← Lenny's Podcast: Product | Career | Growth

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

Episode 2025-09-142025-09-14EnglishMembers only

Guests: Ethan Smith

The ultimate guide to AEO: How to get ChatGPT to recommend your product | Ethan Smith (Graphite)
Listen to this episode00:00 / 1:11:55
Reading settings
Reading preferences
Text size
Download in-depth read
Read offline

English · In-depth read · Current version

Downloads are not available yet.

Text language

In-depth read · Key ideas, reasoning, and examples, organized by topic.

§01AEO Is Search Summarized, Not Search Replaced

Ethan Smith frames Answer Engine Optimization as the work of appearing in large language model answers. He distinguishes it from a simple blue-link ranking: when an LLM answers a question such as which website builder is best, it summarizes several citations, so repeated inclusion across credible sources can matter more than holding the first citation position. The interview presents AEO as a meaningful but not total replacement for search. Smith says Webflow saw a sixfold conversion-rate difference between LLM traffic and Google Search traffic, while Lenny notes that ChatGPT already sends more traffic to his newsletter than Twitter. Those are reported examples rather than a universal benchmark, and the conversation does not establish their sample sizes or measurement method.

The opportunity is especially accessible to early-stage companies. Traditional SEO may require domain authority that takes time to build, whereas a new company can be cited quickly in a Reddit thread, YouTube video, blog post, or other source. Smith also argues that the conversational tail is larger than in conventional search: users ask more specific follow-up questions, including questions that have never previously been searched. He therefore sees a path for specialized products to win narrow, newly emerging queries before they have built a large SEO footprint. The limitation is that the claim is strategic and anecdotal in this section; it does not imply that every new company will gain traffic simply by publishing one citation.

Smith rejects the idea that AEO requires abandoning everything learned from SEO. He says traditional SEO still works, citing high-search-volume landing pages as one Webflow example, but adds two differences: the head depends on being mentioned across multiple citations, and the tail depends on answering detailed follow-up questions about use cases, features, integrations, and languages. In the Webflow work he describes, the strongest additional tactics were YouTube videos and Reddit optimization, alongside ordinary SEO. Lenny correctly tests the implication: appearing first in the answer's citations may not be as valuable as appearing repeatedly across the summary.

Anything can be optimized. You just need to understand the underlying systems and the rules of the game.

Keep reading as a member.

You've read the free preview. Become a member to read the rest, with side-by-side reading and downloads where available.