https://page2pod.com - Publishers with OpenAI deals get cited 48% more by ChatGPT. Thomas Peham of Otterly.AI on what AI search citation testing actually shows.

Most AI search advice is theory. Thomas Peham, CEO and co-founder of Otterly.AI, runs a team with two full-time GEO researchers whose only job is to test it. Across millions of citations from ChatGPT, Claude, Gemini, and Google AI Overviews, a consistent picture emerges: corporate and brand websites supply roughly 40–60% of all citations, niche and trade media outperformed mainstream publishers in 16 of 17 US industries studied, and publishers holding licensing deals with OpenAI pick up about 48% more ChatGPT citations than those without.

The uncomfortable findings are here too. Schema markup lifted Google organic and AI Overviews visibility but did nothing measurable on ChatGPT. Crawl data shows almost no activity on llms.txt files, while sitemaps get hit constantly. A fictional agency with no clients and no history got recommended by ChatGPT purely by earning mentions in the articles ChatGPT was already citing. And 25% of the URLs ChatGPT cites are dead links.

Thomas also shares Otterly.AI's own Claude numbers, which should make anyone in B2B re-examine their analytics: Claude accounts for 0.1–0.2% of their website traffic, but 10.6% of net-new leads name it as where they first heard about the company.

🧪 In This Episode

- Why hiring dedicated GEO researchers became the marketing strategy

- Does schema markup actually influence ChatGPT citations?

- The llms.txt crawl data — and why sitemaps deserve the attention instead

- Where AI citations come from: corporate sites, niche media, and publisher licensing deals

- The fake London SEO agency experiment and what it exposes about off-page manipulation

- Building brand sentiment through reviews, product content, and a central brand entity

- Why YouTube long-form gets cited regardless of view count

- What Claude's citation data reveals about B2B buying you can't see in analytics

A practitioner's look at what's been tested, what held up, and what marketers should stop doing.

Subscribe for new episodes with the people actually running the experiments in SEO, AEO, and GEO.

Which finding would change your roadmap first — the schema results, the llms.txt crawl data, or the 48% licensing gap? Tell us in the comments.

####

Sponsored by Moving Traffic Media → https://www.movingtrafficmedia.com

Follow the Page 2 Podcast

✅ Page 2 Podcast on Youtube → https://www.youtube.com/@thepage2podcast922

✅ Page 2 Podcast on Linkedin → https://www.linkedin.com/company/page2podcast

✅ Page 2 Podcast on Facebook → https://www.facebook.com/ThePage2Podcast/

✅ Page 2 Podcast on Instagram → https://www.instagram.com/page2pod

✅ Page 2 Podcast on Tiktok → https://www.tiktok.com/@jonleeclark

Follow Jon Clark

✅ Jon Clark on Twitter → https://x.com/jonlclark

✅ Jon Clark on Linkedin → https://www.linkedin.com/in/jonlclark

🛠️ Tools & Resources Mentioned

- Thomas Peham → https://thomas-peham.com/

- Thomas Peham on LinkedIn → https://www.linkedin.com/in/thomaspeham/

- Thomas Peham on Twitter → https://twitter.com/tompeham

- Thomas Peham on Instagram → https://www.instagram.com/tompeham/- Otterly.AI → https://otterly.ai - Otterly.AI's Agent Analytics → https://otterly.ai/blog/agent-analytics/ - Otterly.AI's Claude Citation Study: https://www.linkedin.com/posts/thomaspeham_our-claude-citation-study-shows-reddit-activity-7477977546194849792-gu8-/

https://page2pod.com - Entity SEO finally has a number: Lidia Infante on tracking Google's confidence in your brand with the Knowledge Graph API.

Lidia Infante leads SEO and AI search at Pleo, following senior roles at SurveyMonkey, BigCommerce, and Sanity. She walks through the entity work most teams talk about but can't measure — querying the Google Knowledge Graph API for a confidence score, tracking that score in a Google Sheet, and using the movement to prove that third-party profile cleanup and consistent semantic triples actually changed something. That's the case for Answer Engine Optimization (AEO) and Generative Engine Optimization (GEO) work you can defend in a budget meeting.

