from raw data to a shipped change
stop stitching together AI writing tools.
Notion, Claude and a CMS connector are clever, but they don't know your search data and never find out whether a page worked. Cogny drafts from your data, publishes to your CMS once you approve, and measures every page seven days later.
or from your own AI
claude mcp add --transport http cogny https://app.cogny.com/mcpcontent publishing inside your growth system
Most AI content workflows stop at the draft. The writing tool doesn't know which queries you almost rank for, which pages convert, or what your competitors cover. And after the page goes live, nothing checks whether it brought traffic, sign-ups or orders.
Cogny starts from the data. It reads Search Console, GA4 and your brand context, which is built from your website at signup, and drafts the page for a reason you can see. On approval it publishes to WordPress, Webflow, Sanity, Ghost, Framer, Shopify or WooCommerce. On Cloud it can also open a pull request on your site's repository, with a preview deployment.
Every published page becomes a ticket. Seven days later Cogny compares clicks, impressions and conversions before and after, and closes it with a verdict. Pages that underperform get a follow-up proposal instead of being forgotten.
Write, publish, measure, follow up. In one place.
where content workflows break
CMS editors eat the day
Blocks, formatting and SEO fields by hand, for every page.
Context switching
Analytics in one tab, writing in another, the CMS in a third.
No feedback loop
Pages go live and nobody checks whether they brought traffic or leads.
AI without growth context
Generic drafts that don't know your audience, your queries or your gaps.
how it works
- 01
Prompt
Say what to publish, or let Cogny propose a page from your search data.
- 02
Draft
Written with your brand context and the data behind the choice.
- 03
Publish
To your CMS after approval, or as a pull request on Cloud.
- 04
Measure
Seven days later in Search Console and GA4. A verdict, and a follow-up if needed.
published pages, measured
Example rows, not customer data.
inside your growth system
Growth context built in
Search Console, GA4 and Brand & Context inform every draft.
Publish, measure, ticket
Every page is a ticket measured seven days after it goes live.
Follow-up, not forgetting
Underperforming pages get a proposed next change.
One agent, full stack
The same agent can promote the page in ads or email, each change measured.
who this is for
Growth teams
Tired of switching between four tools to publish one page and never seeing whether it worked.
Marketing leads
Already drafting with Claude or Notion, and want the loop closed from publish to result.
Performance marketers
Writing pages for paid campaigns, and need them judged on conversions.
what it doesn't do
- It doesn't publish without approval. Pull requests never merge themselves.
- It doesn't promise rankings. It measures what each page did, misses included.
- Content loops that write on a schedule are a Cloud feature.
the record.
Published case studies, each one linked.
what it costs.
- Free $015 MCP calls a month and 50 one-time chat credits, no card
- Solo $9unmetered in your own AI, fair use
- Cogny AI $18150 credits
- Cogny Pro $901,500 credits
- Cloud $4995,000 credits, for teams
Per month. Solo, Cogny AI and Cogny Pro have the same features and one seat each; only the credits differ. Teams need Cloud. Measurement runs on every plan, even at zero credits. Compare plans →
write, publish, measure, follow up.