Publish to Hosted blog
The route for sites on closed builders, listing platforms and locked-down corporate stacks, and increasingly the one we would pick anyway. You point blog.yourdomain.com at Vupie with a single record we give you. Everything behind it, the rendering, the certificate, the images and the caching, runs on Cloudflare's edge network. You install nothing and you operate nothing.
Setup
How you connect it.
What you need
One DNS record on your own domain. That is the entire dependency.
-
1
Pick the subdomain, usually blog.yourdomain.com.
-
2
Add the CNAME record shown in the site's Integrations panel, pointing at the hostname we give you.
-
3
Wait for propagation. The panel confirms when the hostname resolves and the certificate has been issued.
-
4
Choose the look, and articles start landing on the calendar you approved.
On publish
What Vupie fills in for you.
Not just the body text. The fields that decide whether the page is indexable and linked are set on the way in, because an article nobody can find is not published, it is stored.
Before you pick this route
Worth knowing about Hosted blog.
It runs on the edge, and you run none of it
Rendering, caching and images sit on Cloudflare's network, so pages are served from the location nearest the reader without you provisioning a server, a CDN or a certificate. There is nothing to patch and nothing to keep up.
Built for a perfect performance score
No render-blocking scripts, no third-party tags, no cookie banner you did not ask for, images in modern formats sized at the edge. The target is 100 on Lighthouse performance with green Core Web Vitals on every article, and nothing in the stack is fighting it.
One DNS record is the whole dependency
No plugin, no platform access, no CMS credentials, nothing for your web agency to schedule. If your main site is governed by a change-control process, this is usually the only route that survives it.
You can leave with everything
Cancel and the blog keeps serving on your domain for twelve months, so no URL goes dark while you move. Every article exports in one click throughout, with its images, its schema and its redirect map, to your own stack or anyone else's.
Built in, not configured
Every best practice, on by default.
On every other route your CMS and your theme decide how the page is built, and we can only ask nicely. Here we own the whole stack, so the things that usually take an agency and a performance budget are simply how it ships.
01
Static HTML, served from cache
Articles are rendered ahead of time and served straight from the edge cache, so a reader is not waiting on a database, a plugin chain or a theme. The first byte is a file that already existed.
02
Nothing render-blocking above the fold
Critical CSS inlined, fonts preloaded and subset with a swap fallback, no client-side framework needed to read an article. The text paints before anything else has finished loading.
03
Images that cannot shift the layout
AVIF and WebP with a JPEG fallback, responsive srcset, explicit width and height on every image, lazy below the fold. Cumulative layout shift stays at zero by construction rather than by tuning.
04
No third-party tags by default
No analytics bundle, no tag manager, no chat widget, and therefore no cookie banner unless you add something that needs one. Most of what wrecks a performance score on a marketing blog is simply not there.
05
Crawlable and quotable
Semantic headings, Article and Breadcrumb schema, canonical URLs on your own domain, a clean slug per article, sitemap and robots.txt maintained for you, and AI crawlers explicitly allowed.
06
Transport tuned at the edge
HTTP/3, Brotli, immutable cache headers on assets and early hints on the document. The parts that are pure infrastructure are handled once, for every article, forever.
Core Web Vitals are a ranking signal and a crawl-budget factor, so this is not decoration. It is the part of on-page SEO that can be solved once at the infrastructure level instead of being fought article by article.
The same on every destination
The integration decides where. It does not decide what.
Whichever route you pick, the article behind it went through the same pipeline. Choosing Hosted blog changes the plumbing, nothing else.
Researched before written
Primary sources retrieved and tiered, your own numbers computed from your data, and the two or three questions only you can answer sent by SMS or Slack. The model may cite only from that list.
Reviewed by another vendor
Every draft is checked by a model from a different vendor than the one that wrote it. Sections below threshold are rewritten up to twice, and what still fails is refused rather than published.
Images, links and schema
Images with provenance metadata, internal links into your existing pages, and Article schema, all set through the integration rather than left for you to add.
Index inspection after publish
The URL is checked on days 2, 7 and 14 to confirm it was actually indexed, and AI-visibility tracking watches whether ChatGPT, Perplexity and Gemini start citing it.
Who publishes this way
Usually chosen by these industries.
Connect it once. Then it publishes
without you.
Setup is about five minutes, plus dns propagation, and it is the last time you have to touch it. After that, articles land on the calendar you approved.