...

AEO / GEO case study

Canadian Cannabis Site Increased AI Referral Visibility in 1 Month

A Canadian cannabis company came to us with pages that were visible to people, but weakly structured for answer engines and AI assistants. After one month of AEO and GEO work, ChatGPT referral visibility became measurable in GA4.

The problem

The site had content, but AI systems did not have clean answers to cite.

The pages were written like traditional website copy: useful in parts, but not organized around direct questions, source-backed explanations, entity relationships, or structured data. For cannabis and CBD brands, this is a bigger issue because vague health language, unsupported claims, and thin service pages reduce trust for both users and search systems.

Unclear answer structure

Important information was buried in paragraphs instead of being presented as direct answers, comparison blocks, FAQs, and decision sections.

Weak entity signals

The site did not clearly connect the brand, service areas, cannabis marketing expertise, case studies, contact paths, and proof pages.

Compliance risk

Restricted-market pages need careful language. The goal was to improve visibility without medical claims, inflated promises, or policy-sensitive wording.

How to read the screenshot

The chart shows AI referral visibility becoming measurable.

This is not a vanity traffic claim. The screenshot is a GA4 user acquisition view, filtered by first user source / medium, where ChatGPT sources become visible after the AEO / GEO work.

RangeCustom period: May 17 - Jun 14, 2026.
Sourceschatgpt.com / ai-assistant, chatgpt.com / (not set), and chatgpt.com / referral are visible in the legend.
BaselineThe blue AI-assistant line has no visible activity through the first part of the period.
PeakThe same blue line rises to roughly 20 users per day around June 13, based on the chart scale.
PercentFrom a zero baseline, a normal percent increase is not mathematically meaningful. From the first visible non-zero day to the visible peak, the chart indicates approximately +400% growth.
GA4 line chart showing ChatGPT user acquisition growth from AI-assistant, not set, and referral sources
Provided GA4 screenshot. Exact totals should be confirmed with an exported GA4 table; this page uses the visible chart scale for approximate daily values.

What changed

The pages were rebuilt so humans, search engines, and AI assistants could understand the same facts faster.

Answer-first page architecture

Priority pages were rewritten around real buyer questions: what the service solves, who it is for, what evidence supports it, and what the next step should be.

AEO and FAQ blocks

Each page received concise direct-answer sections, FAQ content, and natural phrasing that AI systems can summarize without stripping away the business context.

GEO and entity clarity

Content was connected to the Canadian cannabis market, service categories, city context, compliance constraints, and proof from relevant case studies.

Schema and trust signals

Structured data, internal links, editorial notes, clear contact routes, and compliance language were added so the site could be understood as a trustworthy source.

The main win was not adding more words. It was making the existing expertise easier to verify, quote, and route into a buying conversation.

Why this matters for cannabis and CBD brands

AI visibility is becoming a second discovery layer beside search.

Cannabis and CBD companies often cannot rely on generic marketing playbooks. Paid ads can be restricted, health claims create risk, and local rules differ by market. AEO and GEO work helps the site explain itself clearly when a buyer asks an AI assistant for options, comparisons, checklists, or agencies with restricted-market experience.

For buyers

Pages become easier to scan, compare, and trust because they answer practical questions before asking for a call.

For search

Structured pages give Google clearer signals about topic coverage, service relevance, authoritativeness, and internal proof.

For AI assistants

Clean answers, schema, and source-backed sections make it easier for AI systems to cite or recommend the business in relevant contexts.

FAQ

Questions this case study answers

Can AEO and GEO guarantee AI citations?

No. AI assistants decide what to cite based on many signals. The goal is to make pages clearer, more verifiable, and easier to reference, not to promise guaranteed AI inclusion.

Why is the percent growth approximate?

The source provided here is a GA4 screenshot, not a GA4 exported table. The visible chart shows the AI-assistant source rising from no visible baseline to roughly 20 users per day, and about +400% from the first visible non-zero point to the peak.

What kind of pages were optimized?

Service, local, proof, FAQ, and educational pages were structured around direct answers, compliance-safe cannabis marketing language, schema markup, internal links, and clear next steps.

Is this only for cannabis companies in Canada?

No. The same framework can be adapted for CBD, dispensary, delivery, ecommerce, and other restricted-market brands in Canada, the USA, and other markets where compliance and trust matter.

Next step

Want to know whether your site is ready for AI-driven discovery?

We can review your cannabis or CBD website for answer quality, AI visibility, schema, internal proof, compliance language, and conversion path issues.