Cloudflare moved its own blog to EmDash and version 1.0 is close
Cloudflare said it migrated its own blog to EmDash, a content management system built to work on Astro and with its own infrastructure. The post shares traffic and performance figures from the migration completed on August 12 and notes that version 1.0 is approaching.
An infrastructure company tested its own content system in production
Cloudflare said it has moved its own blog to EmDash. In the company description, EmDash is a content management system built especially to work on Astro and with Cloudflare.
The migration progressed in stages and completed on August 12. The post explains that the move was not limited to infrastructure, and that the front end also got a refresh including dark mode support.
Per Search Engine Journal reporting dated August 24, 2026, EmDash launched as a 0.1.0 developer preview and is now approaching version 1.0. Cloudflare publishing its own high traffic blog on the system signals that the product is considered production ready.
The traffic and performance figures shared
The post gives the blog load profile as follows: normal load sits in the neighborhood of 75 requests per second, while spikes can go above 5,000 requests per second.
In the first phase after migration the blog served up to 450 requests per second without noticeable issues, and overall handled up to 850 requests per second.
On resilience, the standout number is this: a 28,000 requests per second denial of service attack on August 10 was absorbed without issues.
The example given for traffic volume is the Agents Week period. Per the post, the blog took close to 3 million pageviews across 18 posts over 9 days.
On caching, 99.5 percent of static files and 70 percent of all requests were served from cache. On latency, the post describes the 95th percentile response times as showing a flat and consistent profile instead of the periodic spikes seen in earlier periods.
The architectural claim behind EmDash is plugin security. Per Search Engine Journal the system runs plugins sandboxed through Dynamic Workers and is written entirely in TypeScript.
The default in content infrastructure is softening
The assessment in this section is ours. Corporate blogs have had one default for a long time. For an alternative to be taken seriously, being fast is not enough. The editor experience, the plugin ecosystem and the security model all have to hold up.
That is why Cloudflare moving its own blog matters. A company running its own product on its own critical publication is a different kind of evidence than a lab measurement. What happens under a traffic spike and under attack becomes visible.
The second point is the security model. A notable share of the trouble in the WordPress world comes from plugins. A design that runs plugins sandboxed means that surface gets smaller. It is no coincidence that plugin and admin experience topics were on the WordPress agenda the same month, since competition is concentrating on the same point.
The third point is dependency. EmDash is designed to run with Cloudflare infrastructure. That is an advantage for performance and a tie worth thinking about for portability. The question when picking a content management system is not only speed, it is the cost of leaving.
The fourth is maturity. Version 1.0 has not shipped and no date has been given. Rather than moving a corporate site there today, the healthier route is watching how the plugin ecosystem develops after the release.
What it means for businesses in Türkiye
The assessment below does not appear in the sources, it is our reading. The announcement contains no Türkiye specific information.
In Türkiye the large majority of corporate sites and blogs run on WordPress. That gives agencies and content teams familiar ground, and it also brings a known maintenance load: plugin updates, security patches and performance tuning.
An option like EmDash maturing could reduce that maintenance load in the medium term. For today our practical advice is to wait. Moving a corporate site before 1.0 ships, before the Turkish content and editor experience is tested, and before backup and migration paths are clear would be early.
The second point is team skill. Finding a content editor who knows WordPress is easy in Türkiye. Moving to a new system means training the people who produce content too. If the technical gain is offset by editor slowdown, there may be no net benefit.
The third is where slowness actually comes from. Most of the time a site is slow because of unoptimized images, unnecessary plugins and heavy page construction rather than the content management system. Fixing those three items on the existing site is usually faster and cheaper than switching.
The fourth is watching this space. An infrastructure provider entering content management could change the cost structure of web publishing in the coming years.
The UNALSOFT take
In web design projects we pick the content management system by the content rhythm of the business, not by fashion. A business publishing one post a week and an ecommerce site adding five products a day do not need the same thing. We are watching EmDash with interest, and the security model that runs plugins sandboxed is valuable to us. Even so we are not recommending a migration to a client today: moving before 1.0 ships and before the Turkish editor experience and backup paths are tested can cost more than it returns. We try a new system on our own internal projects first and move it to a client site later. That order never changes.
What foundation is your site standing on?
Let us measure the speed and maintenance load of your current setup.