ServicesAI visibility
Crawl rules, schema, content shape. AI that knows who you are.
When someone asks ChatGPT, Claude, Perplexity, or Gemini for a business like yours, something comes back. We set the crawl rules, write the schema, and shape the content so the model has something accurate to quote about you. Built into every site we ship, also available as a standalone pass on a site you already have.
Crawl rules, schema, content. Included with every site we build. Standalone quoted after a short audit.
Five steps from baseline read to second look.
Ask the chats about you.
We run a fixed set of prompts about your business across ChatGPT, Claude, Perplexity, and Gemini, then save the answers. That becomes the baseline. Most small businesses don't show up at all, or show up with the wrong facts. Either way we know what we are starting from.
Set the crawl rules.
robots.txt rules for each major AI crawler, plus an llms.txt and llms-full.txt written specifically for your business. You decide which models are allowed to read the site. Most owners say yes to all of them.
Build the entity layer.
Schema markup that names your business properly: full Organization or LocalBusiness, sameAs links to your Google Business Profile, social profiles, and any directory listings. The point is to give the model unambiguous proof that the business on the site is the same business it has seen elsewhere.
Shape the content for citation.
Rewrite the About page and the service pages so a model can pull a clean two-sentence answer out of them. Q&A sections where they fit. Plain, sourced facts. Models cite content they can quote without paraphrasing into nonsense.
Re-ask the chats.
Same prompts, same models, four to six weeks later. The before-and-after lands in a short PDF you keep. Anything still wrong gets a second pass.
Two ways in.
Included with every rebuild, or quoted as a standalone pass.
What that looks like
Built into every site we ship. llms.txt, AI bot crawl rules, entity schema, and citation-shaped content are part of the build, not a bolt-on. You get it without asking.
Standalone, for a site we did not build, we run the baseline read across the four major chat models and quote the work from there. The fee depends on how much rewriting the About and service pages need, and whether we can edit the site directly. The audit itself is free.
What is and isn’t in the work.
Included
- Baseline read across ChatGPT, Claude, Perplexity, and Gemini
- llms.txt and llms-full.txt written for your business
- robots.txt rules for the major AI crawlers, you pick which
- Schema markup for entity clarity (Organization, LocalBusiness, sameAs)
- About page rewrite shaped so a model can summarize you cleanly
- Service pages broken into Q&A sections where it fits
- Internal linking pass between entity-shaped pages
- Before-and-after comparison across the four major chat models
- Short PDF documenting what changed and what to watch
Not included
- Wikipedia article creation
- Paid placement inside any AI product
- AI content writing as a retainer
- Directory submission services
- Anything that promises a guaranteed mention or position
- Off-site SEO of any kind
- Automating your day-to-day workflows; that is AI & automation, a separate service
What owners ask before they let us at the site.
Don’t see yours? Ask us directly.
Isn't this just SEO?
There is overlap, but the signals are different. Google rewards links, time on page, and structured data tied to ranking. Chat models reward content they can read in one pass and quote cleanly, plus entity-level proof of who you are. Doing one well does not guarantee the other.
Will this make ChatGPT mention me?
Sometimes, for the right kind of query. If someone asks an AI chat for a business like yours in your city, the model needs something to point at. We make sure that something is accurate, current, and on your site instead of on a competitor's. We promise the work that makes a mention possible. Nobody legitimate guarantees the mention itself.
Should I block AI crawlers instead?
Most small businesses should not. The trade is simple: let models read the site and you might get cited, block them and you will not. There are reasons to block (paywalled content, sensitive customer data on the page) and we will tell you when one of those applies. For a service business with a public site, allowing the crawl is the right call.
What is llms.txt?
A short text file at the root of your site, written for language models, that tells them what your business is and which pages matter. It is not yet a formal standard but the major chat companies have started reading it. We write yours to match how we want the site summarized.
Can you do this on a site you didn't build?
Yes. The schema and the crawl-rule files we can usually add directly. The page rewrites depend on whether we can edit the site or have to send the copy to your current developer. We will tell you which after a short look at the platform.
Will any of this hurt my Google ranking?
No. Entity schema, llms.txt, and cleaner content are independently good for Google as well. The two jobs reinforce each other when done properly. They only conflict if someone is keyword-stuffing or cloaking content, neither of which we do.
How do you measure it?
We run the same set of prompts across the four major chat models before and after, and save the responses. The before-and-after sits in a short PDF you keep. We are honest that this is not as clean as a Google ranking report. The field is too young for that. The comparison shows clearly whether the chats now know who you are.
Curious what the chats currently say about you?
Send the URL. We run the prompt set across the four major chat models and email back what each one knows about you today. No call, no charge.