Plausible Analytics

Plausible is the best-known of the lightweight privacy-first analytics tools. It is built in Elixir on ClickHouse, sets no cookies and uses no local storage, and counts unique visitors with a daily identifier — a hash of a daily salt, the site domain, the visitor’s IP, and their user agent. The salt is rotated and deleted every 24 hours and raw IPs and user-agent strings are never stored, so there is no persistent identifier and no cross-day, cross-site, or cross-device visitor stitching by design. Everything is stored as anonymized aggregates: browser, OS, device type, and geography down to city.

The data model is pageviews plus custom events. On top of that sit goals, custom properties (metadata attached to events and pageviews), and prebuilt trackers for outbound links, file downloads, form submissions, and 404 pages. Funnels, ecommerce revenue goals, and the Stats API are gated to the Business plan on cloud; Starter covers one site with three-year retention, Growth adds sites and team members, Business extends retention to five years. Everything renders on a single dashboard page — there is no report builder, which is most of the appeal and most of the ceiling.

Implementation is a one-line script, with official routes for WordPress, Google Tag Manager, and most site builders (Shopify, Webflow, Squarespace, Wix, Ghost, Hugo, Next.js). Because many adblockers block third-party analytics domains, the docs describe a first-party proxy setup to serve the script and events endpoint from your own domain — plan for this if undercounting matters to you. Programmatic access comes via the Stats, Events, and Sites APIs, and there is a Looker Studio connector plus a Google Search Console integration for search-query data alongside your stats.

Migration from GA4 is a real, supported path: Plausible imports historical GA4 data day-by-day in aggregate (up to five properties into one dashboard), stitched to end where your Plausible data begins. Expect gaps — imported data is day-level only, some dimensions (exit pages, for example) are not exposed by the GA API, and imported numbers will not reconcile exactly with native ones because the two products count visitors differently. Import before GA4’s retention window expires.

The company is EU-based and states that all visitor data is processed and stored in the EU on infrastructure owned by European companies; it publishes its position on operating without cookie-consent prompts — worth validating against your own counsel rather than taking as given. The AGPLv3 Community Edition ships as Docker images and makes the core product self-hostable, but it is a long-term release published twice per year, is community-supported only, and lacks funnels, ecommerce revenue goals, SSO, and the Sites API; its bot filtering is also more basic than the cloud’s.

Use it when you need clean traffic, referrer, and conversion reporting with minimal script weight and a defensible privacy posture. Look past it when you need user-level journeys, long-window retention and cohort analysis, or deep segmentation across many custom dimensions — the daily-rotating identifier rules those out on purpose.