Open-source Yandex Metrica alternatives

PostHog is an open-source platform that bundles the tools product teams use to understand and improve what they ship. It captures event data, user sessions and errors in one place, and its newer self-driving mode aims to turn signals such as errors, rage clicks and failed queries into researched reports and pull requests that engineers review and merge.

The product suite includes product analytics with autocapture or manual instrumentation and SQL analysis, web analytics with a dashboard resembling Google Analytics, session replays for web and mobile, feature flags, experiments with statistical measurement, surveys, error tracking, logs, a data warehouse and a customer data platform. The back end is Python and the front end is React and TypeScript.

PostHog's repository license is listed as 'Other' on GitHub, so check which parts are open source and which are enterprise-only. It can be self-hosted, and the company also runs a managed cloud service. It suits product and engineering teams that prefer a single tool over separate analytics, flagging and replay vendors.

Umami is a privacy-focused web analytics platform and a common open-source alternative to Google Analytics. It brings traffic, campaign performance, user behavior, conversions and revenue into one dashboard, and it does so without cookies and without cross-site surveillance of visitors.

Topics show it also supports audience segmentation, cohort analysis and user journey views alongside standard statistics and charts. The app is built with TypeScript and runs on Node.js 18.18 or newer with a PostgreSQL database of version 12.14 or newer. Optional settings let you change the API base URL, enable two-factor authentication by providing an encryption key, and switch on an MCP endpoint, which is disabled by default.

Umami is MIT licensed and can be installed from source or run on your own infrastructure; the project also offers a hosted cloud version for teams that do not want to maintain servers. A detailed getting-started guide lives at umami.is/docs. It suits site owners and product teams who want clean, compliant analytics they control.

Plausible Analytics is an open-source web analytics tool built around privacy. It is a lightweight, cookie-free alternative to Google Analytics that measures traffic rather than individuals: it keeps no personal data or IP addresses and uses no persistent identifiers, and the project states compliance with GDPR, CCPA and PECR. The company is made and hosted in the EU.

The dashboard is deliberately simple, with the key insights on a single page and no layers of menus or custom reports to build. The tracking script is tiny, which helps pages load faster, and you can also send custom events to measure goals. The stack uses Elixir with Phoenix, PostgreSQL and ClickHouse, with Tailwind CSS for the interface.

Plausible is AGPL-3.0 licensed and comes in two forms: a managed cloud service run by the company, which is paid, and a self-hosted community edition you operate yourself. A live demo and documentation let you try the dashboard first. It suits site owners and marketers who want straightforward, compliant analytics without cookie banners.

Matomo, formerly known as Piwik, is an open web analytics platform written in PHP with MySQL storage. You download and install it on your own web server, and after a short installation you receive a JavaScript tracking tag to paste into the sites you want to measure, then see reports in real time. It positions itself as a free-software substitute for Google Analytics.

Privacy is built in, and its mission statement stresses giving every user full control of their data. According to the README it is used on more than 1,400,000 websites. Requirements are PHP 8.1 or newer, MySQL 8.0 or newer or MariaDB 10.6 or newer, and the PDO MySQL or MySQLi extension, and it is independent of operating system and web server. Topics also reference marketing and mobile analytics.

Matomo is released under the GPL-3.0 (or later) license. Self-hosting keeps data on your own servers, and the organization also offers Matomo Cloud as a managed option, with a marketplace of plugins. It suits organizations with data-protection requirements, such as public bodies and companies in privacy-regulated markets, that need dependable analytics they control.

Rybbit is an open-source web and product analytics platform that avoids cookies and aims to be easier to read than traditional analytics suites. It is presented as a privacy-friendly alternative to Google Analytics and can be set up in a couple of minutes with a tracking script added to a site.

It covers the usual web metrics, including sessions, unique users, pageviews, bounce rate and session duration, and goes further with session replays, funnels, retention and user journey dashboards. Data can be filtered across more than fifteen dimensions, custom events carry JSON properties, locations are tracked down to city level with map views, and a real-time dashboard is included. Organizations and an unlimited number of sites are supported.

Rybbit is licensed under AGPL-3.0 and written in TypeScript. You can use the hosted service for the quickest start or deploy it on your own VPS for full control, and a live demo shows it running on a production site. The README compares it with GA4, Plausible and Cloudflare analytics on openness, self-hosting and privacy.

