About This Site

A reference on how answer engines retrieve, select and attribute the sources behind a generated answer, written for people who have to make decisions about their own content.

What this site covers

King of Answer Engine documents four systems that compose answers from retrieved web content: Google AI Overviews, ChatGPT search, Perplexity and Claude. Each entry describes what the vendor publishes about its crawlers and its behavior, and says clearly where public documentation stops.

The site exists because most writing on this subject blends three things that should be kept apart: what the engines document, what practitioners have observed, and what someone is selling. Keeping them separate is the entire editorial approach here. Where something is documented, it is attributed to the vendor's own page. Where something is inferred from observation, it is labeled as inference. Where nothing is known publicly, the text says so instead of filling the gap.

Who publishes it

This site is published by Allan Oliveira, an SEO and AEO specialist and the founder of the agency SEOMais. He writes about answer engine optimization at allanaeo.com.

The site takes no paid placement. No vendor described here has reviewed, funded or approved any of it, and there are no affiliate arrangements with any of them. Links to vendor documentation exist so a reader can verify a claim, not as endorsement.

How claims are sourced

Every factual statement about a product on this site is drawn from that vendor's own published documentation, linked at the foot of the relevant page. These products change often, and documentation that was accurate on the review date may not be accurate when you read it. Each page carries the date it was last reviewed. When a claim and a vendor's current documentation disagree, the vendor's documentation is right and this site is out of date.

Corrections

If something here is wrong, it should be corrected rather than defended. The fastest way to establish that is a link to the vendor documentation that contradicts it.

What this site is not

It is not a ranking of the engines, a tool review, or a newsletter about model releases. Those exist and several are good. This site covers the narrower question of how these systems treat the sources they read, which changes less often than the products do and is poorly documented in one place.

It also makes no claim to insider knowledge. Everything here is drawn from public documentation, and where public documentation stops, this site stops rather than continuing in a confident voice.