A site that updates itself, and checks every page before it does
A new affiliate site publishes fresh data twice a day and new content every morning, with no one at the keyboard. How the automation works, and the one rule that makes it safe to leave alone.
A content site that depends on daily data has a problem: someone has to update it every day, forever. On a new US content and affiliate site we built in three weeks, we took that person out of the loop. The automation went live on 4 October.
Three runs a day
- Fresh data, twice a day. The site fetches the latest official data and publishes it. The second run catches anything published late.
- New content every morning. Three new posts from the site’s cast of characters, written with AI.
- A new guide three days a week, written by AI from official pages downloaded for the purpose, taken from a queue of planned topics.
The rule that makes it safe
An automatic run does not get a shortcut. Every run builds the whole site and checks every page against the same rules as a human change: required notices present, no banned claims, partner offers disclosed, AI-written content labelled. If any page fails, nothing goes live.
Small details that keep it quiet
- Only publish when something changed. A run with nothing new ends without deploying.
- Everything is saved. New content is stored with the site’s history, so any post can be traced and undone.
- Grounded in sources. Guides are written from downloaded official pages, not from what the model happens to remember.
How we ship
Every automatic run builds the whole site and checks every page before anything goes live. Automation is only safe if it faces the same gate as people do.
The examples in this article come from real engagements. Client details withheld; every figure comes from the client’s own data.
Read the case study →