Most pricing pages give you tiers. PostHog gives you a calculator. The headline numbers from posthog.com/pricing in May 2026: 1M product-analytics events per month free, then $0.00005 per additional event on the hosted Cloud, plus per-product pricing for session recording, feature flags, surveys, and experiments. Or you can run the entire stack yourself, free, on your own infra. This guide does the event-volume math for both tracks so you can pick correctly the first time.

How we got these numbers. All figures pulled from posthog.com/pricing and the docs at posthog.com/docs in May 2026. PostHog updates per-event prices and free tier limits about twice a year. Always re-check before committing.

Track 1: PostHog Cloud (the hosted SaaS)

Cloud is the default for almost everyone. You sign up, drop the JS snippet on your site, and start getting data within minutes. Pricing is per product (analytics, recording, flags, surveys, experiments), each with its own free allotment and per-unit overage rate.

Product analytics: 1M events free, then $0.00005 per event

That per-event price is small but it adds up: 1 million events of overage = $50. So a SaaS pulling in 2M events/month total ends up at $50/month for analytics. 5M events lands at roughly $200, though sliding-scale discounts kick in earlier in practice.

The thing that surprises people: PostHog charges only above the included tier. There’s no $50/month base fee — if your real volume is 800K events/month, you legitimately pay $0 forever, with full feature access. That’s aggressive compared to Mixpanel (Free plan caps at 100K events/month and has feature limits).

Session recording: priced per recording

Session recording is a separate product with its own meter. The 2026 pricing on posthog.com/pricing is roughly 5,000 recordings free per month, then $0.005 per recording after that, again on a sliding scale. A SaaS recording every session at moderate traffic blows past 5,000 recordings within a couple of weeks. This is the single line item that surprises Cloud customers most often.

Feature flags: free up to a high limit

Feature flags are extremely cheap on PostHog — effectively free up to 1M requests per month, then a small per-request fee. For a typical solo SaaS using flags for staged rollouts, you almost never pay for this product. LaunchDarkly at $20/seat/month for the same use case is dramatically more expensive.

Surveys, experiments, and the rest

Each has its own meter. Surveys are priced per response (a few cents each above a small free quota). Experiments are priced via the events they generate, not separately. For a solo founder, surveys and experiments rarely cost more than a few dollars/month if used intentionally.

Track 2: Self-hosted open source PostHog

The same product, free forever, if you run it yourself. PostHog open source is at github.com/PostHog/posthog under the MIT license for most of the platform (some enterprise features are source-available, not open source). You deploy it via Docker or Kubernetes, point it at a Postgres + ClickHouse cluster, and pay your hosting bill instead of PostHog’s bill.

The catch: PostHog is a real piece of infrastructure. ClickHouse alone is non-trivial to operate. PostHog’s self-hosted docs guide you through it, but you’re signing up for an ongoing maintenance load: backups, version upgrades, scaling the database when event volume grows, watching disk usage. For a solo founder, that’s real time.

Realistic event-volume math by SaaS stage

The single biggest mistake founders make is underestimating event volume. Modern PostHog autocapture — the default config that records every click, page view, and form interaction — generates 5–20× more events than founders expect. Here’s what it actually looks like at common stages.

Pre-launch landing page

A simple marketing site with autocapture on, getting 2,000 visits/month, generates 30,000–80,000 events. Page views, scroll events, button clicks, form interactions. Cost: $0 on Cloud.

$1K MRR SaaS

50–100 paying customers, plus marketing site traffic. Logged-in app sessions are event-heavy — every interaction with a button, form, or chart fires events. Realistic monthly volume: 100,000 to 500,000 events depending on how interactive the product is. Cost: $0 on Cloud, easily.

$10K MRR SaaS

500 customers means 5–10× the in-app activity, plus much heavier marketing site traffic from your growing audience. Realistic volume: 1 to 3 million events per month. Cost: $0 to ~$100/month on Cloud, depending on whether you cross the 1M free line and by how much. Add session recording on top — if you record everything — and you can easily double that.

Cloud cost at $10K MRR (typical)
~$50–$200/month
Product analytics overage + selective session recording. Most of the cost comes from recordings, not analytics events.

