iiinova — where your digital self keeps evolving

Jon Clark @jonclark.co · 30/09/2026

6/ Full conversation on the Page 2 Podcast: youtu.be/lQUL8tZEo0k Local SEO folks, what's changed most in your approach since AI search showed up?

youtu.be

Local SEO Experiments That Actually Worked: Maps, AI, Reddit & Reviews 🧪 | Andrew Shotland

YouTube video by The Page 2 Podcast: An SEO Podcast

010

Jon Clark @jonclark.co · 30/09/2026

5/ On Reddit and AI visibility: a month of seeded brand mentions spiked AI visibility, and it vanished when they stopped. Now they monitor relevant subreddits and flag threads for clients to join genuinely.

101

Jon Clark @jonclark.co · 30/09/2026

4/ On Apple Maps ads: few advertisers yet, and one client is paying roughly 11-12 cents a click. Tracking is the hard part. You can't separate paid from free clicks in analytics, so he adds a UTM to the profile URL and watches referrals.

100

Jon Clark @jonclark.co · 30/09/2026

3/ On fast leads for a new service business: open a legitimate second location (a real sublet, signage, a business license, a verified Google Business Profile). Proximity is the top factor for many local queries! One HVAC client's new office did $10K in its first month.

100

Jon Clark @jonclark.co · 30/09/2026

2/ On reviews: his research with Local Falcon across ~7M local queries found the first ~200 reviews influence rankings in local packs, Google AI Overviews, and ChatGPT. After that the gains flatten, but more reviews seem to widen the radius where a business ranks.

100

Jon Clark @jonclark.co · 30/09/2026

Local SEO isn't just the Google Business Profile anymore. @localseoguide.com walked us through what's moving local visibility across Google, ChatGPT, Apple Maps, and Reddit. A few highlights 🧵

Andrew Shotland discussing Local SEO on the Page 2 Podcast with Jon Clark and Joe DeVita

110

Reposted by Jon Clark

Page 2 Podcast | SEO Podcast @page2pod.com · 30/09/2026

Two AI tools worth trying daily: Fathom for call notes so you stay present, and Whisper for voice-to-text everywhere — Slack, terminal, even 'voicing' code instead of typing it. Voice AI is a genuine speed unlock. What's the one AI tool you can't work without?

012

Reposted by Jon Clark

Glenn Gabe @glenngabe.bsky.social · 24/09/2026

Check it out now. It showed up this morning. Very cool addition in GSC -> Multimodal Search Type Filter In Google Search Console Performance Report And if you focus on health & medical, check it out soon. You might see a lot of clicks and impressions there. :) www.seroundtable.com/google-searc...

2112

Jon Clark @jonclark.co · 23/09/2026

The reviews page on your own site is worthless. @greggifford.bsky.social's proof: filter out your staff's IPs and traffic goes to zero. Customers only trust reviews on sites you don't control. The only version that works is video. Is he wrong? Full @page2pod.com episode: youtu.be/XXAsZESr4a0

youtu.be

Local SEO in AI Search: What Actually Works 🔍 with Greg Gifford

YouTube video by The Page 2 Podcast: An SEO Podcast

140

Reposted by Jon Clark

Page 2 Podcast | SEO Podcast @page2pod.com · 21/09/2026

New episode of Page 2 Podcast is live with Greg Gifford. He breaks down what actually proves local relevance to Google across sales, service, and parts departments, and why category selection matters more than most dealers think. What's the local SEO myth you hear most often?

022

Jon Clark @jonclark.co · 21/09/2026

5/ AI search is probabilistic, not deterministic. There's no rank tracking, so tools count citations instead. Most people never click those links anyway. They read the answer, then search on their own. The recommendation is what matters. Full episode: youtu.be/XXAsZESr4a0

youtu.be

Local SEO in AI Search: What Actually Works 🔍 with Greg Gifford

YouTube video by The Page 2 Podcast: An SEO Podcast

010

Jon Clark @jonclark.co · 21/09/2026

4/ Review gating (asking "rate us 1-5" and only sending the happy ones to Google) breaks Google's rules and FTC law. Google is removing 400-500 reviews at a time from businesses that told customers what to write. Those don't come back.

