← All Guests
ES

Ethan Smith

Getting ChatGPT to recommend your product, covering AI product work, product design, and practical product lessons.

September 14, 2025·14,213 words
AI & Machine LearningGrowth & MetricsLeadership & ManagementProduct StrategyStartup BuildingDesign & UXEngineeringPricing & MonetizationSales & GTMCareer & Personal GrowthUser PsychologyData & Analytics

Episode

The ultimate guide to AEO: How to get ChatGPT to recommend your product | Ethan Smith (Graphite)

Summary

Ethan Smith, CEO of Graphite, delivers a comprehensive breakdown of Answer Engine Optimization (AEO) — how to get your product recommended by ChatGPT, Claude, and other LLMs. He draws on original research showing 100% AI-generated content doesn't rank in either Google or LLMs, while AI-assisted human content does. He covers citation mechanics, AEO differences across B2B and commerce, and why controlled experiments are essential.

Key Takeaways

1

Focus on citation volume across diverse sources (Reddit, YouTube, affiliate sites, help center pages) rather than ranking a single authoritative page — LLMs synthesize from many citations.

2

Run a proper controlled experiment before investing in AEO: intervene on 100 questions, leave 100 as control, and only scale reproducible tactics.

3

100% AI-generated content does not rank in Google or get cited in LLMs — AI-assisted content edited by humans is the right approach.

4

LLM traffic converts at dramatically higher rates than Google traffic (Webflow saw 6x). Track via post-conversion surveys and branded search lift.

5

Optimize your help center for AEO by moving it to a subdirectory, adding dense cross-links, and creating content for long-tail questions with zero competition.

Notable Quotes

But that's not the case in the LLM, because the LLM is summarizing many citations, and so you need to get mentioned as many times as possible.

AI & Machine Learning
00:00:15

And I personally prefer Answer Engine Optimization versus Generative Engine Optimization, because generative, you can generate images and videos and things other than an answer. Whereas answer is more narrowly defined, so my personal preference is we're talking about optimizing LLMs.

AI & Machine LearningGrowth & Metrics
00:06:52

Companies that we work with started in January and it started, one, because of more adoption, but two is because the answers became a bit more clickable.

General
00:07:57