I audited 50 business websites for AI visibility — here's what I found
I checked 50 local business and SaaS websites to see if AI models could recommend them. The results were bad.
0 out of 50 had an llms.txt file
3 out of 50 had proper schema markup
47 out of 50 were actively blocking AI crawlers in robots.txt
When someone asks ChatGPT "who's the best plumber in Phoenix" or "what's the best project management tool for agencies" — these businesses literally cannot be recommended. They're invisible.
The fix isn't complicated. Three systems make your site readable and citable by AI:
System 1: llms.txt — A plain-text file at your domain root that tells AI exactly what you do, where, and for how much. 30 minutes to create.
System 2: Markdown Mirrors — Clean, script-free versions of your pages that AI can parse without fighting through cookie banners.
System 3: Sitemaps + Search Console — Tell crawlers which pages exist and matter, then use the data to find quick wins.
The businesses that set this up now get a structural advantage. AI models favor sites that are clean, direct, and easy to summarize. Most business websites are the opposite.
I built a complete playbook covering all three systems plus a SaaS-specific track. No coding required for the first three systems.
This window won't last forever — but right now, it's wide open.
