Umami

Open-source web analytics, honestly reviewed. No marketing fluff, just what you get when you self-host it.

TL;DR

• What it is: Open-source (MIT) web analytics platform — the data stays on your server, no cookies are set, and Google never sees your traffic [1][2].
• Who it’s for: Developers, indie hackers, and privacy-conscious founders who want clean traffic insights without feeding their visitors’ behavior to a surveillance ad network. Also non-technical founders who have someone to deploy Docker for them [1][4].
• Cost savings: Google Analytics is free in the same sense that Facebook is free. You pay with your visitors’ data. Self-hosted Umami runs on a $5–10/mo VPS with unlimited data, no pageview caps, and no data-sharing with third parties. Umami’s cloud pricing was not available in the sources reviewed.
• Key strength: Genuinely lightweight, genuinely private, and fast to set up. No cookies means no consent banner — a real operational win for EU-facing sites [1][4].
• Key weakness: Event tracking is limited and has frustrated users who need funnel analysis or fine-grained filtering. It’s a pageview-and-referrer tool first. If you want product analytics depth, you’re looking at the wrong tool [4].

---

What is Umami

Umami is a web analytics platform built specifically to be a cookieless, privacy-first alternative to Google Analytics. It tracks page views, referrers, device types, browser agents, and live visitor counts — and it does nothing else it doesn’t need to. All data goes to your own database. No third-party servers, no cookies, no ad network fingerprinting [1][2].

The project was created by three brothers — Mike, Brian, and Francis Cao — and has grown to 35,719 GitHub stars under an MIT license [merged profile][4]. That star count puts it in a different tier from most self-hosted analytics tools: this isn’t a hobby project with 200 contributors. It has real adoption.

The pitch in its own words: “A simple, fast, privacy-focused alternative to Google Analytics.” That’s it. No AI dashboards, no behavioral scoring, no data monetization plays. The simplicity is the feature.

What you get when you install it: a Next.js application backed by PostgreSQL (or MySQL/MariaDB), a lightweight JavaScript tracking snippet you drop into your site, and a clean dashboard showing the metrics that actually matter for most website owners — traffic volume, where visitors came from, what pages they hit, what devices they’re on, and who’s on the site right now [1].

---

Why people choose it

The case for Umami falls into two categories: the privacy argument and the cost argument. They’re related but distinct.

The privacy argument. Google Analytics is technically free, but you’re giving Google behavioral data about every visitor who hits your site. For EU-based sites, this has become legally risky — GDPR enforcement has resulted in fines for sites that pass data to US-based analytics providers without proper consent mechanisms. Umami sidesteps this entirely: no cookies, no personal data collection, no data leaving your infrastructure. The FreeCodeCamp article [1] is explicit about this: Umami is “compliant with privacy regulations like GDPR and CCPA, making it ideal for privacy-focused organizations seeking transparent, user-controlled analytics.”

The practical win: no cookie consent banner needed. For sites that get meaningful EU traffic, this alone reduces friction on every page load.

The cost argument. Google Analytics is free, so the cost comparison is unusual. The real savings aren’t against GA — they’re against the privacy-compliant SaaS alternatives you’d have to pay for if you want similar data without the Google relationship. Plausible charges $9/mo for 10,000 pageviews; Fathom charges $14/mo for 100,000 pageviews; Simple Analytics starts at $9/mo. Self-hosted Umami on a $6 Hetzner VPS handles unlimited pageviews with no caps [3][4]. If you’re already running a VPS for other services, Umami adds essentially no marginal cost.

What G2 reviewers say. The real-user reviews captured in the Simple Analytics article [4] are consistent: people value the simplicity and the setup speed. One verified G2 user: “Very easy to use and provides superb value for free… it respects the German GDPR (DSGVO). It will not be blocked by uBlock or other adblockers.” Another: “Very simple to set up via Docker. Possible to rename the script for your website, so that it won’t be blocked by uBlock.”