When self-hosted is worth it

Cloud is the default. Self-hosted only becomes the right answer when one of these is true:

  • Privacy or compliance requirements. Healthcare, finance, EU data-residency, or government work where customer data legally cannot leave your infrastructure. Self-hosted PostHog keeps everything in your VPC.
  • You’re consistently over 10 million events/month. At that volume, Cloud overages start adding up to thousands per month, and the ops cost of running PostHog yourself becomes economical. You’ll need either devops time or a contractor — we covered the math on outside help in when to hire your first contractor.
  • You already have devops capability. If your team runs Kubernetes for other reasons, adding PostHog is incremental. If PostHog would be your first piece of self-hosted infra, the learning curve dwarfs the savings.

For 95% of solo founders reading this, none of these apply. Cloud is the right call.

When hosted PostHog clearly wins

Cloud is the obvious choice when:

  • You’re under 5 million events per month and have no compliance constraints — the bill stays in the “rounding error” range
  • You don’t want to spend any time on ops — the value of your weekend is more than the Cloud bill
  • You want the latest features without waiting for a self-hosted release — PostHog ships to Cloud first
  • You’re a one-person team — the time to operate a ClickHouse cluster is the most expensive thing you have

Add-ons that change the math

The Cloud bill rarely matches the headline analytics rate because of the per-product add-ons. Three to watch closely:

  • Session recording at ~$0.005 each. If you turn on recording for every session, this becomes the largest line item on your PostHog bill within weeks. The right move is conditional recording — record only logged-in users, only on key flows, or only for a percentage of sessions. PostHog supports all of these.
  • Feature flags. Mostly free at solo-founder volume, but if you call posthog.feature_flags.get_all() on every page load for an unauthenticated marketing site, you can rack up flag-evaluation requests fast.
  • Surveys, priced per response. Cheap individually, but if you launch a survey to your full mailing list at $10K MRR, even pennies-per-response can hit $50–$100 in one campaign.

Hidden volume drivers (what actually fills your event meter)

Three things drive PostHog volume way more than founders realize.

Autocapture is generous by default

PostHog’s autocapture sends events for clicks, form submissions, page views, and rage-click detection. On a moderately interactive marketing site, that’s 10–15 events per page view, not one. Turning off autocapture on marketing pages — and switching to manual posthog.capture() calls for the events you actually care about — can cut your event volume by 80%.

Group analytics multiplies events

If you use group analytics (per-organization tracking for B2B SaaS), every event is associated with both a user and a group. The event count itself doesn’t double, but query volume can, which affects retention pricing on certain product features.

Session recording sample rate

Default is 100% of sessions. If you set it to 10% (still statistically valuable for spotting UX issues), you cut recording costs by 10×. Few founders adjust this and most should.

How PostHog compares to alternatives in 2026

PostHog’s bundle of analytics + recording + flags + surveys is unique in the market. The closest alternatives:

  • Mixpanel. Better cohort analytics UI, dramatically more expensive past the free tier ($25/month minimum on the Growth plan in 2026), and no session recording or feature flags.
  • Amplitude. Strong product analytics, weak everything else. Higher floor than PostHog Cloud.
  • FullStory + LaunchDarkly + a survey tool. The unbundled stack. Costs roughly 5–10× PostHog at the same volume because each tool prices independently.
  • Plausible / Fathom. Cheap and lightweight, but they’re web analytics, not product analytics. They don’t track logged-in user behavior the way PostHog does.

The bundle math is why PostHog dominates the solo-founder analytics segment. We covered the qualitative side in our PostHog review; the broader stack lives in the solo founder tech stack guide. If you’re still figuring out which metrics belong on the dashboard at all, start with SaaS metrics that matter.

Bottom line on PostHog pricing

Use PostHog Cloud, on the Free tier, until you cross 1 million events per month. That moment usually arrives around $5K–$15K MRR for an interactive SaaS. After that, expect $50–$200/month all-in for most solo founders, with session recording typically the largest line item.

Self-hosted PostHog is technically free but operationally expensive for a solo founder. Don’t pick it as a cost-saving move — pick it only when compliance or extreme volume forces your hand. And before either path, take fifteen minutes to disable autocapture on marketing pages and set session recording to 10% sample rate. That single config change is worth more than upgrading any tier.

Related reading

Get one SaaS build breakdown every week

The stack, prompts, pricing, and mistakes to avoid — for solo founders building with AI.