Cloudflare moves corporate blog to EmDash CMS with latency cut of nearly half
Cloudflare Inc. moved its corporate blog to EmDash, a content management system built on Astro, on Aug 12, cutting p95 page-latency by roughly half compared with its legacy platform.
Source: Cloudflare Blog · August 24, 2026 at 9:01 PM · AI-assisted report
Single-source
KUALA LUMPUR, 25 AUGUST 2026 —
Cloudflare Migrates Blog to In-House CMS in Bid to Prove Scalability, Performance
Market Impact
KUALA LUMPUR, Aug 24 — Cloudflare has completed the migration of its official blog to EmDash, an in-house content management system (CMS) built on Astro, as part of a broader effort to validate performance, scalability, and usability under real-world conditions. The move, executed on August 12, 2026, represents a key milestone in Cloudflare’s long-standing practice of using its own products internally before offering them to enterprise customers.
The migration was not merely a redesign—though the blog now features a modernized interface, dark mode, and improved navigation—but a full-scale stress test of EmDash under Cloudflare’s demanding traffic patterns. The blog typically handles around 75 requests per second (RPS) under normal load, but has recorded spikes exceeding 5,000 RPS during viral post releases. Such variability demands a platform capable of handling sudden surges without degradation in performance or user experience.
“At Cloudflare, we are our own first, most demanding customer,” the company said in a blog post. “We validate scale, security, and usability on our own massive infrastructure before a paying customer ever touches the product.” This internal philosophy, codified in the company’s engineering standards known as the Codex, ensures that any product adopted by Cloudflare must first prove itself in the harshest production environment.
EmDash, still in pre-1.0 development at the time of migration, underwent rigorous testing. While the platform performed well overall, several gaps were identified—most the lack of scheduled post functionality, which was only resolved in EmDash version 0.19.0. “This gap was understandable given the early version of EmDash, but it was also definitely something we didn’t want to be finding out after the scheduled time for a post,” the company noted.
The migration also provided an opportunity to modernize the frontend experience. The blog’s interface was rebuilt using Cloudflare’s Kumo design system, aligning it with the visual language of the company’s homepage, dashboard, and marketing sites. Native support for light and dark modes was introduced, with theme switching tied to system preferences and accessibility standards strictly enforced.
Long-standing UX issues were addressed, including the relocation of the email subscription form from the top-right corner—where users frequently mistook it for a search bar—to a dedicated call-to-action block at the end of posts.
Technical enhancements included the addition of a dynamic table of contents on longer posts and a “Discuss Online” sidebar to facilitate community engagement across social and developer platforms. These changes were designed to improve readability and interactivity, particularly for technical audiences who rely on the blog for in-depth insights.
To ensure zero downtime during the migration, Cloudflare deployed a proxy Worker using Cloudflare’s edge network. This Worker intelligently routed traffic between the legacy blog and the new EmDash-powered site by setting a version cookie on incoming requests. In the event of a 500 error, the system automatically fell back to the legacy platform. The use of service bindings allowed direct worker-to-worker communication, reducing latency and improving response times for end users.
The rollout was executed in phases, beginning with just 1% of total traffic and gradually increasing to 5%, 15%, and beyond as system health was validated. By the end of the day, 100% of traffic had been successfully shifted to the new platform. Early performance data indicates significant improvements in response times, with the company reporting faster page loads and reduced load on backend systems.
Cloudflare’s caching architecture played a critical role in maintaining performance. With multiple layers of caching in place—ordered by proximity to the user—the company now serves 99.5% of static files and 70% of all requests from cache. This setup not only enhances speed but also improves resilience, ensuring consistent availability even under heavy load.
The migration of the Cloudflare Blog to EmDash underscores the company’s commitment to self-validation and continuous improvement. By using its own infrastructure as a testing ground, Cloudflare aims to deliver more reliable, scalable, and user-friendly platforms to its global customer base. The lessons learned from this project will likely inform future iterations of EmDash and other in-house tools.
As for regional implications, Malaysian enterprises and digital publishers—particularly those managing high-traffic content platforms—may take note of Cloudflare’s approach. The use of edge computing, phased rollouts, and automated fallback mechanisms offers a model for managing large-scale content migrations with minimal disruption. With the rise of AI-driven content and real-time analytics, platforms that can scale seamlessly while maintaining performance will be critical to staying competitive in the digital economy.
Looking ahead, Cloudflare plans to further refine EmDash based on feedback from the migration. While the company has not disclosed specific timelines for future updates, the successful transition of its flagship blog suggests that EmDash is on track to become a, enterprise-ready CMS. For now, readers of the Cloudflare Blog can enjoy a faster, more intuitive experience—one that reflects the company’s own standards for performance and reliability.