100

Jon Clark @jonclark.co · 21/09/2026

3/ A landing page for every city in the state is doorway spam, not local SEO. He saw a prospect with a page for every city in Texas, identical copy, city swapped. Four or five suburbs in your own metro is closer to right. And own your backyard before you expand.

100

Jon Clark @jonclark.co · 21/09/2026

2/ Your primary GBP category is the single most powerful lever you have for the map pack, and most businesses waste it on the generic option. If you're the only Ford dealer for 100 miles, you already rank for that. Pick the category you're actually fighting over.

100

Jon Clark @jonclark.co · 21/09/2026

1/ @greggifford.bsky.social has been doing local SEO since before "AI search" was a phrase. Four things from his @page2pod.com episode that most agencies still get wrong about Google Business Profile, reviews, and AI answers 🧵

120

Jon Clark @jonclark.co · 14/09/2026

“Video is becoming table stakes for SEO.” That’s Cindy Krum on the latest @page2pod.com. If you’re traveling to #BrightonSEO today, give it a listen before her talk on “Video SEO to help fast-track AIO rankings.” Consider it the pre-game. 🎧 youtu.be/Qr9PFu3GH6M?...

youtu.be

Why Video Is Now Table Stakes for SEO 🎥 with Cindy Krum

YouTube video by The Page 2 Podcast: An SEO Podcast

110

Reposted by Jon Clark

Page 2 Podcast | SEO Podcast @page2pod.com · 14/09/2026

Vertical video isn't ranking in AI Overviews yet. Cindy Krum, CEO of MobileMoxie, explains on @page2pod.com that Google tends to test new formats like images and video shorts in their own tab before folding them into search. Integration into AI Overviews looks like a matter of time.

031

Jon Clark @jonclark.co · 10/09/2026

Does this have to be done from desktop?

000

Jon Clark @jonclark.co · 08/09/2026

Everyone has an AI-search theory. Otterly AI is actually testing them. One wild result: publishers with OpenAI licensing deals were cited by ChatGPT ~48% more often. Plus: fake brands can earn recommendations, schema may not help ChatGPT, and YouTube citations are rising.

Thomas Peham CEO of Otterly AI on The Page 2 Podcast.

211

Jon Clark @jonclark.co · 01/09/2026

AI is making it easier to create the average answer. Which might make having an actual opinion more valuable than ever. @lidia-infante.com on why going against consensus could become a real advantage. 🎧 Listen to the new @page2podcast.bsky.social episode + subscribe. youtu.be/wbSavynl0S4?...

132

Jon Clark @jonclark.co · 30/08/2026

AI-search hot take: LLMs.txt probably isn’t the hack you want it to be. Clear copy + consistent facts + accessible content + third-party consensus are harder and far more useful. 🎧 Listen to the episode with @landwehr.bsky.social + subscribe to @page2podcast.bsky.social youtu.be/ZntIGuNx29c?...

000

Jon Clark @jonclark.co · 26/08/2026

Also in the conversation: fan-out queries running site: commands on domains that don't exist, and how his team turned that into an expired-domain finder. Full @page2podcast.bsky.social episode: youtu.be/ZntIGuNx29c?... Which of the three above are you still doing anyway?

youtu.be

What Fan-Out Queries Reveal About AI Search 🔍 with Malte Landwehr

YouTube video by The Page 2 Podcast: An SEO Podcast

000

Jon Clark @jonclark.co · 26/08/2026

The boring thing that works: describe yourself identically everywhere. Not this .. Website: "SEO agency for lawyers." Founder's LinkedIn: "digital visibility in the jurisdictional industry." Press release: "search visibility for law offices." To an LLM, that's three companies.

100

Jon Clark @jonclark.co · 26/08/2026

3. llms.txt as a visibility play. He's never seen one lift visibility in an answer. Its real job is telling actual agents where your API, MCP and docs live. His line: he doesn't know anyone strict about measurement who thinks llms.txt is a lever.

100

Jon Clark @jonclark.co · 26/08/2026

2. Schema as a delivery mechanism for facts. Published test: when the page text and the schema disagree, the text wins. When the fact lives ONLY in schema, ChatGPT answered ~20% of the time and Claude 0%. Schema is a forcing function for data hygiene. Not a pipeline.