She also gets practical about the rest of the job: using LLMs to break a research library into individual tagged facts and place them inside existing content, correlating clicks on candidate brand terms against real signups to decide what your product should even be called, and why disagreeing with consensus may be the last cheap differentiator in a landscape full of AI-generated averages. Then the part nobody puts in a conference deck — legal approval layers, 16-market localization, and the business questions a growth leader has to answer before proposing any strategy at all.

🧩 In This Episode

- How do you track Google's confidence in your brand entity?

- The CORE framework: consistency, optimization, repetition, engagement

- Breaking research reports into tagged facts with Python and the Gemini API

- What should your product actually be called? Correlating brand terms to signups

- Why going against consensus still ranks when everything else is an average

- Getting content through legal without six-month lead times

- The questions a new growth leader should answer before proposing a strategy

A rare conversation that connects entity theory to the org chart, the legal queue, and the P&L that decide whether any of it ships.

👉 Subscribe for new practitioner interviews on SEO, AEO, and GEO every week.

💬 Have you run your brand through the Knowledge Graph API yet? Tell us what came back — and which entity Google thinks you are.

Find Lidia

✅ Lidia Infante on LinkedIn → https://www.linkedin.com/in/lidiainfante/

✅ Lidia Infante on X → https://x.com/LidiaInfanteM

✅ Lidia Infante on Bluesky → https://bsky.app/profile/lidia-infante.com

✅ Lidia's website → https://www.lidia-infante.com/

✅ Lidia Infante on Youtube → https://www.youtube.com/@LidiaInfanteM/

✅ Lidia Infante on Instagram → https://www.instagram.com/lidiainfantem/

####

Sponsored by Moving Traffic Media → https://www.movingtrafficmedia.com

Follow the Page 2 Podcast

✅ Page 2 Podcast on Youtube → https://www.youtube.com/@thepage2podcast922

✅ Page 2 Podcast on Linkedin → https://www.linkedin.com/company/page2podcast

✅ Page 2 Podcast on Facebook → https://www.facebook.com/ThePage2Podcast/

✅ Page 2 Podcast on Instagram → https://www.instagram.com/page2pod

✅ Page 2 Podcast on Tiktok → https://www.tiktok.com/@jonleeclark

Follow Jon Clark

✅ Jon Clark on Twitter → https://x.com/jonlclark

✅ Jon Clark on Linkedin → https://www.linkedin.com/in/jonlclark

🛠️ Tools & Resources Mentioned

- Lidia's Synthetic Research Pannel → https://www.linkedin.com/posts/lidiainfante_i-am-a-girl-with-a-claude-subscription-and-activity-7490867156474171392-xnRo/

- SEO and Brand Sitting in a Tree K-I-S-S-I-N-G → https://speakerdeck.com/lidiainfante/seo-and-brand-sitting-in-a-tree-k-i-s-s-i-n-g

https://page2pod.com - Fan-out queries expose the exact words LLMs want to see about your brand — Peec AI's Malte Landwehr on how to read them.

Malte Landwehr is CPO and CMO at Peec AI, and he spent two decades in search before this — VP Product at Searchmetrics, then VP SEO at idealo, running enterprise SEO across roughly 130 million pages. His core argument: LLMs reward consensus. If your website, your founder's LinkedIn, your press release footer, and your G2 profile describe you three different ways, the model has nothing consistent to trust. Describing yourself identically everywhere you have control is the highest-leverage AI visibility work most brands still haven't done.

He also gets specific about what doesn't hold up. Answer Engine Optimization (AEO) and Generative Engine Optimization (GEO) have picked up tactics that don't survive testing — schema markup can't carry facts your body copy leaves out, duplicate .md versions of pages went uncited across hundreds of thousands of chats, and llms.txt is guidance for AI agents rather than a visibility lever.

What works is duller: declarative sentences, concrete numbers, and content that answers what the fan-outs are already asking about you.

🔎 In This Episode

- Why LLMs look for consensus — and what happens when your site, LinkedIn, and G2 profile don't match

- How should you do keyword research when fan-out queries add terms you never wrote?

- What ChatGPT's site: fan-outs reveal about whether OpenAI has its own link graph

- Does schema markup still influence AI citations? What the testing actually shows

- Markdown copies, .md URLs, and llms.txt — where each one belongs

- Product feeds, Yelp/Resy/OpenTable, and how local visibility works inside ChatGPT

- Why the biggest lever is usually the wrong first project

A practitioner's map of what to change, what to measure, and what to ignore in AI search right now.