OpenReplay is an open-source experience platform that you host on your own infrastructure. It combines session replay, DevTools-level debugging context, product analytics and live cobrowsing, so teams can see how real users navigate, where they struggle and what errors they hit without sending sessions to a third party.

Because everything stays in your own cloud, it is pitched at teams that cannot hand customer data to an external vendor. Data can be masked, obscured or skipped at capture time before it reaches your servers, and a private mode redacts everything by default. A lightweight tracker sends data asynchronously to limit the impact on page performance, and SDKs cover frameworks such as React, Angular, Vue, Svelte, Next.js and React Native.

The README states it can be self-hosted for free on AWS, GCP, Azure and other platforms. The repository lists its licence as Other and its topics also reference feature flags, web vitals and real-user monitoring, so it spans debugging, product and support use cases.

OpenPanel is a web and product analytics platform, released as open source, that sets out to combine the depth of Mixpanel with the simplicity of Plausible. It is presented as an alternative to Mixpanel and Google Analytics, with optional self-hosting and cookieless tracking by default.

Its feature list covers funnels and cohorts, user profiles with session history, session replay that has privacy controls, real-time dashboards, built-in A/B testing, event and funnel alerts, custom dashboards, revenue tracking for purchases, subscriptions and lifetime value, SDKs and an API for web, mobile and server-side tracking, integrations such as Google Search Console, and an MCP server so AI clients can query user data. A comparison table in the README sets it against Mixpanel, GA4 and Plausible.

OpenPanel is written in TypeScript and licensed under AGPL-3.0. It can be self-hosted for full control of data, and a hosted version with sign-in is also available. The project emphasizes transparent pricing with no hidden costs or usage limits. It suits product teams that want privacy-friendly analytics with product-level depth.

Countly is a product analytics and customer engagement system for organizations that insist on owning their data. Teams use it to see how people behave in mobile, web, desktop and connected products, to improve product experiences, and to automate and personalize how they engage with customers.

The repository topics reflect a broad toolset: web and mobile analytics, crash analytics, push notifications, feature flags, remote configuration, user feedback, dashboards and user journeys. Privacy is a central theme, with references to GDPR, COPPA and HIPAA, and the README stresses that, unlike SaaS-only tools, it can run on your own premises or inside a private cloud. SDKs are provided for many platforms, and the project has a Discord community.

The server code is JavaScript and published under a license that GitHub lists as 'Other', so check the repository for the terms that apply to the edition you use. Countly also offers hosted and enterprise options through the vendor. It suits product and marketing teams that need analytics without sending data to a third party.

Clarity is a behavioral analytics library written in TypeScript with two main goals: privacy and performance. It watches DOM mutations and user interactions in the browser to generate aggregated insights about how visitors view and use a website, across modern devices and browsers. It is released under the MIT license.

The data it captures supports session replay and heatmaps, which help product and design teams see where features succeed or fail. The README says the same code powers Microsoft's hosted behavioral analytics service at clarity.microsoft.com, where a live demo is also available, so the repository represents the client-side collection library rather than a full self-contained analytics backend.

The community is invited to help build a privacy-first, high-performance behavioral analytics library. It suits web teams who want to understand user behavior, improve conversion and retention, and review the code that runs on their pages before adding it to a site.

Swetrix is an open-source web analytics platform presented as a privacy-respecting alternative to Google Analytics. It is cookieless and designed to comply with GDPR without consent banners, with anonymized data and no cross-device tracking, and it has been developed in the open since 2021.

Beyond traffic statistics such as top pages, locations, devices, traffic sources and UTM campaigns, it offers custom events and properties, sessions and user flows, funnels, real-user performance metrics, client-side error tracking and session replays. The README says some features, such as network intelligence and session replays, are available on the cloud service. A lightweight tracking script feeds a real-time dashboard.

Swetrix is built with TypeScript, NestJS, Redis, MySQL and ClickHouse and is licensed under AGPL-3.0. You can try it on Swetrix Cloud, which is free to start, or self-host it with Docker. A live demo and documentation are available. It suits website owners and product teams who want analytics, error tracking and performance data in one privacy-focused tool.