

pricing: Tiered enterprise

Plausible + warehouse

Lightweight analytics pattern for lean teams.

pricing: Subscription + infra

verdict

PostHog is the best analytics stack for solo founders, full stop. Install it on day one, use the free tier — 1M events/month is more than enough for your first year. Add session replay when you're debugging specific user flows, and feature flags when you're ready to A/B test pricing or features. You won't outgrow the free plan until you have a real business.