Subscribe for new episodes with the practitioners actually doing the work → https://www.youtube.com/@thepage2podcast922?sub_confirmation=1

Comment: what's the strangest term you've seen appended to a fan-out query for your brand?

Connect with Malte Landwehr

✅ Malte on LinkedIn → https://www.linkedin.com/in/landwehr

✅ Malte on X → https://x.com/maltelandwehr

✅ Malte's Personal website → https://www.maltelandwehr.de

✅ Peec AI → https://peec.ai

####

Sponsored by Moving Traffic Media → https://www.movingtrafficmedia.com

Follow the Page 2 Podcast

✅ Page 2 Podcast on Youtube → https://www.youtube.com/@thepage2podcast922

✅ Page 2 Podcast on Linkedin → https://www.linkedin.com/company/page2podcast

✅ Page 2 Podcast on Facebook → https://www.facebook.com/ThePage2Podcast/

✅ Page 2 Podcast on Instagram → https://www.instagram.com/page2pod

✅ Page 2 Podcast on Tiktok → https://www.tiktok.com/@jonleeclark

Follow Jon Clark

✅ Jon Clark on Twitter → https://x.com/jonlclark

✅ Jon Clark on Linkedin → https://www.linkedin.com/in/jonlclark

https://page2pod.com - Clicks are disappearing. Derek Hobson makes the case that direct traffic is the AI search signal most teams aren't measuring.

Derek Hobson — SEO strategist, writer behind SEO Humorist, and former Senior Director of SEO and AI Search at Brainlabs — argues that when organic clicks shrink, the proof of impact moves elsewhere: direct sessions, log files, and query fan-out data. On several of his accounts, direct and organic tracked in parallel for years until direct broke upward right as clicks fell and impressions climbed. His read: people research a brand inside an LLM, then go straight to the site. He can't prove it yet, and he says so out loud to clients — but directionally, it's the number he'd watch.

The second argument is organizational. Off-page SEO can't sit in an optional package anymore now that answer engines scrape, aggregate, and synthesize third-party sources. That means SEO, paid, PR, and social have to actually talk to each other — and most companies are still built around separate teams, separate budgets, and separate reporting.

Jon Clark and Joe DeVita also get into automating reporting with Apps Script, why LLM referral numbers in analytics are almost certainly wrong, and why Derek thinks every SEO should investigate a win as aggressively as a loss.

📊 In This Episode

- Is direct traffic the AI search signal most teams are ignoring?

- Who actually owns the direct channel — and why nobody claims it

- Why off-page SEO stopped being an optional line item

- Query fan-out and grounding queries: what the Gemini API and Microsoft Clarity now expose

- Why LLM referral traffic looks impossibly small in analytics

- Building a paid-and-organic gap analysis that starts real cross-team conversations

- Automating monthly data pulls with Apps Script and pattern-matching keywords with embeddings

- Investigating positive results the same way you'd investigate a drop

A practical look at what "proving the work" looks like when the click is no longer the unit of measurement.

Subscribe for new Page 2 Podcast episodes with the practitioners actually doing this work.

Comment: who owns direct traffic at your company — and have you seen the same uptick Derek describes?

Sponsored by Moving Traffic Media → https://www.movingtrafficmedia.com

Follow the Page 2 Podcast

✅ Page 2 Podcast on Youtube → https://www.youtube.com/@thepage2podcast922

✅ Page 2 Podcast on Linkedin → https://www.linkedin.com/company/page2podcast

✅ Page 2 Podcast on Facebook → https://www.facebook.com/ThePage2Podcast/

✅ Page 2 Podcast on Instagram → https://www.instagram.com/page2pod

✅ Page 2 Podcast on Tiktok → https://www.tiktok.com/@jonleeclark

Follow Jon Clark

✅ Jon Clark on Twitter → https://x.com/jonlclark

✅ Jon Clark on Linkedin → https://www.linkedin.com/in/jonlclark

Find Derek Hobson

✅ SEO Humorist → https://seohumorist.com

✅ Derek Hobson on LinkedIn → https://www.linkedin.com/in/dereknhobson/

✅ Derek Hobson on YouTube → https://www.youtube.com/@seohumorist

✅ Derek Hobson on Reddit → https://www.reddit.com/user/SEO_Humorist/

✅ Derek Hobson on Instagram → https://instagram.com/seohumorist✅ Derek Hobson on Bluesky → https://bsky.app/profile/seohumorist.bsky.social✅ Derek Hobson on X → https://x.com/Seo_humorist✅ Derek Hobson on TikTok → https://tiktok.com/@seohumorist

