Betterlytics

Self-hosted Google Analytics alternative, honestly reviewed. New player, real features, small community — here’s what you actually get.

TL;DR

• What it is: Open-source (AGPL-3.0) cookieless web analytics platform built on Rust and ClickHouse, positioned as a privacy-first Google Analytics replacement with session replay and uptime monitoring bundled in [README][website].
• Who it’s for: Founders and developers who want more than just pageviews — session replay, user journeys, and Core Web Vitals — without stitching together three separate tools, and without cookie consent banners [5].
• Cost savings: Google Analytics 4 is free but ships your data to Google’s ad infrastructure. Plausible Cloud starts at $9/mo. Betterlytics Cloud starts at $6/mo with a free tier. Self-hosted: your VPS cost only [5].
• Key strength: Feature density at low price. Session replay, user journeys, funnels, uptime monitoring, SSL monitoring, and Core Web Vitals — all in one dashboard — for less than Plausible charges for pageview-only analytics [5][README].
• Key weakness: 147 GitHub stars. That’s not a typo. This is a very early-stage project with no substantial third-party reviews, an unproven production track record, and a tiny community. You’re betting on a small team and a young codebase [GitHub profile].

---

What is Betterlytics

Betterlytics is a self-hostable analytics platform that tracks website traffic without cookies, without personal data, and without requiring a consent banner. The pitch on its homepage is blunt: “A Better Alternative to Google Analytics” — cookieless, GDPR-ready, no third-party data sharing [website].

Under the hood, it’s a Rust backend writing events to ClickHouse (a columnar database built for analytics throughput), with a Next.js dashboard on top and PostgreSQL for metadata. The tracking script weighs under 2KB [README]. That stack is legitimately well-chosen for the problem — ClickHouse handles billions of rows comfortably, and Rust minimizes the memory overhead of the ingestion service. Whether a 147-star project has stress-tested it at serious scale is a different question.

What separates Betterlytics from the minimalist privacy-analytics crowd (Plausible, Umami) is scope. It doesn’t just count pageviews and call it a day. It includes session replay, user journey visualization, conversion funnels, Core Web Vitals monitoring, uptime monitoring with SSL certificate checks, and notification integrations for Slack and Discord [README][5]. For teams that currently pay separately for analytics, a session replay tool, and an uptime monitor, that bundling is the real value proposition.

The project is hosted at https://github.com/betterlytics/betterlytics under AGPL-3.0 license. The license matters: AGPL requires that any modifications you deploy on a server must be released as open source. That’s more restrictive than MIT (what Umami uses) and roughly equivalent to Plausible’s license posture.

---

Why people choose it over Google Analytics, Plausible, and Umami

There are essentially no independent third-party reviews of Betterlytics as of this writing. The project is too new. What we can do is lay out the positioning argument and let you weigh it.

Versus Google Analytics 4. The core privacy argument is the same one the whole privacy-analytics space makes: GA4 processes your visitors’ data on Google’s infrastructure, uses it to train advertising models, and requires consent banners in the EU [2]. Betterlytics collects no personal data and sets no cookies, so no consent banner is legally required. The self-hosted option means the data never leaves your server at all [README][website]. If you’re on GA4 because it’s free — Betterlytics has a free cloud tier and a self-hosted option that runs on any $5–10 VPS. The real question is whether you need Google’s SEO tooling and attribution integrations, which have no equivalent here.

Versus Plausible. This is the most useful comparison. Plausible is the established market leader in privacy-first analytics — self-funded, profitable, 20K+ GitHub stars, and a famously simple UI. Both tools are cookieless, GDPR-ready, lightweight, and open-source under similar AGPL licenses [2][5]. Betterlytics’ own comparison page makes the case honestly: Plausible intentionally keeps its scope narrow. No session replay, no uptime monitoring, no individual user journey visualization — by design [5]. Plausible starts at $9/mo for 10,000 pageviews with no free cloud tier. Betterlytics starts at $6/mo and includes a free tier [5]. If you’re evaluating Plausible and you also need session replay, Betterlytics is the more feature-complete option at lower starting price. The honest caveat: Plausible is a proven, stable, widely-used product with years of production hardening. Betterlytics is not, yet.