That last point matters more than it sounds. The script rename feature means privacy-conscious visitors running ad blockers still send analytics data. With Google Analytics or similar tools, adblockers strip the tracking entirely. Umami’s self-hosted origin URL makes this a configuration decision rather than a platform limitation [4].

The honest limitations from the same reviews. The same G2 reviewers flag real gaps: “It’s a bit limited in regards to ‘ordering’ or ‘filtering’ options.” Another: “No option to send events. No option for funnels.” [4] These aren’t edge cases. If you want to track button clicks, form submissions, checkout flows, or conversion paths, Umami’s event tracking is present but immature enough that reviewers explicitly call it out as a weakness [4].

---

Features

Based on the README and article descriptions:

Core analytics:

• Page view tracking with historical data [1]
• Real-time live visitor count [1]
• Referrer tracking — where traffic comes from [1]
• Device, browser, and OS breakdown [1]
• UTM parameter tracking (utm_source, utm_medium, etc.) [4]
• Multiple websites per Umami instance with separate dashboards [1]

Privacy architecture:

• No cookies set by default [1][2]
• No personal data collected or stored [2][4]
• All data stays on your database [1]
• GDPR and CCPA compliant by design [1][4]
• Tracking script can be renamed to avoid adblocker detection [4]

Deployment and operation:

• Docker and Docker Compose deployment (single command) [README]
• Source build via Node.js 18.18+ and pnpm [README]
• PostgreSQL 12.14+ required (MySQL/MariaDB also supported per [1])
• Lightweight tracking script added to target websites [1]
• Multilingual interface [2]

What’s not there (or limited):

• Event tracking is present but users consistently report it as limited [4]
• No funnel analysis [4]
• Limited filtering and ordering on the dashboard [4]
• No A/B testing, no feature flags, no session replay — this is strictly pageview analytics

---

Pricing: SaaS vs self-hosted math

Self-hosted Umami:

• Software license: $0 (MIT) [merged profile]
• VPS to run it: $5–10/mo (Hetzner, Contabo, DigitalOcean)
• Database: bundled via Docker Compose, or connect to an existing Postgres instance
• Pageview caps: none — unlimited

Umami Cloud: Umami offers a managed cloud version, but cloud pricing was not captured in the sources reviewed for this article. Check https://umami.is/pricing for current tiers.

The actual SaaS comparison isn’t Google Analytics — it’s the privacy-compliant alternatives:

Tool	Monthly cost	Pageview cap	Open source
Plausible.io	$9/mo	10,000 pv	Yes (AGPL)
Fathom Analytics	$14/mo	100,000 pv	Closed source
Simple Analytics	$9/mo	10,000 pv	Partial
Umami self-hosted	~$6/mo (VPS)	Unlimited	Yes (MIT)

Source for competitor pricing: [3]

The math is favorable if your site gets significant traffic. A site doing 200,000 pageviews/month would be on Plausible’s higher tiers ($19–$29/mo range) or Fathom’s growth tier. Umami self-hosted handles that for the cost of the VPS you’re probably already paying for.

The honest caveat: if you’re coming from Google Analytics, your current cost is $0. The switch to Umami costs you $5–10/mo plus setup time. The value proposition is privacy and data ownership, not dollar savings over GA.

---

Deployment reality check

The README’s primary path is Docker Compose, which runs Umami plus a PostgreSQL database in two containers. The FreeCodeCamp article [1] documents a seven-step install process from source (clone, configure, build, start). The Docker path is considerably shorter.

What you actually need:

• A Linux VPS with Node.js 18.18+ (if building from source) or Docker installed
• PostgreSQL 12.14+ (bundled in Docker Compose or external)
• A domain name and reverse proxy (Nginx or Caddy) for HTTPS
• The tracking snippet pasted into your website’s HTML <head>

What the setup actually looks like for Docker:

docker compose up -d

Umami provides the compose file. Database is included. Default credentials are admin/umami — change these immediately [README].

