- Know the six things AI needs to find on a dispensary website
- Stop paying for the parts Google says do nothing
- Generate your schema, llms.txt and robots.txt in two minutes
of Perplexity's answers and 92% of Gemini's cited at least one dispensary's own website. ChatGPT did in 12%, which is why lessons 3, 4 and 7 matter too.
200 AI answers, 50 cities, October 2026.Your website is the only source about your store that you fully control. When an AI tool reads it, it is looking for the same things a new customer wants in the first ten seconds: what this place is, where it is, when it is open, what it sells, and why someone would pick it. If those answers are in plain text on the page, they get used. If they are inside an image, a pop-up or an embedded menu, they do not exist.
Six things AI needs to find
- Who, what and where, on the first screen. One plain sentence near the top of your homepage: "Your store is a licensed dispensary in [neighborhood], [city]." Then your address, today's hours and a menu button. That sentence is what AI tools quote.
- A real menu in the HTML. Category pages for flower, pre-rolls, vapes, edibles and concentrates, with product and brand names as text. An iframe-only menu is an empty page to a crawler.
- An age gate that does not block reading. The gate should sit on top of the page, with the real content underneath in the HTML. No redirects, no blank page behind the button.
- Answers to real questions. The questions from lesson 2, each answered in the first two sentences: what to bring, purchase limits, delivery area, parking, payment, first visit.
- Proof you are real. An About page with your license number, how long you have been open, the neighborhood, and the people behind the counter. A reviews page from lesson 5.
- Your one-line reason, everywhere. The thing you want AI to say about you from lesson 3, in your homepage headline, page titles and About page.
Structure the site like your Google profile: the homepage is the store, category pages are the aisles, and each location gets its own page if you have more than one.
What you can skip
A lot of "AI SEO" sold to dispensaries is files and markup that Google says it ignores. Here is the honest version:
| Thing | Does Google's AI use it? | Worth doing? |
|---|---|---|
| Crawlable pages with real text | Yes. This is the whole game | Yes, first |
| robots.txt allowing AI search bots | Google uses its normal crawler. Others need their own bots allowed | Yes, two minutes |
| Schema markup (Store) | Not required for AI features, still read for facts and rich results | Yes, five minutes |
| llms.txt | No. Google confirmed in June 2026 | Optional. Never pay for it |
| Rewriting pages into tiny "AI chunks" | No. Google reads whole pages | No |
Let the right bots in
Each AI company runs separate crawlers for search and for training. Block the search ones and you disappear from that tool's answers. OpenAI says OAI-SearchBot decides whether your site can appear in ChatGPT search. Blocking GPTBot, its training crawler, does not change that. Check yours by opening yoursite.com/robots.txt in a browser. If you see Disallow: / under OAI-SearchBot, PerplexityBot or Bingbot, fix it today.
The generator
Fill in your facts once. The generator writes three files: schema for your homepage, an llms.txt and an AI section for your robots.txt. If you came from your back office, your details are already in. Everything stays in your browser.
Your store facts
Saved on this device1. Schema for your homepage
Paste this into the <head> of your homepage only. Most website builders have a "custom code" or "header code" box for it.
2. llms.txt
Save as a plain text file named llms.txt at the root of your site, so it opens at yoursite.com/llms.txt. Edit the key page links to match your real pages.
3. robots.txt, AI section
Add these lines to your existing robots.txt. If you do not want your content used to train AI models, keep the two training blocks. Removing them does not change whether you appear in AI search answers.
Check your work
- Paste your homepage address into the Schema Markup Validator at validator.schema.org. It should find your Store with no errors.
- Open
yoursite.com/robots.txtandyoursite.com/llms.txtin a browser. Both should load as plain text. - Ask Perplexity "what is your store known for?" in a week. If the answer quotes your site, it is working.
Common questions
Does schema markup help me rank in AI answers?
Not directly. Google says there is no special markup for its AI features. Schema still helps machines read your facts cleanly and keeps you eligible for rich results, so it is worth five minutes, but it will not rescue a thin site.
Is llms.txt worth adding?
It is optional. Google said in June 2026 that Search does not use it. Some AI tools and agents may read it. It takes two minutes with the generator below, so add it if your site makes it easy, and do not pay anyone for it.
Does my age gate stop Google and AI from reading my site?
It can. If your age gate redirects visitors to another page or hides the content until a button is clicked in a way crawlers cannot pass, they see nothing. A gate that sits on top of a normal page, with the real content in the HTML underneath, is safe.
Is my embedded menu hurting me?
If your menu only exists inside an iframe or an app widget, search engines and AI tools see an empty page where your products should be. Real category pages with product names in the HTML give them something to read and quote.
Finished this lesson?
Mark it done. Your progress saves in this browser, no account needed.
