01How to get your business recommended by AI: what it means
To get your business recommended by AI, you need your name on the pages AI tools read when they search the web for an answer. Two ideas make this work: GEO and AEO.
Here is how buying has changed. A person asks an AI who is the best at something. The AI searches the web, reads a stack of pages, then writes one answer and names a few businesses.
The person never sees the links. They only see the answer.
This changes the question you ask. It used to be, how do I rank my page? Now it is, what pages is the AI reading, and how do I get my name on them?
- GEO (generative engine optimization): getting onto the pages AI reads first.
- AEO (answer engine optimization): getting picked from those pages when the AI writes its answer.
Julian Goldie puts it simply: GEO gets you in the room, and AEO gets you chosen. Most business owners still write one blog post and wait. Meanwhile the AI reads lists, community threads and comparison pages.
02Real results from the best AI SEO community search
Julian Goldie shows his own numbers for the question, "who's the best AI SEO community." The AI Profit Boardroom is named first in several places. He is also open that it is not number one everywhere.
- Number one inside ChatGPT.
- Number one inside Gemini.
- Number one inside Google AI Mode.
- Number five inside Google AI Overviews in two spots.
When he searches the same phrase on Google, the site ranks number one, is recommended in AI Overviews, and shows up several times across different sites. It also ranks organically.
The traffic grew too. The same site went from zero to 5,800 impressions a day inside Google's AI features. On September 21 it was about 2,500 a day.
By October 4 it was 5,800.
One page ranked number one in AI search within 15 hours of going live. Another ranked when it was one day old. The site is only a domain rating 22, so you do not need a lot of backlinks.
You need quality ones.
03What AI tools do before they answer: three patterns
Julian looked at how AI tools answered the question and found three patterns. Each one tells you where to put your effort.
- Every AI searched the web first. None answered from memory. You do not need the AI to have known you for years. You need to be on the pages it finds today.
- List pages get cited most. The top result was a listicle. The next biggest group was community pages like Reddit threads. Normal blog articles barely showed up.
- His name came from more than one site. The AI Profit Boardroom site was cited. So was another of his sites. A third site linked straight back to the Boardroom.
The AI kept bumping into the same name from different pages. That repeat exposure is part of why the name showed up in the answer.
This is the core idea behind how to get your business recommended by AI. Be on the pages it reads, and be there more than once.
04Move 1: target query fan outs
When you ask an AI a question, it does not run just one search. It splits your question into smaller searches. These are called query fan outs.
For "what's the best AI SEO community," the AI might search:
- AI SEO community reviews
- AI SEO course with coaching
- Best SEO school group
Each small search pulls up different pages. You never see these searches, but they decide which pages the AI reads. Most people try to rank for the big question.
The AI builds its answer from the little ones.
Julian uses a tool called RadarKit for this. It tracks the fan out searches each AI tool ran. He connects it to Claude and asks what searches the AI ran for a question, and which ones his sites do not have a page for yet.
The result is a list of targets that few others are looking at.
05Move 2: turn fan outs into listicles
Next, you make the kind of content AI already cites. In RadarKit, open the prompt and go to responses. You will see a breakdown of content types based on citation data.
- If lists win, create listicles.
- If community wins, post on Reddit.
- If articles and blog posts win, make more of those.
List pages work because the comparing is already done. That makes them easy for the AI to read and quote. For each fan out, Claude writes a list page, such as the best SEO communities, the best SEO groups at school, or the best SEO courses for live coaching.
Claude does not start from scratch. It follows Julian's SEO skill, a checklist of about 100 checkpoints. It is built into the agent OS.
Add a "best for who" angle. AI answers say one option is best for beginners and another is best for coaching. Put something real from you in every list, like a result, a case study, a number or a tool you tested.
06Move 3: publish across many sites with the agent OS
One list page on one site is a decent start. But the AI kept finding Julian from different websites. So he publishes five or ten unique articles across different sites and platforms.
Here is the flow inside the agent OS:
- Open the generate section.
- Enter a keyword, usually one of the fan out pages.
- Paste in the case study you want to feature.
- Click generate. Claude writes five different list pages for five sites, with different titles and angles but the same honest answer.
- The pages deploy to your sites.
Most of his sites run on Netlify, where Claude builds and pushes pages live. A few run on WordPress, where it can publish directly. Pages then go to an indexing tool called Indexceptional so Google finds them fast.
This matters because if the AI has not found your page, it cannot rank or quote it. Some pages also link back to the main site, so the AI points people where you want.
07Tools and the agent OS setup
The agent OS is Julian's system for automating this process. It is built with Claude Code CLI. You can plug in Claude, Hermes and OpenClaw into one dashboard, then create and generate content.
- Search OS: pulls in your Google Search Console data and finds the GEO keywords closest to it.
- SEO skill: the checklist Claude follows when it writes pages.
- RadarKit: the AI search tracker that shows prompts, responses, citations and fan outs.
- Indexceptional: gets pages found by Google quickly.
These tools work together. Search data shows what you already rank for. The tracker shows what AI cites.
Claude writes the pages, and the system deploys them.
The agent OS comes inside the AI Profit Boardroom, along with a zip file, SEO automation systems and daily updates.
08Tips, limits and who this is for
Keep a few limits in mind. Julian is clear that he is not number one everywhere. Some Google AI Overviews results show his site at number five.
Results vary by tool.
- Do not rely on one blog post. AI reads lists, community threads and comparison pages.
- Do not skip the unique part. Add case studies, numbers or tools you tested.
- Do not publish and forget. Make sure pages get indexed.
- Sort options by who they are best for.
This approach is for business owners who know that buyers now ask AI who is best. If your name is not in the answer, they never know you were an option.
If you want help, the AI Profit Boardroom includes the agent OS, four weekly coaching calls, a community map and AI SEO, GEO and AEO training. Julian also offers a free one-to-one GEO and AEO audit at JulianGoldie.com.
