Building Yeti Editorial Studio — From a Website URL to a Researched Editorial Feature

One of the things I have been working on recently is Yeti Editorial Studio, a new SaaSYeti application designed to turn a website into something more useful than another generic AI-written blog post.

The idea started with a fairly simple question:

Could AI take a website URL, research the subject properly and turn what it finds into a balanced, publication-style editorial feature?

That has gradually grown into Yeti Editorial Studio.

Rather than simply asking AI to write about a business, product or person, the application is being developed around an editorial workflow. It can research the subject, create a longer-form feature of around 1,500 words and prepare it for publication in the SaaSYeti Spotlight.

SaaSYeti Spotlight

Spotlight is becoming a small digital magazine inside the SaaSYeti ecosystem.

Articles can include:

• a professionally structured editorial feature;
• an AI-created hero image and supporting visuals;
• magazine-style typography and layout;
• sources and research behind the story;
• a preview before anything is published;
• Facebook and Open Graph sharing using the same hero artwork;
• verified profile claims for people featured in Spotlight.

One rule I particularly wanted to build into the system is that a person can delete their own private articles, but once an article has been published in Spotlight it becomes part of the publication record and cannot simply disappear later.

That makes Spotlight feel more like a real publication rather than just another content generator.

Making it easier to request a feature

I have also been working on the public Spotlight request process.

Instead of presenting somebody with a large intimidating form, the current version starts with three very small steps:

What would help? → Your website → About you

Once somebody completes the request, the system can immediately use AI to create a practical personalised tip based on the information they provided.

So even before I have reviewed their request, they receive something potentially useful.

The request also clearly asks permission before a future article can be published or shared with members of my Build SaaS with AI Facebook community.

Where this is heading

The next big part is research.

I do not want Editorial Studio to rely entirely on whatever information happens to be visible on one website.

I am therefore separately building a shared Yeti Research Engine using Crawlee and Playwright.

The aim is for that engine to investigate several useful sources, collect evidence and then pass structured research back to applications such as Editorial Studio.

Eventually the workflow should look something like:

Website → Research → Evidence → Editorial Story → Spotlight Publication

What I find interesting about this project is that it started as an article generator, but it is gradually becoming something quite different:

an AI-assisted editorial and research system.

And, as with most of the SaaSYeti projects I am building, the useful parts of the technology should eventually be reusable across several other Yeti applications rather than being locked inside one product.

Yeti Editorial Studio:
https://editorial.saasyeti.com