Litemetrics vs PostHog
PostHog is a full product analytics platform: events, funnels, cohorts, session replay, feature flags, experiments, surveys, and more. Litemetrics is the opposite philosophy: a focused embedded analytics SDK with one job (give your SaaS customers a dashboard inside your product). If you only need the dashboard, PostHog is often more than you want to operate.
Two different products
PostHog is a platform. It is hugely capable, well-funded, and the right pick when you want one tool that does product analytics, replays, flags, and experiments together. The price is operational weight: PostHog needs ClickHouse plus Kafka plus Zookeeper plus Postgres plus Redis plus a few services. The hosted plan starts free; serious usage runs into hundreds or thousands per month.
Litemetrics is a single binary plus one database. There are no feature flags, no replays, no experiments. There is a tracker, a collector, a query API, and a React dashboard. That is the entire surface area.
Side by side
| Capability | PostHog | Litemetrics |
|---|---|---|
| License | MIT (some features paid) | MIT (everything) |
| Self-host complexity | ClickHouse + Kafka + Zookeeper + Postgres + Redis + multiple services | One container + one DB (ClickHouse, Postgres, or MongoDB) |
| Tracker bundle size | ~50 KB | ~3.5 KB |
| Product features beyond analytics | Replays, flags, experiments, surveys | None (intentionally focused) |
| Embeddable React dashboard | iframe (no native React) | Native React components, themeable |
| White-label / multi-tenant for SaaS customers | Possible with engineering effort | Built in (site_id isolation, secret keys) |
| Hosted pricing | Free up to 1M events, then per-event tiered | Self-host only (free) |
When PostHog is the right pick
- You want product analytics, replays, and feature flags in one platform.
- You are tracking your own product internally, not exposing analytics to customers.
- You have the operational appetite to run (or pay for) a multi-service stack.
When Litemetrics is the right pick
- The customer-facing analytics dashboard is the use case, not internal product analytics.
- You want one container plus one database, not five services.
- Bundle size matters because you ship the tracker to your users (3.5 KB vs 50 KB on a slow connection makes a real LCP difference).
- You want native React components for the dashboard, not an iframe.
Can I run both?
Yes. A common pattern is PostHog for internal product analytics (your team) and Litemetrics for external customer-facing analytics (your users seeing their own data). The two have non-overlapping jobs.
Try the live demo or jump to Quickstart.