If you want to learn how to rank in AI search, meaning how to get your brand recommended by large language models like ChatGPT and Google’s AI mode, I have three people you should be following. But first, a word of warning about courses.
People often ask me how to learn GEO. They’re usually looking for a course or certification. The problem is this is a brand new field. There is no course I’d pay for right now. At TJ Digital, we’ve been running AI optimization experiments for our clients for years, and the landscape shifts too fast for any static curriculum to keep up. A Coursera GEO course updated in late 2025 covers the basics, but best practices can change week by week as Google and OpenAI tweak their models.
Everyone who’s actually good at this is learning in real time. We’re running experiments, documenting what works, and comparing notes with others doing the same.
Does SEO Experience Help with GEO?
Having a background in SEO gives you a huge advantage. Most of the foundational systems are the same. Content quality, site structure, E-E-A-T signals, and technical accessibility all still matter. AI systems tend to use the same trust and relevance signals that search engines do.
But there are real differences. Traditional SEO measures rankings and clicks. GEO measures whether AI tools cite your brand when generating answers. Traditional SEO focuses on your website. GEO requires presence across multiple platforms, because LLMs pull from YouTube, Reddit, forums, review sites, and more.
| Traditional SEO | GEO | |
| Goal | Rank in top search positions | Get cited in AI-generated answers |
| Key Metrics | Rankings, clicks, organic traffic | AI citation frequency, brand share of voice |
| Content Focus | Keywords, title tags, backlinks | Self-contained paragraphs, schema, Q&A format |
| Presence | Mainly your website | Multi-platform (YouTube, Reddit, forums, reviews) |
| How Users Find You | Click a blue link on a search results page | AI retrieves and synthesizes your content into an answer |
The overlap means SEO practitioners aren’t starting from zero. But there are new opportunities, and the only way to find them is through experimentation.
Why Courses Won’t Cut It (Yet)
The consensus among practitioners is that GEO moves too fast for static courses. A certification in GEO is not yet a meaningful industry credential. Most courses are one-time snapshots, and some already contain outdated details by the time you finish them.
The better approach: absorb the foundational concepts from whatever resources you can find, then validate everything with live AI queries. Ask ChatGPT and Gemini questions relevant to your business. See what gets cited and what doesn’t. Run your own tests. That hands-on feedback loop is worth more than any certificate.
SearchPilot’s GEO testing guide recommends using A/B testing methodology: create a control group of pages with no changes and a variant group with your GEO optimizations. Apply one change at a time, measure AI-specific outcomes, and iterate.
The 3 GEO Experts Worth Following
These are the only three people I feel truly understand GEO at a fundamental level and are openly sharing their insights.
1. Ethan Smith (Graphite)
Ethan is the CEO of Graphite, an AI-powered growth agency. He’s been in SEO for 18 years and now focuses entirely on AI search optimization.
I recommend checking out his episode on Lenny’s Podcast. It’s one of the first things I share with anyone interested in GEO. In that episode, he lays out a practical playbook for getting cited by ChatGPT, including optimized landing pages, YouTube content, and strategic answers on Reddit.
What makes Ethan valuable is that he’s not just sharing theories. He’s practicing it daily. Graphite’s research found that leads coming from ChatGPT convert at roughly 6 times the rate of traditional Google search traffic. That stat alone should get your attention.
Ethan’s approach is pragmatic: target high-intent queries with concise answers, and track your “AI share of voice” to measure results.
2. Kevin Indig (Growth Memo)
Kevin runs the Growth Memo newsletter, and I subscribe to the paid version. The free version is also excellent.
I’m currently producing an 11-part video series covering the 11 AI search ranking factors based on a single issue of his newsletter. That’s how dense and useful his work is.
Kevin has always been a valuable resource in the SEO community, and he’s one of the few SEO professionals who actually understands what has changed with GEO and what hasn’t. He runs real experiments on how AI systems read content, then translates that data into actionable frameworks. His 2026 “State of AI Search Optimization” memo breaks down the retrieval, citation, and trust factors that determine whether LLMs surface your brand.
If you want data-backed GEO guidance instead of speculation, Growth Memo is where you start.
3. Mike King (iPullRank)
Mike is the founder of iPullRank and the creator of what he calls “Relevance Engineering.” He goes deeper on a technical level into how Google’s systems work than any SEO I know.
Now he’s applying that same technical understanding to Google’s AI mode and AI overviews. Mike argues that search is becoming passage-level and probabilistic. Instead of fixed rankings, AI “fans out” queries into variations, retrieves documents via vector embeddings, and assembles answers by combining passages from multiple sources. Understanding this shift is critical for anyone serious about GEO.
Mike recently did a webinar with Joy Hawkins that you can find on YouTube. I recommend watching the entire thing. He keynoted “The Brave New World of SEO” at SEO Week 2025 and has written detailed analyses of how Google AI Mode handles query processing and reasoning.
How LLMs Actually Choose What to Cite
This is worth understanding because it explains why following these practitioners matters more than following old SEO playbooks.
LLMs don’t rank websites the way Google’s traditional algorithm does. They use a process called Retrieval-Augmented Generation (RAG). First, the model retrieves relevant information via web search or vector embeddings. Then it generates a coherent answer by stitching together the most relevant passages from its retrieval pool.
Research shows only about 12% of citations from ChatGPT and Gemini overlap with Google’s top-ranking links for the same query. Your content can get cited even if it’s nowhere near position one in Google, as long as it’s the most extractable, verifiable source for that specific question.
The factors that seem to matter most:
- Semantic relevance. Your content needs to closely match the user’s intent in meaning, not just keywords.
- Trust and authority. AI cross-checks claims against trusted sources. Brand mentions on authoritative sites, consistent information across platforms, and expert credentials all help.
- Clear structure. AI reads pages in chunks. Clear headings, short paragraphs, and self-contained answer blocks make it easier for models to find and extract useful pieces.
- Multi-platform presence. LLMs don’t just scan websites. They pull from YouTube, Reddit, forums, news articles, and reviews. Having your brand mentioned consistently across these channels builds what some call “co-citation signals.”
Start With Experiments, Not Theory
Here’s what I’d recommend if you’re getting into GEO right now:
- Follow these three practitioners. Subscribe to their newsletters and watch their content.
- Pick 5-10 questions your customers ask frequently. Query ChatGPT, Gemini, and Google AI mode with those questions and document what gets cited.
- Look at the content that is getting cited. Study its structure, its format, where it’s published.
- Make one change to one of your pages based on what you observed. Add a concise answer at the top, restructure a section into Q&A format, or add schema markup.
- Wait a few weeks, then query the same questions again and see if anything changed.
This is the “learning by doing” framework that everyone serious about GEO is using. There’s no shortcut. But the good news is that most of your competitors haven’t even started.
I think by the end of this year, ranking in LLMs will be the primary concern for any business that relies on search traffic. The businesses and agencies that start experimenting now will have a significant head start.
Get Ahead of AI Search
At TJ Digital, AI optimization is our primary focus. We help small and medium-sized businesses show up in ChatGPT, Google AI mode, and other AI platforms where customers are increasingly searching. If you want to see how your business currently performs in AI search, contact us for a free digital marketing audit. No long-term contracts, no sales pitch. Just a clear look at where you stand and what you can do about it.