https://page2pod.com - Sophie Brannon bans new hires from touching AI for three months. Here's how an SEO agency builds AI workflows without losing the thinking underneath them.

Sophie Brannon is co-founder and director of StudioHawk US, the Atlanta arm of the global SEO agency, which she launched from scratch and grew to $1.4M in its first nine months. Her position: AI is genuinely compressing agency delivery timelines — content briefs, IA mapping, and audits that took weeks now take days — but a junior who reaches for ChatGPT before they understand the work never develops the strategic judgment that becomes their actual job. So new StudioHawk hires do everything manually for their first three months, then earn access back.

She also lays out the commercial side nobody's solved yet: token budgets replacing headcount budgets, the possibility of agencies pricing retainers on estimated token usage instead of hours, and why AI visibility means very little until you can tie it to traffic and revenue — across roughly 100 brands StudioHawk surveyed, AI search accounted for about 0.07% of total revenue.

Jon Clark and Joe DeVita dig into the agency mechanics too: why StudioHawk runs direct-to-specialist with no account managers, why they refuse long-term contracts, how link costs get absorbed while digital PR runs on a points-based money-back guarantee, and what Sophie still won't hand to a model.

🤖 In This Episode

- Why new hires are banned from using AI for their first three months

- What do clients actually ask about AI search — and how to answer without jargon

- The AI search tracking stack: GA4 AI organic, Peak, SEMrush AI, Ahrefs Rank Tracker

- Which AI search metrics matter: visibility, citations, mentions, and the revenue line underneath them

- Digital PR's points system — a link is 1 point, an unlinked mention is 0.5, with a pro-rata money-back guarantee

- Will agencies charge clients based on token usage instead of hours?

- Automated content briefs and IA structure mapping built with n8n and Claude

- The three things Sophie won't automate: strategy, creative, and client comms

A rare look at the operating decisions behind an SEO agency scaling in the AI era, from someone 18 months into building one.

👉 Subscribe for new episodes with the practitioners actually doing this work: https://www.youtube.com/@thepage2podcast922

💬 Would you ban your junior hires from using AI for their first three months — or is that training them for a world that no longer exists? Tell us in the comments.

Find Sophie Brannon

✅ Sophie Brannon on LinkedIn → https://www.linkedin.com/in/sophie-brannon-081bb294/

✅ Sophie Brannon on Instagram → https://www.instagram.com/sophiebrannon_seo/

✅ StudioHawk US → https://studiohawk.com

✅ Sophie Brannon's Personal Website → http://sophiebrannon.com

####

Sponsored by Moving Traffic Media → https://www.movingtrafficmedia.com

Follow the Page 2 Podcast

✅ Page 2 Podcast on Youtube → https://www.youtube.com/@thepage2podcast922

✅ Page 2 Podcast on Linkedin → https://www.linkedin.com/company/page2podcast

✅ Page 2 Podcast on Facebook → https://www.facebook.com/ThePage2Podcast/

✅ Page 2 Podcast on Instagram → https://www.instagram.com/page2pod

✅ Page 2 Podcast on Tiktok → https://www.tiktok.com/@jonleeclark

Follow Jon Clark

✅ Jon Clark on Twitter → https://x.com/jonlclark

✅ Jon Clark on Linkedin → https://www.linkedin.com/in/jonlclark

https://page2pod.com - Content strategy that moves rankings: Angela Skane took her blog from page 3 to page 1 by optimizing what she already had — not publishing more.

Angela Skane is SEO & Content Strategy Manager at Newfold Digital, where she leads global content strategy for Network Solutions and a portfolio of brands serving small business owners. Fresh off her first SEO Week stage talk, she brings a framework most search teams never look at: the way TikTok Shop creators hook, hold, and convert an audience in under 60 seconds.