100

Jon Clark @jonclark.co · 26/08/2026

1. Markdown copies of your pages. He ran it across hundreds of thousands of chats. In 0% of them did an LLM cite the .md version over the HTML page. What you get instead: duplicate content, double the crawl, a diluted internal link graph.

100

Jon Clark @jonclark.co · 26/08/2026

@landwehr.bsky.social has tested most of the AI search checklist people are selling right now. Three of the most popular items have never produced a measurable visibility lift for him. The thing that does work is boring enough that nobody builds a product around it.

210

Jon Clark @jonclark.co · 21/08/2026

We investigate traffic drops like detectives. Traffic spikes? Pats on the back and “nice job, everyone.” That’s backwards. If you don’t know why you won, you can’t reliably do it again. youtu.be/YrE21LqTU2o?... 🎧 More from @seohumorist.bsky.social on @page2podcast.bsky.social. #SEO

youtu.be

Direct Traffic Is Your Missing AI Metric 📈 - Derek Hobson on Why You Should Report SEO + Direct

YouTube video by The Page 2 Podcast: An SEO Podcast

021

Reposted by Jon Clark

SEO Humorist @seohumorist.bsky.social · 17/08/2026

Super excited I got to chat with @jonclark.co and Joe DeVita on the @page2podcast.bsky.social!! We've all seen people scramble to explain why performance is down and sometimes just "accept" good results, but if you can’t explain the win, you can’t replicate it. youtu.be/YrE21LqTU2o

youtu.be

Direct Traffic Is Your Missing AI Metric 📈 - Derek Hobson on Why You Should Report SEO + Direct

YouTube video by The Page 2 Podcast: An SEO Podcast

022

Reposted by Jon Clark

Page 2 Podcast | SEO Podcast @page2pod.com · 30/07/2026

On content in regulated industries like legal and health: center their fears, treat the constraints as a box, and shape your strategy within it. The output won't be exactly what you envisioned, and that's okay if it still serves the user. How do you work with compliance?

121

Reposted by Jon Clark

Page 2 Podcast | SEO Podcast @page2pod.com · 27/07/2026

An "About us" H2 with a class and ID costs an AI agent 12-15 tokens to ingest. The same content in markdown costs 3, an 80% cut. Yoast's @alexmoss.bsky.social compares it to mobile-first indexing: agents don't read HTML the way we do. Writing markdown-first for AI search yet?

041

Reposted by Jon Clark

Page 2 Podcast | SEO Podcast @page2pod.com · 21/07/2026

A team-building take for AI search: don't hire AEO/GEO "experts" with no SEO foundation. Everyone on the SEO team already does AEO work - you evolve their skills. The real shift is closer work with social and more in-house digital PR. How is your team adapting? Full episode with @lilyray.nyc

143

Reposted by Jon Clark

Page 2 Podcast | SEO Podcast @page2pod.com · 20/07/2026

Googlebot renders a page at a fixed viewport, then checks the page height once during rendering. Infinite scroll and lazy loading pages that need multiple scrolls to reveal content mostly lose everything past that first check. If you work in technical SEO or JavaScript rendering, worth a listen!

112

Reposted by Jon Clark

Page 2 Podcast | SEO Podcast @page2pod.com · 16/07/2026

Something hopeful about AI search: nobody (not even 20-year veterans) knows what's next. That levels the field. Whoever experiments and learns in the open has real opportunity. Is this the best time to enter SEO/AEO, or the hardest? Full episode with @garrettsussman.bsky.social out now

011

Reposted by Jon Clark

Page 2 Podcast | SEO Podcast @page2pod.com · 14/07/2026

Local SEO truth: review recency is critical. 10,000 reviews won't save you if the last one was months ago... it signals you may not be operating. The review card with a QR code, handed over at close, beats software for small businesses. What's your review engine?

031

Reposted by Jon Clark

Page 2 Podcast | SEO Podcast @page2pod.com · 06/07/2026

Why Local Service Ads won't scale: they serve for the same keywords as your PPC, so Google runs them ~15% as often, mostly in high-conversion hours and near-me searches. Google won't kill its own golden goose. Anyone found a way around the LSA ceiling? Full episode🎙️ www.youtube.com/watch?v=YJUL...