What can go sideways:

The FreeCodeCamp article [1] uses Sevalla (a cloud platform) as an example deployment target, which suggests the raw VPS path isn’t entirely trivial for non-technical users. A non-technical founder who has never touched a Linux server will need help. The Docker path is straightforward if you know Docker; the source build path has more surface area for failure.

G2 reviewers don’t complain about installation — the consistent friction point is the dashboard’s filtering and ordering limitations once you’re running it, not the setup [4].

Realistic time estimate: 15–30 minutes for a developer on a fresh VPS with Docker installed. 2–4 hours for a non-technical user following a guide, including domain and HTTPS setup. If you’re not comfortable on a Linux server, one-time deployment help will pay back fast.

One notable convenience feature: the tracking script can be renamed to avoid adblockers. This is a configuration option, not a workaround that requires custom builds [4]. For a publisher trying to get accurate traffic numbers rather than “users who didn’t install uBlock,” this matters.

---

Pros and Cons

Pros

• Actually MIT-licensed. No commercial restrictions, no “fair-code” carve-outs. Self-host, fork, embed in your product — no legal friction [merged profile][1].
• No cookies, no consent banner needed. This is operationally valuable for EU sites. The absence of a cookie consent modal is a user experience improvement on every page load [1][2][4].
• 35,719 GitHub stars. Not a one-person hobby project. Strong signal of real adoption and community longevity [merged profile].
• Genuinely lightweight tracking script. Multiple sources confirm it doesn’t slow sites down [4][2].
• Adblocker bypass via script rename. Accurate data collection without requiring visitors to whitelist your domain [4].
• GDPR and CCPA compliant by design. Not compliant-with-settings; compliant by default architecture [1][4].
• Clean, simple dashboard. Repeatedly praised as easy to read and navigate [4][2].
• Multiple sites per instance. Run one Umami install for all your projects [1].
• Docker Compose deployment. One command to a working instance [README].

Cons

• Event tracking is limited. G2 reviewers explicitly call this out — no funnels, limited event options. If you need conversion tracking beyond pageviews, you’ll hit the ceiling fast [4].
• Limited filtering and ordering. Dashboard UX is praised for simplicity but criticized for not letting you slice data the way you’d want for analysis [4].
• Not for product analytics. No session replay, no A/B testing, no feature flags, no behavioral cohorts. If you’re trying to optimize a SaaS product rather than understand traffic, this isn’t your tool [4].
• Non-technical founders need help with setup. The Docker path is genuinely simple, but “simple for a developer” is not the same as “simple for a founder who’s never SSH’d into a server” [1].
• Cloud pricing opaque. The self-hosted value is clear; the managed cloud offer wasn’t documented in the sources available.
• PostgreSQL or MySQL dependency. Not a file-based database — you’re running a full database server, which adds operational overhead for teams that don’t already manage databases [README][1].
• Limited support options. SaaSHub reviewers note limited support compared to commercial alternatives [2]. Community support via Discord and GitHub issues is what you get on the free tier.

---

Who should use this / who shouldn’t

Use Umami if:

• You’re a developer or founder who wants honest traffic data without Google owning it.
• Your site has EU visitors and you’re tired of cookie consent banners slowing down every page load.
• You’re paying $9–$29/mo for Plausible, Fathom, or Simple Analytics and you’d rather own the infrastructure.
• You have multiple sites and want one dashboard for all of them without paying per-property fees.
• You care about accurate data and know that adblockers hollow out Google Analytics numbers for a privacy-conscious audience.

Skip it (stay on Google Analytics) if:

• You need deep conversion funnels, custom segments, and attribution modeling — GA4 is genuinely powerful for this use case despite its privacy costs.
• You have no technical help and can’t deploy Docker. The tool is fast once it’s running, but getting there requires basic server skills.
• Your compliance team needs enterprise support contracts and SLAs — Umami self-hosted has none.

Skip it (try PostHog instead) if:

• You’re building a SaaS product and need session replay, feature flags, A/B testing, and event funnels alongside traffic analytics. PostHog is also MIT-licensed and self-hostable [3].