Her argument is that attention is earned moment by moment, in an article the same way it is in a video. Every page needs a job. Every section has to keep the reader moving. And a page that technically ranks can still fail if it doesn't speak to the person actually making the decision. Along the way: why traditional demographic personas are too thin, why environmental data about your audience changes how you frame every problem, and why her team allows zero AI-generated content on the blog while using AI heavily for research and synthesis.

🎧 In This Episode

- The four-phase creator arc — hook, name your audience, appeal to emotion, close the action

- Layering psychological triggers by journey stage instead of picking one and running it all year

- What does a "job" for a page actually look like, and how do you tell when a page isn't doing it?

- Page 3 to page 1 in 18 months by optimizing existing content over publishing volume

- Environmental-data personas: small business failure rates, cost pressures, and what else is on the buyer's mind

- Do 3,000-word guides still work? Her top-converting article says yes

- Where AI belongs in the workflow — research and data synthesis, never the paragraph

If your content is ranking but not converting, this one is about the gap between the two.

Subscribe for new practitioner interviews on SEO and Answer Engine Optimization (AEO) every week.

Angela says to name your audience explicitly in the copy. Have you tried it — and did engagement actually move? Tell us in the comments.

⏱️ Chapters

0:00 Intro: Attention Is Earned2:08 Meet Angela Skane3:36 What Is Newfold Digital?5:46 From Data Journalism to SEO7:18 Why TikTok Shop Caught Her Eye9:14 The Four-Phase Creator Framework11:40 Applying TikTok Signals to Articles15:07 Which Emotion Fits Which Stage?19:55 Every Page Needs a Job23:03 Are Traditional Personas Still Useful?28:16 How Do You Diagnose a Weak Page?31:09 Do 3,000-Word Guides Still Work?36:39 Should AI Write Your Blog Content?39:28 SEO Pods, Briefs, and Research40:59 How Do You Report Content Value?44:19 How the Teams Actually Collaborate47:11 Rapid-Fire Round

Connect with Angela

✅ Angela Skane on LinkedIn → https://www.linkedin.com/in/angelaskane/

✅ Network Solutions → https://www.networksolutions.com

Sponsored by Moving Traffic Media → https://www.movingtrafficmedia.com

Follow the Page 2 Podcast

✅ Page 2 Podcast on Youtube → https://www.youtube.com/@thepage2podcast922

✅ Page 2 Podcast on Linkedin → https://www.linkedin.com/company/page2podcast

✅ Page 2 Podcast on Facebook → https://www.facebook.com/ThePage2Podcast/

✅ Page 2 Podcast on Instagram → https://www.instagram.com/page2pod

✅ Page 2 Podcast on Tiktok → https://www.tiktok.com/@jonleeclark

Follow Jon Clark

✅ Jon Clark on Twitter → https://x.com/jonlclark

✅ Jon Clark on Linkedin → https://www.linkedin.com/in/jonlclark

🛠️ Tools & Resources

- TikTok Shop → https://shop.tiktok.com - 2739 — the TikTok Shop coaching community Angela learned the framework → https://www.tiktok.com/discover/2739-membership?lang=en - Dustin Davis — creator who breaks down TikTok content through a psychology lens - Angela's article on the eight human desires in psychology → https://searchengineland.com/persuasive-content-human-psychology-478824 - "What to do after you buy a domain name" — her top-converting article (3,000+ words) → https://www.networksolutions.com/blog/bought-domain-now-what/

https://page2pod.com - Agentic commerce, llms.txt, and markdown for AI: Yoast principal SEO Alex Moss explains how brands get selected by AI agents — not just clicked.

Alex Moss, principal SEO at Yoast and co-founder of FireCask, joins Jon Clark and Joe DeVita to unpack what it takes to be chosen by an AI agent: concise, quotable content, clear entity relationships, structured data, and a single machine-readable source of truth on your own domain. He calls the discipline "token optimization" — making information load faster for models the way Core Web Vitals made pages load faster for people.

Alex also breaks down the standards war behind Answer Engine Optimization (AEO) and Generative Engine Optimization (GEO) — llms.txt, agents.md, Google's UCP and OKF, OpenAI's ACP — and why the tech for agentic commerce is ready even if consumers aren't. His advice is refreshingly practical: experiment, follow the engineers building these systems, and adopt one standard at a time.

