Why leave Google Analytics now
Google Analytics relies on a data collection mechanism that transfers browsing data to servers located in the United States. Since the invalidation of the Privacy Shield in 2020 by the Court of Justice of the European Union (Schrems II ruling), such transfers can only legally take place with robust contractual safeguards — Standard Contractual Clauses (SCCs) — supported by a specific impact assessment. In practice, the vast majority of websites using Google Analytics lack this documentation. The rulings by the Austrian DSB (January 2022) and the French CNIL (February 2022) gave concrete form to this risk: they established that Google Analytics, in its standard configuration, transfers personal data (IP addresses, browser identifiers) outside the European Union without a sufficient legal basis. Germany, the Netherlands, Norway and Sweden have adopted convergent positions. The GDPR's accountability principle (Article 5.2) requires data controllers to demonstrate compliance at all times. Migrating to a self-hosted tool hosted in the EU directly satisfies this requirement: data never leaves your infrastructure, and you remain its sole controller. That is precisely what Plausible Analytics and Matomo offer — two open-source tools, deployable on your own VPS, with no dependency on a US third party.
Comparison criteria
- GDPR legality: native compliance with European regulations, no data transfer outside the EU, documentation available for the processing register.
- Visitor privacy: whether a cookie consent banner is required, minimised data collection, configurable retention period.
- Tracking script performance: size of the JavaScript file injected on each page, impact on Core Web Vitals (LCP, CLS, FID) and Lighthouse score.
- Feature depth: breadth of available reports — page views, sources, conversion funnels, A/B testing, heatmaps, audience segmentation.
- Ease of installation: Docker deployment complexity, number of required services (database, cache, reverse proxy), estimated time to production.
- Total cost of ownership: software licence (open source vs commercial), VPS infrastructure cost, monthly maintenance time and updates.
- Third-party integrations: native connectors with CMS platforms (WordPress, Shopify), marketing tools (Zapier, Make), e-commerce APIs and advertising platforms.
- Learning curve: complexity of onboarding for non-technical users, official documentation quality, community and available support.
Plausible Analytics: lightness first
Plausible Analytics is a web traffic analysis tool built around a single principle: measure the essentials without relying on any cookies or individually identifying visitors. The tracking script weighs less than 1 KB — twenty times lighter than Google Analytics 4 and twenty-two times lighter than Matomo's default script. This directly translates into reduced page load latency and measurable improvement in Core Web Vitals. On the legal front, Plausible sets no tracking cookies and collects no IP addresses. The tool aggregates browsing data via ephemeral cryptographic sessions, never associating them with a persistent profile. This architecture allows Plausible to be used without a consent banner in most European jurisdictions — a direct competitive advantage for conversion-focused websites. The project is available in two modes: Plausible Cloud, hosted in Europe (Germany, via Hetzner) from €9 per month for 10,000 monthly visits, and Plausible CE (Community Edition), the self-hosted version distributed under the AGPL-3.0 licence. The latter is free, actively maintained and deployable via Docker Compose in under fifteen minutes on a VPS with 256 MB of available RAM. The real-time dashboard is clean: popular pages, traffic sources, countries, devices and simple goal conversions.
Plausible Analytics vs Matomo: comparison table across 8 criteria
| Criterion | Plausible Analytics | Matomo |
|---|---|---|
| JS script / performance impact | < 1 KB, cookie-free | 22 KB + opt-in cookie |
| GDPR compliance out-of-the-box | Cookie-free, no consent required | Requires consent banner + specific configuration |
| Feature depth | Essential (pages, sources, countries, goals) | Advanced (funnels, A/B, heatmaps, segmentation) |
| Docker installation | Simple (official image + PostgreSQL) | Complex (MariaDB, Redis, nginx, PHP configuration) |
| RAM required (self-hosted) | ~256 MB | ~2 GB (full LAMP stack) |
| Self-hosted licence | AGPL-3.0 (free) | GPL-3.0 (free) or Matomo Cloud (paid) |
| Real-time dashboard | Yes, native | Yes (via additional plugin) |
| API & integrations | Basic REST, webhooks | Full REST + WooCommerce, WordPress, Zapier, Tag Manager |
Matomo: power for data teams
Matomo (formerly Piwik) is the oldest and most comprehensive alternative to Google Analytics in the open-source ecosystem. Founded in 2007, it offers near-complete functional parity with Universal Analytics: multi-step conversion funnels, heatmaps, session recording, A/B testing, advanced audience segmentation, multi-touch attribution and e-commerce reports compatible with WooCommerce and Shopify. Matomo's richness comes at a cost: complexity. A self-hosted deployment requires a full stack — MariaDB or MySQL, PHP 8.x, nginx or Apache, and ideally Redis for caching. The JS tracking script exceeds 22 KB and, in its default configuration, sets a session cookie that French CNIL considers subject to consent. A proper GDPR configuration requires enabling IP anonymisation (last two octets), reducing the retention period, configuring opt-out, and documenting these choices. Matomo has also offered a 'cookieless' option since version 4.0, at the cost of a slight loss of precision on multi-page sessions. In this mode, combined with hosting on a EU VPS, Matomo can operate without prior consent — provided you validate the configuration with your Data Protection Officer (DPO). For marketing teams that need attribution reports, complex goal tracking or integrations with CRM tools, Matomo remains the open-source benchmark. Its marketplace offers over 100 official and community plugins.
