Umami vs Google Analytics
Self-host swap-in for Google Analytics. · Self-host Google Analytics · Umami on os-alt
Umami is one of the open-source self-host replacements for Google Analytics — license MIT, 10min docker-compose to stand up, and $5 vps — next. Compare against Google Analytics's Free (GA4); $150k+/yr for GA360 / Analytics 360 below.
| Umamiopen-source | Google Analyticspaid SaaS | |
|---|---|---|
| Category | Web analytics | Web analytics |
| License / pricing | MIT | Free (GA4); $150k+/yr for GA360 / Analytics 360 |
| Starting price | $0 self-host | free tier |
| GitHub | umami-software/umami | closed source |
| Setup time | 10min docker-compose | SaaS — sign up + bill |
| Monthly cost | $5 VPS — Next.js + Postgres or MySQL. | Free (GA4); $150k+/yr for GA360 / Analytics 360 |
Switching from Google Analytics to Umami
Same constraint as Plausible — GA's free tier doesn't release raw event data. Run Umami in parallel with GA for an overlap, then cut over. Umami v2 exposes a `/api/send` endpoint compatible with simple custom-event reporting if you migrate event scripts.
- Good fit for
- Devs who want a single-binary-feel analytics stack and a clean dashboard.
- Weak at
- No funnel builder in OSS edition; cohort analysis is shallow.
Other open-source self-host alternatives to Google Analytics
Umami also replaces
- Fathom Analytics— Privacy-first web analytics
In a terminal? npx os-alt google-analytics prints Google Analytics's self-host options —
how the CLI works →
FAQ
Is Umami a free alternative to Google Analytics?
Yes — Umami is open source under MIT. Self-host cost: $5 VPS — Next.js + Postgres or MySQL.. Google Analytics pricing anchor: Free (GA4); $150k+/yr for GA360 / Analytics 360.
How long does Umami take to set up vs Google Analytics?
Self-hosting Umami: 10min docker-compose. Google Analytics is a hosted SaaS — sign up and you're in.
What is Umami good at, and what is it weak at?
Good fit for: Devs who want a single-binary-feel analytics stack and a clean dashboard.. Weak at: No funnel builder in OSS edition; cohort analysis is shallow..