🤖 In This Episode

• How Yoast decides which SEO defaults ship on — and why schema stays gated behind an API
• Why markdown beats HTML for LLM ingestion (an 80% token reduction on one real example)
• What does it actually mean to be "selected" by an AI agent instead of clicked?
• Is llms.txt worth adopting now? Alex's insurance-policy framework for unproven standards
• The Walmart–ChatGPT experiment and Alex's timeline for real agentic commerce adoption
• AI information pages, entity maps, and building a source of truth machines can trust
• The team you'd hire to launch an agentic-commerce-ready eCommerce brand today

Whether you're placing bets on llms.txt or ignoring the acronyms entirely, this conversation gives you a working framework for which AI-search standards deserve your time.

Subscribe for new practitioner-level conversations on SEO, AEO, and GEO every week.

💬 Would you buy anything through a ChatGPT session today — and what's the biggest purchase you'd trust an AI agent with? Tell us in the comments.

Guest links:

• Alex Moss → https://alex-moss.co.uk
• Yoast → https://yoast.com
• FireCask → https://firecask.com

Sponsored by Moving Traffic Media → https://www.movingtrafficmedia.com

Follow the Page 2 Podcast
✅ Page 2 Podcast on Youtube → https://www.youtube.com/@thepage2podcast922
✅ Page 2 Podcast on Linkedin → https://www.linkedin.com/company/page2podcast
✅ Page 2 Podcast on Facebook → https://www.facebook.com/ThePage2Podcast/
✅ Page 2 Podcast on Instagram → https://www.instagram.com/page2pod
✅ Page 2 Podcast on Tiktok → https://www.tiktok.com/@jonleeclark

Follow Jon Clark
✅ Jon Clark on Twitter → https://x.com/jonlclark
✅ Jon Clark on Linkedin → https://www.linkedin.com/in/jonlclark

#Page2Podcast #AgenticCommerce #AISearch #SEO #AEO

🛠️ Tools & Resources Mentioned

• Yoast SEO → https://yoast.com
• FireCask → https://firecask.com
• llms.txt proposal → https://llmstxt.org
• agents.md
• NLWeb (Microsoft, RV Guha) → https://github.com/microsoft/NLWeb
• Yoast aggregation feature (entity/relationship endpoint)
• InLinks Entity Map (Dixon Jones) → https://inlinks.com
• WordLift (Andrea Volpini) → https://wordlift.io
• Schema.org → https://schema.org
• Shopify → https://www.shopify.com
• WooCommerce → https://woocommerce.com
• Cloudflare markdown /crawl endpoint & content negotiation → https://www.cloudflare.com
• Google UCP, OKF, Knowledge Catalogue & ARD (proposed standards)
• OpenAI ACP (Agentic Commerce Protocol)
• Astro → https://astro.build
• Google Lighthouse
• brightonSEO → https://brightonseo.com
• Reykjavik Internet Marketing Conference (Sept 17)

https://page2pod.com - JavaScript isn't breaking your SEO — rendering is. Merj R&D Director Giacomo Zecchini on what Google and AI crawlers actually see.

Giacomo Zecchini has spent years injecting tracking beacons into Google's rendering service to replace industry folklore with data. His research with Vercel found that virtually every indexable page crawled by Googlebot gets rendered — so the "rendering budget" most SEOs plan around doesn't behave the way the industry assumes. The real failures live in implementation: APIs timing out mid-render, lazy-loaded content that never fires under Google's viewport expansion, and hero images that shove main content thousands of pixels down the page.

Do AI crawlers render JavaScript? Some can — but none do it consistently. Giacomo's testing shows bots like ClaudeBot, Mistral, and DuckAssistBot sometimes render and sometimes fetch raw HTML using the same user agent, which is why Answer Engine Optimization (AEO) starts with content that survives a no-render fetch. He also makes the case for serving Markdown to bots via the Accept HTTP header (a web standard) instead of .md URLs, and shows how one swapped ARIA label made ChatGPT's agent click Cancel instead of Save — turning accessibility into an agentic-web requirement, not a nice-to-have.

