Ethan Smith argues that aggregate organic-search traffic remains broadly stable despite widespread claims that SEO is collapsing. LLM usage is doubling year over year, but it remains much smaller than Google search and appears to be expanding the total discovery market rather than simply replacing search.

SEO traffic is roughly flat across the observed data
LLM usage is growing about 100% year over year
LLM traffic remains approximately one-fifteenth the size of search
Growth in AI discovery does not necessarily imply declining search traffic

“The pie is getting bigger. So SEO's not dying. LMs are growing a lot, but it's still a small fraction of search.”
Ethan Smith · 02:30

“But it seems like most of the strategies that work in answer engine optimization heavily overlap with the search engine optimization.”
Ethan Smith · 03:00

#seo#aeo#search-traffic#llms

Myth Buster10:30

Why SEO Decline Stories Can Misrepresent the Market

Individual websites and keywords can suffer severe traffic losses even when the overall channel remains stable. Ethan warns that negative cases are more likely to be reported, while aggregate data across thousands of sites offers a more reliable picture of market-wide behavior.

Site-level declines do not establish a market-wide trend
Negative traffic stories receive disproportionate attention
Algorithm changes and lower click-through rates can affect individual queries
Aggregate data suggests organic traffic is not broadly collapsing

“Anytime you're looking at a large data set, there's going to be examples of websites going down. There's also going to be examples of websites…”
Ethan Smith · 10:30

“And I have substantial data across thousands of sites where I can see that the uh organic traffic is not decreasing.”
Ethan Smith · 13:30

#seo-data#traffic-analysis#selection-bias#click-through-rate

Myth Buster26:30

Technical AEO Is Simpler Than the Hype Suggests

Most sites do not need an elaborate new technical stack for answer engines. Ethan says the core requirements remain strong internal links, readable and renderable pages, appropriate schema, and avoidance of technical failures that prevent bots from accessing content.

Maintain useful internal links
Ensure bots can render and read each page
Use appropriate schema and markup
Avoid client-side loading patterns that hide important content
Prepare for future action markup on forms and buttons

“Technical SEO, the stuff that is impactful is make sure you have good internal links.”
Ethan Smith · 27:00

“Internal links, schema, markup, and make sure the LLM can read your page. That's really all that there is.”
Ethan Smith · 27:30

#technical-seo#technical-aeo#schema#internal-links

Myth Buster28:00

Why LLMs.txt and Special Markdown Are Not Proven Ranking Tactics

Ethan cautions marketers against treating plausible technical ideas as established facts without evidence. A crawler requesting an LLMs.txt file does not prove the file affects retrieval or recommendations, and major AI companies are capable of parsing ordinary web pages without publishers converting everything to Markdown.

Repeated speculation can become mistaken for established practice
Crawler access does not prove ranking or retrieval impact
There is no cited evidence that LLMs.txt is widely used
Ordinary readable web pages do not need special Markdown versions
Rendering problems can still affect both search engines and LLMs

“That doesn't mean that it's doing anything.”
Ethan Smith · 29:00

“So LMS.txt, as far as I know, is not a thing.”
Ethan Smith · 29:00

#llms-txt#markdown#technical-aeo#seo-myths

Hot Take· 2

Hot Take04:00

Google's Ten Blue Links Are Still Holding Steady

Ethan challenges the assumption that conventional organic results are rapidly disappearing. His historical data suggests traffic from the ten blue links has remained within a relatively narrow range even as AI Mode, Gemini, and other AI interfaces have grown.

Traditional organic-result traffic has been broadly flat
AI interfaces are adding usage to a growing discovery market
A sudden collapse in blue-link traffic is not supported by the cited data
Future decline may be gradual rather than catastrophic

“The data that I have suggests that the amount of traffic coming from the 10 blue links is actually flat.”
Ethan Smith · 04:00

“It's not that it's shifting from search to LLMs, it's that there's an increased usage of both.”
Ethan Smith · 04:30

#google#organic-search#blue-links#ai-mode

Hot Take37:00

Autonomous Agents May Add a New Conversion Layer Above Search

Ethan expects search traffic to remain important while LLM-driven discovery grows as an additive channel. Over a longer horizon, autonomous agents may plan purchases and complete transactions for users, while personalization uses individual preferences to improve recommendations beyond today's generic search and chat results.

Search and LLM discovery may grow concurrently
Autonomous agents could research, coordinate, and transact for users
Agent-mediated commerce will create new organic and paid ecosystems
Personal context can improve hotel, travel, and product recommendations
Meaningful agent adoption may take longer than optimistic forecasts suggest

“It's additive. It's not taking from search and moving to LLMs, it's adding on top of it.”
Ethan Smith · 37:30

“I think that it'll probably take longer than people think.”
Ethan Smith · 38:30

#autonomous-agents#personalization#future-of-search#agent-commerce

Explainer· 2

Explainer07:30

How New Brands Can Win AEO Before They Can Win SEO

Answer engines create a faster route to visibility for young companies because recommendations often depend on third-party citations rather than the brand's own domain authority. New entrants can target both prominent commercial questions and extremely specific long-tail prompts where few useful answers exist.

