HOW FAST IS THE WEB? EXPLORE BILLIONS OF REAL-USER MEASUREMENTS WITH BEACON

**Hacker Pranks Exclusive: Unveiling the Web's Performance Secrets with BEACON**

As tech enthusiasts, we often take for granted the seamless online experiences we enjoy on our high-end devices. But what about the billions of users who struggle with slow loading times, connectivity issues, and poor website performance? A recent breakthrough by Cloudflare, a leading cybersecurity company, sheds light on the web's performance secrets with the BEACON dataset. This anonymized collection of billions of real-world performance measurements across 10,000 of the largest websites on the Cloudflare network offers unparalleled insights into the web's performance. In this article, we'll dive into the fascinating world of web performance, exploring the key findings, implications, and opportunities for researchers, developers, and security experts.

**Closing the Perception Gap with Objective Data**

The BEACON dataset is a culmination of Cloudflare's efforts to collect telemetry data for its customers, providing a detailed understanding of how real users experience their sites. By expanding this dataset to include every major browser engine, Cloudflare has created a massive, publicly available resource for researchers and developers. The dataset includes Core Web Vitals, a set of metrics measuring perceived performance on the web, including:

1. **LCP (Large Contentful Paint)**: The time it takes for the main content to load. 2. **INP (Interaction to Next Paint)**: The time it takes for an interaction (like a click) to become visible on the screen. 3. **Cumulative Layout Shift**: The visual stability of a webpage.

These metrics provide a comprehensive view of web performance, allowing researchers to examine not only the speed but also the responsiveness of websites.

**Global Performance Variations**

The BEACON dataset reveals striking variations in web performance across different regions, browser engines, and industries. WebKit, the browser engine used in iOS, performs best overall but struggles in certain countries. In 46 countries where WebKit represents more than 10% of traffic, its LCP, INP, or both are at least 10% worse than those of Blink-based browsers like Chrome, Edge, and Opera.

Interestingly, some industries, such as Government and Politics, Health, and Safe for Kids, outperform others, like Ads, Religion, and Weather. This data highlights the importance of understanding the specific performance needs of different sectors and optimizing accordingly.

**Soft Navigations and Application Architecture**

Cloudflare's support for Google Chrome's Soft Navigations API provides accurate LCP reporting for client-side navigations used in single-page applications. The results show that soft navigations render two to three times faster than hard navigations at every percentile. However, this advantage is offset by the heavier initial load of the landing page, which may not be justified if users rarely progress beyond it.

**Combining Data for Deeper Insights**

One of the most exciting aspects of the BEACON dataset is its potential for combination with other sources. Researchers can join BEACON with datasets like the World Bank Group's measure of GDP per capita to explore the correlation between economic conditions and web performance. This approach has already been applied to the Cloudflare Radar platform, which will feature correlations between the Internet Quality Index (IQI) data and BEACON data.

**Publicizing the BEACON Dataset**

Cloudflare's commitment to transparency and open data is evident in the BEACON dataset. The company has stripped out any potential customer website identifiers, ensuring that the community can access valuable insights without compromising user trust. The dataset is publicly available on Google BigQuery, and Cloudflare encourages researchers, developers, and security experts to contribute to the project.

**Conclusion**

The BEACON dataset offers a unique opportunity for the web performance community to come together and address the challenges of slow loading times, poor responsiveness, and connectivity issues. By leveraging this massive collection of real-world performance measurements, researchers can identify areas for improvement and optimize website performance for a more seamless user experience. Join the conversation and contribute to the BEACON project by accessing the dataset on Google BigQuery and sharing your findings in the Cloudflare community forum or on Discord.

Stay tuned for future updates on the BEACON project and the web performance community's efforts to make the web faster and more accessible to everyone.