⚙️ In This Episode

• Why "JavaScript problems" are really rendering and implementation problems
• The rendering budget myth, tested against live Googlebot data
• How does Google's viewport expansion handle lazy loading and infinite scroll?
• Auditing rendering backwards: from rendered page to rendering tree
• Which AI crawlers actually render JavaScript — and why it changes per fetch
• Serving Markdown via the Accept header vs. .md URLs vs. llms.txt
• Why bad ARIA labels make AI agents take the wrong action on your site

A masterclass in the infrastructure layer of search visibility — what machines can actually access, interpret, and act on.

Subscribe for practitioner-level SEO, AEO, and GEO conversations every week.

💬 Giacomo serves Markdown to any bot that asks for it — Google's advice notwithstanding. Would you serve Markdown to AI crawlers on a client site, or is HTML-only still the right call? Tell us in the comments.

Connect with Giacomo Zecchini✅ Follow Giacomo Zecchini on LinkedIn: https://www.linkedin.com/in/gzecchini/✅ Follow Giacomo Zecchini on X / Twitter: https://x.com/giacomozecchini✅ Follow Giacomo Zecchini on Bluesky: https://bsky.app/profile/giacomozecchini.bsky.social✅ Follow Giacomo Zecchini on Mastodon: https://mastodon.social/@giacomozecchini✅ Follow Giacomo Zecchini on SpeakerDeck: https://speakerdeck.com/giacomozecchini✅ Follow Giacomo Zecchini on SlideShare: https://www.slideshare.net/giacomozecchini✅ Giacomo Zecchini's Personal Website: https://giacomozecchini.com

Sponsored by Moving Traffic Media → https://www.movingtrafficmedia.com

Follow the Page 2 Podcast✅ Page 2 Podcast on Youtube → https://www.youtube.com/@thepage2podcast922✅ Page 2 Podcast on Linkedin → https://www.linkedin.com/company/page2podcast✅ Page 2 Podcast on Facebook → https://www.facebook.com/ThePage2Podcast/✅ Page 2 Podcast on Instagram → https://www.instagram.com/page2pod✅ Page 2 Podcast on Tiktok → https://www.tiktok.com/@jonleeclarkFollow Jon Clark✅ Jon Clark on Twitter → https://x.com/jonlclark✅ Jon Clark on Linkedin → https://www.linkedin.com/in/jonlclark

🛠️ Tools & Resources Mentioned• Merj Official Site: https://merj.com• Google JS Rendering Capabilities Research with Vercel: https://vercel.com/blog/how-google-handles-javascript-throughout-the-indexing-process• AI Cralwer's Accept Language Research: https://www.linkedin.com/posts/merj-com_this-is-something-most-multilingual-sites-activity-7434943896385822720-T2u1• Merj's take on serving Markdown Files: https://www.linkedin.com/posts/merj-com_ai-agents-are-reading-your-site-right-now-activity-7466111771209142272-3xEy