Versus Umami. Umami is MIT-licensed (freer than AGPL), has ~22K GitHub stars, a large community, and a dead-simple setup. It doesn’t do session replay, user journeys, or uptime monitoring [1]. If you want minimal viable privacy analytics and nothing more, Umami’s MIT license and community size make it the safer self-hosted bet. Betterlytics wins on features; Umami wins on maturity and license permissiveness.

Versus PostHog. PostHog is the full product analytics suite — session replay, feature flags, A/B testing, event pipelines, data warehouse sync. It’s a different category. If you’re a funded startup tracking product metrics, PostHog is the tool. If you’re a founder who just wants to know where traffic comes from and why users drop off, PostHog is overkill and complex to self-host. Betterlytics sits in the middle ground: more than Plausible, far less than PostHog.

---

Features

From the README and official feature pages:

Core analytics:

• Page views, unique visitors, bounce rate, session duration [README]
• Traffic sources, referrer tracking, UTM campaign parameters [README][website]
• Geographic analytics with world map visualization [website]
• Device, browser, and OS breakdown [README]
• Real-time event stream [website]

Advanced analytics:

• Session replay — anonymized, pixel-perfect recordings [README][website]
• User journey visualization — multi-step path flows [README][5]
• Conversion funnels with drop-off rates [README][5]
• Outbound link tracking [README]
• Custom events [README][5]
• Annotations [README]
• Time period comparisons [README]

Performance and monitoring:

• Core Web Vitals: LCP, INP, CLS, FCP, TTFB [README][website]
• Uptime monitoring with configurable check intervals [website]
• SSL certificate monitoring and expiry alerts [website]
• Slack and Discord notification integrations [5]

Privacy and compliance:

• Cookieless tracking — no cookies set, no consent banner required [README][website]
• GDPR, CCPA, PECR compliant by design [README][5]
• EU-based cloud infrastructure [website]
• No cross-site tracking, no third-party data sharing [README]
• Data anonymization [README]

Access and security:

• Role-based access control [README]
• Two-factor authentication [README]
• OAuth via Google and GitHub [README]

Developer experience:

• Under 2KB tracking script [README][5]
• Single <script> tag installation [website]
• React and Next.js SDK components with TypeScript [README][5]
• Self-hosting via Docker [README][website]
• Data retention up to 5 years (cloud) [5]

The feature table is genuinely impressive for a 147-star project. Whether session replay at scale, uptime monitoring reliability under load, and Core Web Vitals accuracy have been battle-tested in production by real teams is not something the available sources answer.

---

Pricing: SaaS vs self-hosted math

Betterlytics Cloud:

• Free tier: available for smaller sites [5]
• Paid plans: start at $6/month [5]
• Full pricing breakdown not published in scraped pages — check https://betterlytics.io/pricing directly for current tiers

Betterlytics self-hosted:

• Software: $0 (AGPL-3.0) [README]
• Infrastructure: $5–15/mo for a VPS
• Setup: Docker-based, documentation at https://betterlytics.io/docs/installation/self-hosting

Plausible for comparison [5][2]:

• No free cloud tier
• Starts at $9/month for 10,000 pageviews
• Self-hosting is free (AGPL), but requires more setup than Betterlytics claims to

Google Analytics 4:

• Free, but the product is the data — your visitors’ behavior feeds Google’s advertising infrastructure
• No self-hosting option

Concrete math for a content site with 50,000 pageviews/month:

Plausible Cloud at that scale runs roughly $19/mo. Betterlytics Cloud would be in the $6–20 range depending on the tier (exact pageview tiers not confirmed in available data). Self-hosted on a $6 Hetzner VPS: $6/mo for unlimited traffic, plus the hour or two to set it up. You’d also be getting session replay and uptime monitoring that would otherwise cost you $20–50/mo from a dedicated tool like LogRocket or UptimeRobot.