Third-party mentions can influence answer-engine recommendations quickly
Off-site visibility can shorten time to impact
Highly specific prompts create openings with little existing competition
AEO is more accessible to young brands than authority-dependent SEO

“And you could get mentioned tomorrow and win a question like that.”
Ethan Smith · 07:30

“So if you're the only answer to a question, then you can win that.”
Ethan Smith · 17:00

#aeo#startups#citations#long-tail

Explainer31:30

AEO Citations Target Exact URLs, Not Just Authoritative Domains

Traditional SEO rewards backlinks from authoritative pages across the web, while answer-engine visibility depends more directly on the pages cited for a target prompt. A useful mention may not require a hyperlink, but it needs to appear on the exact URL the answer engine consults.

AEO mentions can work without hyperlinks
The cited URL matters more than domain authority alone
A mention on an uncited page may provide no prompt-level benefit
Lists can make brand mentions easier for models to identify
Off-site AEO outreach resembles backlink outreach but targets different pages

“In LMs, it just needs to be a mention.”
Ethan Smith · 31:30

“So it the only thing that matters are the specific URLs that are appearing in the citations.”
Ethan Smith · 32:30

#citations#backlinks#off-page-seo#off-page-aeo

Tool· 2

Tool21:00

Where to Find the Questions Customers Actually Ask AI

Prompt-panel data can reveal common head queries, but it is less useful for discovering sparse, highly specific questions. Ethan recommends mining Reddit, support conversations, sales-call transcripts, direct audience feedback, and chatbot interactions, then using an LLM to cluster and summarize recurring questions.

Use Reddit to find natural product and problem questions
Analyze customer-support and sales-call transcripts
Collect direct audience feedback after publishing content
Review questions submitted to brand or expert chatbots
Use an LLM to summarize recurring themes and question patterns

“Reddit is a great place to look for that because people are actually asking questions on Reddit about does HubSpot do this or I have…”
Ethan Smith · 22:00

“Looking at your customer support and sales conversations also, because you're getting a ton of people asking, does HubSpot do this, or I'm I'm struggling…”
Ethan Smith · 22:30

#customer-research#prompt-research#reddit#sales-calls#support-data

Tool35:00

Turn Paid Search Keywords Into an AEO Citation Target List

Ethan outlines a practical way to identify commercially valuable prompts and the pages influencing their answers. Begin with paid-search money terms, convert those keywords into natural questions, track the resulting answers using scraped interface data, and prioritize outreach to the most frequently cited URLs and domains.

Start with paid-search keywords that signal commercial value
Use competitor paid-search terms when first-party data is unavailable
Transform keywords into natural-language questions
Use scraped answer data rather than relying on API output
Identify the top cited URLs and domains for each category
Tailor outreach to the sources that actually appear

“What you want to do is you want to start with the prompts that you care about.”
Ethan Smith · 35:00

“Then you get your top citations and your top domains, and then you can have the answer to your question.”
Ethan Smith · 36:00

#prompt-tracking#paid-search#citation-research#aeo-tools

Takeaway· 3

Takeaway07:00

Why Early-Stage Startups Should Delay Heavy SEO Investment

SEO can remain valuable while still being a poor initial distribution channel for a new startup. Ethan recommends that seed-stage companies prioritize channels with faster feedback, then increase their search investment as authority and organizational resources grow.

Seed-stage companies should spend little or no time on SEO
Paid, viral, community, and Reddit channels can produce faster early feedback
Series A companies can begin increasing search investment
Series B companies are better positioned to invest heavily in SEO

“if you're an early stage founder, you should not be spending significant time on search engine optimization because it's hard to do.”
Ethan Smith · 07:00

“And the later you are, the easier it is to do because you have more authority.”
Ethan Smith · 07:00

#startups#seo-strategy#distribution#company-stage

Takeaway40:00

Unique Content and Video Will Become AI's Competitive Currency

As model algorithms converge in quality, proprietary and distinctive source material may become the stronger differentiator. Ethan expects richer video understanding to unlock concepts and context beyond titles and transcripts, potentially making video one of the most valuable inputs for future answer engines.

Better input data can outweigh small differences in model quality
Exclusive expert knowledge can improve generated answers
Video contains context that titles and descriptions do not capture
Multimodal systems can extract concepts beyond transcript text
YouTube and other video sources may gain greater citation importance

“The best input's going to out produce the best output.”
Ethan Smith · 40:00

“It might be the most important input into the LM.”
Ethan Smith · 41:30

#unique-content#video#training-data#multimodal-ai

Takeaway42:30

Replace AEO Guesswork With Reproducible Experiments

Because many AEO claims remain unproven, Ethan closes by urging marketers to test interventions rather than accept confident speculation. A useful result should show a measurable effect and then be reproduced before it is treated as reliable guidance.

Treat many current AEO claims as hypotheses
Design experiments around measurable visibility changes
Reproduce observed effects before drawing conclusions
Share experiment data to accelerate collective learning

“The way to know what works is to set up an experiment and to see an effect and to reproduce that.”
Ethan Smith · 42:30

“Everyone should be doing experiments and sharing their experiment data so that we can all learn and figure out the answers faster.”
Ethan Smith · 42:30

#experimentation#aeo#measurement#evidence

Frameworks in this episode

Citation-URL Outreach Playbook

Criteria-Combination Answer Architecture

Long-Tail Question Mining Loop

Minimal Technical AEO Checklist

Reproducible AEO Experiment Loop

Stage-Based Search Investment Ladder

Usage-and-Fit Engine Prioritization Rule

Insights & moments

⚡Myth Buster· 4

The Data Says SEO Traffic Is Not Dying