MACHINE 404 / Do it yourself

How to check if AI can read your website

You do not need us for this part. Six checks, a browser, ten minutes. We publish them because the number of people who will do it themselves and the number who will pay somebody are different people, and pretending otherwise is how this industry got its reputation.

The six

1. Look for the data. Open your homepage, view source, search for application/ld+json. Nothing there means no machine has been told who you are. 2. Break a URL. Ask for a page that does not exist. Anything other than a 404 is a soft 404. 3. Read your robots.txt. It is at /robots.txt. Does it name any AI crawler at all, and does it declare your sitemap?

4. Find a price. Is the number text, or is it inside an image or a PDF? 5. Turn JavaScript off and reload. What is left is roughly what most AI crawlers read. 6. Ask an assistant about your own business by name and see what it says. If it is vague, wrong, or about somebody else, that is the answer your customers are getting.

What the free scan adds

It runs eight of the eighteen checks against your address in about ten seconds and shows you the result as data, the way an agent would hold it: what it found, what came back null, and how many of the eight fields exist at all. No email, no call, no account.

The other ten need a server to run properly, and the score out of 100 comes from all eighteen. That part is the paid report, along with the instructions written for your own platform and a re-scan afterwards so the before and after is a number instead of a claim.