Stack
PostHog
A usage-based product analytics platform that combines event tracking, session replay, feature flags, experiments, surveys, and error tracking, available as cloud or self-hosted software.
What it actually does
PostHog records product events and turns them into funnels, retention curves, trends, and user paths, so you can see what people do inside an app rather than just how many arrive. It layers session replay, feature flags, A/B tests, surveys, and error tracking on top of the same event data.
Billing is usage-based with a monthly free allowance across each product (events, replays, flag requests, surveys, exceptions). The free tier resets every month and needs no credit card, so many early-stage products pay nothing until they gain real traction.
The platform is open source and can be self-hosted, and billing limits let you hard-cap spend so a traffic spike cannot produce a surprise invoice. The main cost risk is autocapture and replays quietly stacking multiple meters.
PostHog vs its main rivals
The tools people actually weigh against PostHog: Mixpanel, Amplitude, Heap. Same criteria for every column, including where PostHog loses.
| PostHog |  Mixpanel | AAmplitude | HHeap | |
|---|---|---|---|---|
Pricing & access | ||||
| Free tier | Yes1M events + 5k replays | Yes1M events | Yes50k MTUs | YesLimited |
| Starting price | $0 usage-based | $0, then $0.28/1k | $49/mo (Plus) | Custom quote |
| Open source / self-host | YesYes | NoNo | NoNo | NoNo |
Product suite | ||||
| Session replay | YesYes | YesYes | YesYes | YesYes |
| Feature flags / experiments | YesYes | NoNo | YesYes | NoNo |
| Error tracking / surveys | YesYes | NoNo | YesPartial | NoNo |
Analytics depth | ||||
| Point-and-click funnel polish | YesGood | YesBest-in-class | YesBest-in-class | YesYes |
| Funnels / retention | YesYes | YesYes | YesYes | YesYes |
Developer & API | ||||
| REST / query API | Free | Free | Free | Paid |
Scaling & limits | ||||
| Cost at ~5M events/mo | ~$200 | ~$1,120 | MTU-based | Quote |
| Spend / billing caps | YesYes | YesPartial | NoNo | NoNo |
| Data retention (paid) | 7 yr | 5 yr | Varies | Varies |
- The monthly free tier (1M events, 5k replays) covers most pre-launch and early products entirely
- One tool replaces separate analytics, replay, feature-flag, A/B, error-tracking, and survey subscriptions
- Billing limits let you hard-cap spend, and the platform can be self-hosted
- Costs stack across products, so multiple active meters can reach a few hundred dollars a month at moderate scale
- Autocapture inflates event counts quickly and each replay past the free 5k costs extra, burning the free tier if left unconfigured
- Point-and-click funnel and cohort ergonomics trail dedicated tools like Mixpanel and Amplitude
Pay less for it
5 ways found1M events and 5k replays every month, no credit card, resets monthly
Set per-product spend caps on day one so a spike cannot surprise-bill you
The open-source edition removes per-event fees if you run your own infrastructure
PostHog for Startups grants sizable platform credits to eligible early companies
Umami — If you only need traffic counts and not replays, flags, or product analytics, a free privacy dashboard is far cheaper
StackTracker tracks what you actually pay for PostHog and every other tool, flags overpayment, and shows the dollars you would save by switching.
Is it the right tool for you
- You need analytics, session replay, and feature flags in one tool
- You are pre-launch or early and want to pay $0 until real traction
- You want the option to self-host for data control
- You only need a simple traffic dashboard → use Plausible
- You need the deepest point-and-click funnel and cohort analysis → use Mixpanel
- You want revenue-per-channel tied to Stripe → use DataFast
Track what PostHog and the rest of your stack cost
StackTracker adds up every subscription, plus your hours, so you see the real number.
Prices and limits last verified 2026-07-20.