Umami is a simple, fast, privacy-focused alternative to Google Analytics

Original Article ↗ Hacker News Discussion ↗

Real‑world use and reliability

  • Several self‑hosters report that Umami is good for hobby/personal sites but unreliable for “set and forget” use: frequent breaking changes (API, DB, frontend) have left some instances silently broken, pushing users to simpler tools.
  • Others say it “just works” for low‑volume sites and praise that self‑hosting isn’t crippled vs. the paid SaaS.
  • One heavy‑traffic user migrated from Matomo to Umami and found it far more performant at scale (tens of millions of visits/day), though team/user management is described as confusing or “weird.”

Features, UI, and usability

  • UI is widely liked for simplicity, but some find it “too basic” or immature for serious business use: missing global time ranges, easy localhost exclusion, and richer admin options.
  • Session duration and bounce handling (for long single‑page reads) is a general pain point across lightweight tools; workarounds require custom events.
  • Multiple commenters note that many “privacy analytics” tools share nearly identical dashboard layouts, viewed as either sensible standardization or just cloning.

Self‑hosting, deployment, and stack

  • Docker deployment is commonly used and considered straightforward; some integrate via language SDKs.
  • Complaints appear about needing Yarn/Docker/Node to build/run from source; some users drop the tool over that friction.
  • Umami supports bot filtering with an env flag to disable it when bot data is desired.

Privacy model, cookies, and legal debates

  • Big debate over “no cookies” claims: several argue the real issue is data collection and identifiers, not cookie mechanics.
  • Techniques like hashing IP+UA with a daily salt (Plausible‑style) are scrutinized: some say this respects privacy and avoids PII storage; others say any stable fingerprint or online identifier falls under GDPR/ePrivacy and may still require consent.
  • There’s disagreement on whether “privacy‑friendly analytics without banners” is legally sound or marketing spin; even lawyers and regulators are portrayed as inconsistent.
  • Some insist the only truly privacy‑respecting approach is to track as little as possible or not at all.

Ad‑blocking and tracking ethics

  • Umami’s documentation on bypassing ad/tracking blockers is strongly criticized: opponents argue that intentionally evading explicit user opt‑out is incompatible with a “privacy‑focused” label and amounts to spyware behavior.
  • Others see value in first‑party, self‑hosted analytics vs. shipping data to Google, even if it still tracks individuals to some degree.

Comparisons and ecosystem

  • Alternatives repeatedly mentioned: Plausible, Matomo, Pirsch, Goatcounter, UXWizz, onedollarstats, and others, each with different trade‑offs (features, legal posture, pricing, performance, self‑hosting).
  • Vendor representatives stress legal reviews and anonymization strategies, but some commenters question whether inexpensive legal opinions are sufficient risk coverage.

Naming and branding

  • Several comments dive into the name “Umami” and the broader trend of picking short, foreign words; opinions are split between finding it practical/memorable and seeing it as uncreative or confusing.