https://page2pod.com - In this episode Marcos Ciarrocchi, co-founder of Graphite, breaks down why Graphite believes that 95% of SEO work is often waste and why the real leverage comes from identifying the small percentage of initiatives that actually drive growth. The conversation dives deep into how SEO strategy is shifting from keyword-first thinking to topic-first systems built around search intent, topical authority, AI prompts, and answer engine visibility.The discussion also explores how AEO and SEO may not be separate disciplines, but two versions of the same strategic question: what does the user want, and are you the best answer?🤖 In This Episode• Why Graphite believes 95% of SEO work does not drive meaningful impact• How Marcos defines a “topic” as a cluster of intent, not just keywords• Why topic graphs are living systems that change as search behavior evolves• How AI prompts, question clusters, and query fan-outs connect to traditional SEO and content marketing • The role of topical authority in deciding what content a site can realistically rank for• Why smaller sites may need to target more specific topic variants before expanding• How internal linking helps redistribute authority across large websites• Why content structure, FAQs, headings, and direct answers matter for AI citations• How RAG, citations, off-site reputation, and digital PR influence AEO visibility• Why consistency across owned and earned channels is critical for AI searchThis episode is packed with practical SEO strategy insight for marketers, SEOs, and growth teams trying to understand how search strategy should evolve in the age of AI answer engines.Subscribe to the Page 2 Podcast for more conversations with the people shaping the future of SEO, AEO, and digital visibility.What do you think matters more for AI search visibility right now: owned content, citations, or off-site reputation? Drop your thoughts in the comments.🛠️ Tools & Resources Mentioned• Marcos Ciarrocchi on LinkedIn - https://www.linkedin.com/in/marcosciarrocchi/• Graphite - https://graphite.io• Graphite’s 2026 Topical Authority Study - https://graphite.io/five-percent/topical-authority-white-paper• Graphite’s AEO Guide - https://graphite.io/five-percent/aeo-is-the-new-seo• Graphite’s Entity Position Tracking Guide - https://graphite.io/five-percent/how-to-track-entity-position-in-aiFollow the Page 2 Podcast✅ Page 2 Podcast on Youtube → https://www.youtube.com/@thepage2podcast922✅ Page 2 Podcast on Linkedin → https://www.linkedin.com/company/page2podcast✅ Page 2 Podcast on Facebook → https://www.facebook.com/ThePage2Podcast/Follow Jon Clark✅ Jon Clark on Bluesky → https://bsky.app/profile/jonclark.co✅ Jon Clark on Linkedin → https://www.linkedin.com/in/ppcmarketing/### Sponsored by Moving Traffic Media → https://www.movingtrafficmedia.comAbout Moving Traffic MediaMoving Traffic Media is a digital marketing agency that specializes in SEO, answer engine optimization, generative engine optimization, and performance media. In plain English: Moving Traffic Media helps brands show up in traditional search and in AI Answers. And they back that up with everything around it, from paid media and social to web development built for how AI agents actually browse. Moving Traffic Media was co-founded by Jon Clark and Joe DeVita and is headquartered in New York, with a second office in Miami. So if your brand isn't showing up in AI answers the way it should, contact Moving Traffic Media at: https://www.movingtrafficmedia.com.

⁠https://page2pod.com⁠ - While we're between seasons, we're revisiting some of the most impactful conversations we've had on The Page 2 Podcast. 

In this episode of Page 2 Podcast, Jon Clark and Joe DeVita sit down with Duane Forrester, founder and CEO of UnboundAnswers.com and a longtime search industry leader who helped launch Bing Webmaster Tools and Schema.org.

Duane shares his perspective on the future of search optimization, the rise of AI-driven visibility, and why the best SEOs may need to think less like button-pushers and more like managers of intelligent systems.

From structured data and LLM verification layers to chunking, semantic overlap, semantic density, query fan out, and the debate over GEO vs. AIO vs. SEO, this conversation digs into what actually matters for brands trying to remain discoverable in an AI-first search environment.

🤖 In This Episode

• Why Duane believes AI has fractured search, not killed it
• How structured data and Schema.org may influence LLM understanding
• Why “GEO vs. AIO vs. SEO” may be a distraction from the real work
• What semantic overlap and semantic density mean for content strategy
• How chunking can help content become more useful to AI systems and humans
• Why content creators may be better positioned for AI search than they think
• How people management skills apply to working with AI tools like ChatGPT
• Why search optimization may still have decades of relevance ahead
• What Google AI Overviews could mean for organic traffic, paid ads, and click quality
• Duane’s prediction for AI-generated ads inside future search experiences

This episode is a must-watch for SEOs, content strategists, marketers, founders, and agencies trying to understand how AI search is changing visibility, trust, and optimization.

Subscribe to Page 2 Podcast for more conversations with the people shaping the future of SEO, AI, search, and digital marketing.

What do you think: should we still call this SEO, or is AI search creating a completely new discipline? Drop your thoughts in the comments.

Sponsored by Moving Traffic Media → https://www.movingtrafficmedia.com

Follow the Page 2 Podcast
✅ Page 2 Podcast on Youtube → https://www.youtube.com/@thepage2podcast922
✅ Page 2 Podcast on Linkedin → https://www.linkedin.com/company/page2podcast
✅ Page 2 Podcast on Facebook →  https://www.facebook.com/ThePage2Podcast/

Follow Jon Clark
✅ Jon Clark on Bluesky → https://bsky.app/profile/jonclark.co
✅ Jon Clark on Linkedin → https://www.linkedin.com/in/ppcmarketing/