Over a year: Plausible + a separate uptime monitor ≈ $400+. Betterlytics self-hosted ≈ $72. That math works — if the self-hosted setup holds up under your traffic.

---

Deployment reality check

The README’s install path is Docker Compose. The documentation promises a self-hosting guide at https://betterlytics.io/docs/installation/self-hosting. The stack requires ClickHouse and PostgreSQL — both are bundled in the default Docker Compose setup.

What you need:

• A Linux VPS with at least 2–4GB RAM (ClickHouse is memory-hungry; plan for at least 4GB if you have real traffic)
• Docker and docker-compose
• A domain and reverse proxy (Caddy or nginx) for HTTPS
• Roughly 30–90 minutes for initial setup

Where things can get complicated:

• ClickHouse is not a trivial dependency. It’s powerful and fast, but it consumes more RAM than SQLite-backed tools like Umami. On a $4 VPS it will struggle under any real load.
• Session replay data adds storage requirements that pageview-only analytics don’t — budget for this.
• The project is 147 stars old. The documentation and Docker Compose configuration may have rough edges that a more mature project (Plausible, Umami) has already ironed out through community bug reports. If you hit a wall during setup, the community is small enough that GitHub issues may be your only escalation path.
• No third-party deployment guides or community tutorials were available as of this review. With Plausible or Umami, a web search turns up dozens of step-by-step walkthroughs for every major hosting platform. Betterlytics doesn’t have that yet.

The website claims “ready in 30 seconds” for cloud and “manual installation via Docker” for self-hosted — which is accurate but undersells the ClickHouse memory requirements for a budget VPS [website].

---

Pros and cons

Pros

• Feature density at low price. Session replay, user journeys, funnels, uptime monitoring, SSL checks, Core Web Vitals — one dashboard, starting at $6/mo cloud or VPS cost only for self-hosted [README][5]. Competitors charge separately for each of these.
• Genuinely cookieless. Not “cookie-optional” or “with consent mode” — fully anonymized by default, so no consent banner required anywhere [README][website][5].
• Modern tech stack. Rust ingestion + ClickHouse is the right architecture for high-throughput analytics. The choice signals technical seriousness [README].
• Cheaper than Plausible on cloud, more features included, and a free tier Plausible doesn’t offer [5].
• EU-hosted cloud — meaningful for European founders with GDPR exposure [website].
• Self-hosting option with full data ownership — data never touches anyone else’s infrastructure [README].
• AGPL open source — you can audit the code, which matters for privacy tools [README].

Cons

• 147 GitHub stars. This is an early-stage project with minimal production validation. Plausible has ~20K stars and years of production hardening. Umami has ~22K stars and MIT license. Betting your analytics infrastructure on a 147-star project carries real risk [GitHub profile].
• No third-party reviews. As of this writing, no independent reviewer has published an assessment. The only comparison content is Betterlytics’ own vs. pages — which are marketing, not journalism [5].
• ClickHouse memory overhead. The performance stack is right for scale, but overkill and resource-intensive for a $4 VPS. Expect 2–4GB RAM minimum for comfortable operation. Umami and Plausible are leaner on small instances.
• AGPL license is more restrictive than MIT. If you want to embed this in a commercial product or resell it, you need to release your modifications. Check with a lawyer if that matters for your use case [README].
• Unproven session replay at scale. Session replay is storage-intensive and latency-sensitive. Whether Betterlytics’ implementation holds up under thousands of daily sessions on a self-hosted instance is not documented anywhere.
• Small community means slow bug resolution, few tutorials, and limited ecosystem integrations compared to established tools [GitHub profile].
• Data retention capped at 5 years on cloud — Plausible offers unlimited retention on their paid plans [5].

---

Who should use this / who shouldn’t

Use Betterlytics if:

• You want to replace three tools (analytics + session replay + uptime monitor) with one, and budget is tight.
• You’re a developer comfortable with Docker who can handle a slightly rough self-hosting experience on a young project.
• You’re on Plausible’s paid plan and you’re frustrated that you can’t see user sessions or journeys without adding another tool.
• Privacy compliance (GDPR, CCPA) is non-negotiable and you don’t want to rely on Google’s infrastructure.
• You’re willing to be an early adopter in exchange for a lower price and more features.