Skip it (try Matomo instead) if:

• You need enterprise-grade analytics with complete GA4 feature parity, including ecommerce tracking, goal conversions, and detailed segmentation. Matomo’s self-hosted version is more complex to run but significantly more capable [3].

Skip it (try Plausible instead) if:

• You want the privacy-first approach but don’t want to manage infrastructure, and you’re OK paying $9/mo for a European-hosted SaaS with a strong privacy track record [3].

---

Alternatives worth considering

From the SaaSHub alternatives page [3] and the articles:

• Plausible.io — the strongest direct competitor. Also privacy-focused, no cookies, GDPR-compliant by default. Self-hostable under AGPL (more restrictive than Umami’s MIT). Cloud starts at $9/mo for 10,000 pageviews. The EU hosting angle is a real differentiator if you want managed cloud without US data transfers [3].
• Matomo — the heavyweight open-source analytics option. More capable than Umami, significantly more complex to self-host and operate. Has both self-hosted (free) and cloud (paid) tiers. If you need full GA4 parity, Matomo is the answer [3].
• Fathom Analytics — closed-source SaaS, privacy-focused, $14/mo for 100,000 pageviews. No self-hosting option, but simpler than Plausible for non-technical users [3].
• Simple Analytics — European-owned SaaS, $9/mo, partial open source. Published an honest comparison of Umami vs. their product [4] — worth reading, with the obvious caveat that they’re selling their alternative.
• PostHog — MIT-licensed, self-hostable, and covers product analytics (events, funnels, session replay, feature flags, A/B tests) in addition to pageview tracking. Much heavier operationally, but if you need the full suite, PostHog is the only self-hosted option that competes with Mixpanel or Amplitude [3].
• GoAccess — terminal-based, no web UI, parses your server logs directly. Zero install overhead, zero database, zero JavaScript. If you want traffic numbers without any client-side code, GoAccess is the purist choice [3].

For a non-technical founder who just wants traffic numbers and no cookie banners: Umami vs. Plausible is the realistic shortlist. Pick Umami if you’re self-hosting. Pick Plausible if you want managed cloud in the EU.

---

Bottom line

Umami does exactly what it claims: clean, private, fast web analytics with no cookies and no Google. It’s not trying to replace Mixpanel or compete with PostHog’s product analytics suite — it’s trying to be the honest alternative to Google Analytics for developers and founders who understand why “free with data harvesting” isn’t actually free. At 35,719 GitHub stars and MIT-licensed, this is not an experimental tool. It’s a mature, well-maintained project with a clear scope.

The real-world limitations are genuine: event tracking and filtering have frustrated users enough to mention it unprompted in reviews [4]. If your analytics workflow needs conversion funnels or behavioral segmentation, you’ll hit the ceiling. But for the 80% use case — knowing how many people visited, where they came from, and what pages they read — Umami gets out of the way and gives you the answer without surveillance infrastructure attached.

If deployment is the blocker, that’s exactly what upready.dev deploys for clients. One-time fee, done, you own the data.

---

Sources

1. Manish Shivanandhan, freeCodeCamp — “How to Set Up Your Own Google Analytics Alternative Using Umami” (November 11, 2025). https://www.freecodecamp.org/news/how-to-set-up-your-own-google-analytics-alternative-using-umami/

2. SaaSHub — “umami reviews. Is umami good?”. https://www.saashub.com/umami

3. SaaSHub — “umami Alternatives & Competitors” (Updated March 17, 2026). https://www.saashub.com/umami-alternatives

4. Ankit Ghosh, Simple Analytics — “Umami Analytics review [2025]” (Published August 17, 2023; edited April 1, 2025). https://www.simpleanalytics.com/resources/analytics-review/umami-analytics-review-and-a-better-alternative

Primary sources:

• GitHub repository: https://github.com/umami-software/umami (35,719 stars, MIT license)
• Official website: https://umami.is
• Documentation: https://umami.is/docs