021

Reposted by Jon Clark

Page 2 Podcast | SEO Podcast @page2pod.com · 02/07/2026

In AI search, keyword research is research. Not the content brief. You group keywords into topic entities, define the buyer and journey stage, then let strategy flow down. Strategy-down, not keyword-up. Anyone still winning with keyword-up in AEO?

031

Jon Clark @jonclark.co · 22/06/2026

Is this AI Overview / AI Mode-like feature on Reddit new?

110

Jon Clark @jonclark.co · 19/06/2026

What it feels like reading AI-generated content.

A bunched up sock in a shoe with a hole. It exemplifies what it feels like reading fully AI-generated content.

000

Jon Clark @jonclark.co · 18/06/2026

🤚 I'm the guy with the laptop :/

010

Jon Clark @jonclark.co · 14/06/2026

TWO QUIRKS WORTH KNOWING 1. Claude's query fan-outs are near deterministic. Same fan-out 65% of the time across users. That makes it testable. 2. And right now, Claude's fan-outs are obsessed with years. Put the year in your page titles. Claude is the most optimizable LLM on the market today.

000

Jon Clark @jonclark.co · 14/06/2026

"How does," "what is," and "steps to" answered from memory. CLAUDE IS NOT CHATGPT Domain overlap between Claude & ChatGPT citations: 8%. Or: 92% of what is cited in ChatGPT not cited in Claude. Claude's results overlap 64% with Google. Your Google work transfers. Your ChatGPT work mostly doesn't

Claude versus ChatGPT: top-cited content

100

Jon Clark @jonclark.co · 14/06/2026

WHEN CLAUDE ACTUALLY SEARCHES ChatGPT web search ~90%. Claude? Just 36.6% What triggers Claude search: Recency prompts ("best XYZ") → 81% Ranking prompts ("top 10...") → 67% Location ("restaurants in...") → 55% Comparisons ("X vs. Y") → 51% Prioritize "best," "top," "near me," & comparison terms.

110

Jon Clark @jonclark.co · 14/06/2026

Claude doesn't re-rank search results. It transposes Brave's top 10 directly into its answers. Rank 1-10 on Brave, and you rank in Claude. That was the biggest takeaway from Zero Click by @profound session this afternoon. My notes:

79.2% of all Claude citations come from Brave ranks 1-10

210

Reposted by Jon Clark

John Mueller @johnmu.com · 08/06/2026

Occasionally we get questions about whether dashes in domain names are ok for SEO (they're ok). So far, I haven't seen anyone ask the other question - HOW MANY DASHES ARE OK? Folks, the answer is apparently 61. Now you can read the link in peace.

6274

Jon Clark @jonclark.co · 25/05/2026

The future of SEO tooling might be: → AI agents making changes → monitoring systems detecting them → automated testing workflows validating results @nickswan.bsky.social is already building toward it with SEOTesting.com. New @page2podcast.bsky.social episode breaks it down 🎙️ youtu.be/-diODWM6Bqc

youtu.be

🚀 Nick Swan Reveals the Future of SEO Testing, AI Workflows & MCP Servers

YouTube video by The Page 2 Podcast: An SEO Podcast

010

Reposted by Jon Clark

Lily Ray 😏 @lilyray.nyc · 22/05/2026

Markdown files double the size of your website and create needless duplication We can instead use content-type: text/markdown but the default is still HTML. This reduces having a duplicate URL Have one url per piece of content. @seo.guru

3181

Jon Clark @jonclark.co · 19/05/2026

AI visibility measurement tip from Aimee Jurenka: Stop single-shot prompt tracking. It's like playing slots. Instead: run 20 prompts × 5 times → aggregate → that's your mention rate. Statistically valid. Borrowed from 1977 voting poll methodology. 🎙️ Full breakdown → youtu.be/z4a_jYdekdY

youtu.be

"The Funnel Got Eaten by LLMs" — Aimee Jurenka on Measuring AI Visibility

YouTube video by The Page 2 Podcast: An SEO Podcast

000