GDPR compliance: what the texts actually say
The GDPR does not mention Google Analytics, Plausible or Matomo. It governs the processing of personal data — and it is the analysis of each tool in its deployment context that determines its compliance. To be lawful, a processing activity must rest on a legal basis (Article 6 GDPR). For analytics cookies, the French CNIL accepts legitimate interest only if the data collected is strictly necessary, minimised and not cross-referenced with other processing activities. In practice, it is Matomo's 'full anonymisation' mode or Plausible's cookieless mode that allow reliance on this legal basis. Transfers of data outside the EU are governed by Chapter V of the GDPR. Hosting your analytics on a VPS located in the European Union — with a provider whose parent companies and sub-processors are also in the EU — eliminates this transfer risk by design. This is the main technical argument for self-hosted solutions over American SaaS tools. Finally, the minimisation principle (Article 5.1.c) requires collecting only data strictly necessary for the declared purpose. If your goal is to measure traffic and popular pages, Plausible is sufficient and naturally minimises collection. If you need conversion funnels to manage a marketing team, Matomo is justified — provided you document this necessity in your processing register.
Tip: combine both tools based on traffic type. Deploy Plausible on your public-facing site for anonymous traffic — lightweight, cookie-free, no consent banner. Activate Matomo only in authenticated spaces (customer portal, back-office) where users have an established contractual relationship with you. This hybrid architecture gives you the lightness of Plausible for SEO and the power of Matomo for conversion analysis, while minimising the personal data surface collected.
How to deploy in 15 minutes with Docker on your VPS
Both tools have official Docker images and maintained docker-compose.yml files. Here are the key steps for a quick deployment on a Debian 12 VPS.
Plausible Analytics CE: clone the official repository (plausible/community-edition on GitHub), copy plausible-conf.env.example to plausible-conf.env and fill in four variables: BASE_URL, SECRET_KEY_BASE (generated via openssl rand -base64 48), DATABASE_URL (PostgreSQL included in the compose file) and CLICKHOUSE_DATABASE_URL. Run docker compose up -d. Plausible listens on port 8000; place nginx as a reverse proxy with a Let's Encrypt certificate via Certbot. First login, create the admin account, add your domain — you are in production.
Matomo: the official matomo image on Docker Hub comes with a MariaDB compose file. Key variables are MYSQL_DATABASE, MYSQL_USER, MYSQL_PASSWORD and MATOMO_DATABASE_HOST. The data volume must be mounted on the host for persistence. After docker compose up -d, Matomo's web wizard guides installation in five steps. Then configure GDPR settings under 'Privacy': IP anonymisation, retention period capped at 13 months, cookieless mode if required.
In both cases, a VPS with 2 vCPU and 4 GB RAM comfortably hosts both tools simultaneously, with headroom for medium-sized site traffic (up to 500,000 monthly page views).
Verdict: which one to choose for your situation
The choice between Plausible Analytics and Matomo depends less on the intrinsic quality of each tool than on your actual needs and operational capacity. Choose Plausible if: you manage a showcase site, blog or portfolio whose primary goal is to measure traffic and popular pages; you do not have a dedicated developer to maintain an application stack; you want GDPR compliance without configuration or a DPO; your VPS is RAM-constrained (256 MB is sufficient); or you value a dashboard readable in thirty seconds. Choose Matomo if: you run an online store and need cart tracking and conversion funnels; your marketing team uses audience segments, attribution reports or A/B tests; you need to integrate your analytics with a CRM or marketing automation tools; or you want to migrate existing reports from Google Analytics with minimal loss of granularity. Choose both if you have a public-facing space (Plausible) and an authenticated space (Matomo) — this is the architecture that maximises compliance while preserving analytical depth where it is justified. In either case, hosting on a ServOrbit VPS located in Europe guarantees that your data never leaves the European Union — which is, in 2026, the necessary condition for any genuinely GDPR-compliant analytics strategy.