Skip it, use Plausible instead, if:

• You want a proven, stable, widely-deployed privacy analytics tool with years of production history and a large community.
• You don’t need session replay or monitoring — Plausible’s minimalism is a feature, not a flaw.
• You want unlimited data retention on cloud.

Skip it, use Umami instead, if:

• You want an MIT-licensed self-hosted option with a large community and no ClickHouse overhead.
• You’re on the smallest VPS tier and need something lean.
• You don’t need session replay, journeys, or monitoring — just clean traffic stats.

Skip it, use PostHog instead, if:

• You’re building a product that needs feature flags, A/B testing, and deep product event analytics.
• You have engineering resources to manage a more complex self-hosted deployment.
• You need a vendor with enterprise support and contractual SLAs.

Stay on GA4 if:

• You’re in a market where Google’s attribution data (Search Console integration, ad performance) is critical to your workflow.
• GDPR consent banners don’t bother you and cookie compliance is handled.

---

Alternatives worth considering

• Plausible — the established leader in privacy-first analytics. AGPL, ~20K stars, starts at $9/mo, no session replay. The safe choice if maturity matters more than feature count [2].
• Umami — MIT-licensed, ~22K stars, simpler stack than Betterlytics, large community. No session replay, no monitoring. The best self-hosted option for lean installs [1].
• PostHog — full product analytics suite with session replay, feature flags, A/B testing. Open-source, significantly more complex to self-host. For product teams, not just traffic analytics.
• Matomo — the longest-running Google Analytics alternative. Full-featured, mature, heavier. Good compliance story for enterprise.
• Fathom — closed-source cloud-only, simple, privacy-first. No self-hosting option.
• GoatCounter — extremely minimal, MIT-licensed, trivial to self-host. No session replay, no funnels, but zero operational overhead.

For a non-technical founder evaluating options: the realistic shortlist is Plausible vs Betterlytics vs Umami. Choose Plausible if you want proven and simple. Choose Betterlytics if you want session replay and monitoring bundled in. Choose Umami if you want MIT license and minimal VPS footprint.

---

Bottom line

Betterlytics makes a compelling paper case: more features than Plausible, lower price, same privacy standards, built on a technically sound stack. If it were three years older with 10K GitHub stars and a hundred independent reviews, this would be a straightforward recommendation. It isn’t. With 147 stars and no third-party validation, you’re betting on a team and a codebase that hasn’t been stress-tested in public. That’s not a dealbreaker for a developer who can debug Docker Compose and contribute a GitHub issue — it’s exactly the kind of early tool that rewards early adopters who can tolerate some rough edges. But for a non-technical founder who needs something that just works for the next three years, the safer bet is Plausible or Umami while keeping an eye on Betterlytics as it matures.

If you do want to self-host Betterlytics and don’t want to deal with the Docker setup yourself, that’s exactly what upready.dev deploys for clients — one-time fee, running on your infrastructure, you own the data.

---

Sources

1. AlternativeTo — Umami (community reviews and comparison data). https://alternativeto.net/software/umami/about/
2. AlternativeTo — Plausible Analytics (community reviews, feature tags, and pricing context). https://alternativeto.net/software/plausible-insights/about/
3. OpenAlternative — Open Web Analytics (blocked by Cloudflare, unavailable). https://openalternative.co/openwebanalytics
4. OpenAlternative — Humblytics alternatives (blocked by Cloudflare, unavailable). https://openalternative.co/alternatives/humblytics
5. Betterlytics — “Betterlytics vs Plausible: More Features, Same Privacy” (official comparison page). https://betterlytics.io/vs/plausible

Primary sources:

• GitHub repository and README: https://github.com/betterlytics/betterlytics (147 stars, AGPL-3.0)
• Official website: https://betterlytics.io
• Pricing page: https://betterlytics.io/pricing
• Features page: https://betterlytics.io/features
• Self-hosting docs: https://betterlytics.io/docs/installation/self-hosting