Google Analytics 4 (GA4) was designed for corporate marketing departments with fifty employees, endless cross-domain tracking requirements, and specialized data engineers. For a solo founder running a lean content site, a digital product storefront, or a subscription micro-SaaS, logging into GA4 is an exercise in pure disorientation.
You wait twenty-four to forty-eight hours for visitor data to normalize. You must slap an ugly, conversion-killing cookie consent banner across your landing page to remain compliant with EU privacy laws. And your visitors’ browsers load a bloated 45KB+ JavaScript tag that slows down your Core Web Vitals.
This frustration is why Umami (github.com/umami-software/umami) has become the default analytics weapon for indie hackers.
It is an open-source, cookie-free, privacy-focused analytics engine built with Next.js, TypeScript, Prisma, and PostgreSQL. It delivers a fast, beautiful dashboard inspired by paid services like Fathom and Plausible—without charging you ten to thirty dollars every single month.
I have run self-hosted Umami instances tracking millions of monthly pageviews. Here is the operational reality of deploying and scaling it for your own business portfolio.
Repository Snapshot (Data verified September 8, 2026)
- GitHub Repository: https://github.com/umami-software/umami
- Stars: 38,673 | Forks: 7,984 | Open Issues: 126
- License: MIT (Permissive commercial use)
- Primary Language: TypeScript (Next.js + Prisma)
- Contributors: ~428 | Last Commit: September 8, 2026
What Umami Does: Real-Time Traffic Without the Surveillance
At its core, Umami does one job exceptionally well: it collects clean, aggregated website traffic data without invading your visitors’ personal privacy.
Here is how it operates:
- No Cookies or Persistent Trackers: Umami generates a daily rotating hash based on the visitor’s IP address and User-Agent string. It does not store personal data, track users across different domains, or set third-party tracking cookies.
- GDPR and PECR Compliant by Design: Because no personal data is retained, you can legally remove cookie consent banners in most European jurisdictions, instantly boosting your mobile conversion rate.
- Lightweight Script: The tracker script (
script.js) is roughly 2 kilobytes gzipped—compared to Google Tag Manager’s 45KB payload. - Multi-Website Management: A single Umami dashboard allows you to generate tracking scripts for unlimited personal domains, client sites, and staging apps with individual access tokens.
The Economics: Self-Hosted Umami vs. Fathom vs. Plausible
If you manage five small bootstrapped websites that collectively generate 100,000 pageviews per month:
| Solution | Monthly Cost | Hosting Setup | Cookie Banner? | Data Ownership |
|---|---|---|---|---|
| Umami (Self-Hosted) | $0 – $6 / month | Free Vercel + Supabase or VPS | No | 100% Yours |
| Fathom Analytics | $24 / month | Fully Managed SaaS | No | Hosted with Fathom |
| Plausible (Cloud) | $19 / month | Fully Managed SaaS | No | Hosted with Plausible |
| Google Analytics 4 | $0 (Ad-supported) | Google Cloud | Mandatory (GDPR) | Stored with Google |
Over three years of running your small software company, paying $20/month for simple privacy analytics burns $720 in cash. That is money that could have covered your domain renewals, email marketing platform, or paid customer acquisition experiments.
With Umami, you can host the Next.js web application on Vercel’s free hobby tier or Coolify, connect it to a free Supabase or Railway PostgreSQL database, and track your first 200,000 pageviews per month for exactly zero dollars.
Commercial License: Fully Permissive MIT
Unlike Plausible Analytics—which is licensed under the more restrictive AGPL-3.0—Umami is licensed under the standard MIT License.
This distinction matters if you plan to integrate analytics directly into customer-facing software:
- Client Portals: If you build a SaaS tool that provides your users with custom subdomains or private storefronts, you can bundle Umami’s tracking and display embedded client metrics dashboards without having to open-source your proprietary application codebase.
- No Legal Ambiguity: You can run it on your own servers, white-label the dashboard with your company logo, and share public or password-protected dashboards with investors, advertisers, or clients.
Where Umami Breaks Down: The Database Bottleneck
While Umami is practically flawless for early-stage sites, high-volume traffic reveals its architectural limitation: PostgreSQL row growth.
Each pageview, custom event (like a button click or checkout trigger), and outbound link click is written as an individual row into your PostgreSQL database. If your site gets hit by a viral post on Hacker News or Reddit and registers 250,000 pageviews in forty-eight hours, your database will record 250,000 new rows with associated indexes.
Here is what happens as your instance scales:
- Database Disk Bloat: An unpruned Umami database tracking moderate traffic easily consumes 5GB to 15GB of disk space within a year.
- Dashboard Query Latency: Generating the “Last 90 Days” analytics chart requires PostgreSQL to aggregate millions of individual event records. On a cheap 1GB RAM database server, opening your dashboard can take six to ten seconds.
How to solve this as a solo operator:
- Enable Umami’s built-in data retention policy in the settings to automatically purge raw session logs older than 90 or 180 days while keeping aggregate statistics.
- If you exceed 500,000 pageviews per month, add a connection pooler (like PgBouncer) or consider migrating to ClickHouse, a columnar database specifically designed for massive analytical queries.
Umami vs. Plausible Analytics: The Bootstrapper’s Choice
Both Umami and Plausible are incredible European-born open-source projects, but they target different technical philosophies:
- Tech Stack: Umami is written in TypeScript and Next.js, running on standard Node.js and PostgreSQL. Plausible is written in Elixir and uses ClickHouse. If you are a JavaScript/TypeScript developer, modifying or debugging Umami takes minutes. Compiling and debugging Elixir/ClickHouse containers on a budget VPS is considerably more demanding.
- License: Umami’s MIT license is dramatically more flexible for commercial builders than Plausible’s AGPL-3.0.
- UI Polish: Both interfaces are remarkably fast and clean. Umami offers slightly more granular filtering for custom event properties and UTM campaign performance.
How I Would Set Up Umami Today
If you want a reliable analytics setup that takes twenty minutes to configure and requires virtually zero maintenance:
- Database: Spin up a free PostgreSQL database on Supabase or Neon.
- Compute: Deploy the official Umami GitHub repository to Vercel with one click (Umami provides an official “Deploy to Vercel” button in their README).
- Domain: Point a custom subdomain like
stats.yourdomain.comortelemetry.yourdomain.comto the Vercel deployment. - Proxy Tracker: Avoid ad-blockers by renaming the script or configuring a reverse proxy in your main domain’s Next.js/Astro middleware so requests route through
/api/sendinstead of directly to a recognizable analytics domain.
For 95% of bootstrappers and solopreneurs, self-hosted Umami is the single most cost-effective way to get high-fidelity, real-time traffic data without compromising visitor privacy or paying a permanent monthly subscription.