Cloudflare has announced the open-sourcing of the BEACON dataset, a massive collection of anonymized Real User Monitoring (RUM) performance data now publicly available on Google BigQuery. The dataset contains billions of real-user measurements, offering an unprecedented view of web performance in the wild. Developers can query metrics such as Core Web Vitals (LCP, FID, CLS) and newer soft navigation metrics, broken down by browser, region, and other dimensions.
Key details of the release:
- BEACON dataset is now publicly accessible on Google BigQuery, allowing anyone to run SQL queries against billions of anonymized RUM records.
- Includes real-world Core Web Vitals metrics like
LCP,FID, andCLS, as well as emerging soft navigation metrics that capture performance during client-side route changes. - Provides performance breakdowns across browsers (e.g., Chrome, Firefox, Safari) and regions, enabling granular analysis of how performance varies globally.
- Data is anonymized to protect user privacy, with no personally identifiable information included.
This release matters to developers because it democratizes access to large-scale performance data that was previously locked within individual organizations or proprietary tools. By querying BEACON, teams can benchmark their own site’s performance against global baselines, identify regional or browser-specific bottlenecks, and validate optimizations with real-world evidence. It also supports research into web standards and helps drive improvements in user experience across the web. Whether you’re debugging a slow checkout flow or evaluating the impact of a new caching strategy, BEACON offers a data-driven foundation for decision-making.