Page 2 Podcast | SEO Podcast

@page2pod.com

The Page 2 Podcast is an interview show about the people doing SEO and AEO work: the experiments they've run, what the data showed, and how they got here. Hosted by @jonclark.co and Joe DeVita

Angela Skane on the Page 2 Podcast: she would rather optimize the content she already has than publish 20 more articles. 100 pages sitting on page eight of Google do not make a site stronger, they just make it heavier. Has a publishing quota ever worked for you?

The hard truth for clients: traffic isn't coming back. SEO was rankings and traffic; AI search is visibility and downstream conversions. People research in the LLM, exit, then run a branded search. How are you handling attribution in answer engine optimization?

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?

Andrew Shotland: proximity is the top factor for a lot of queries, and getting physically closer to customers beats 10 reviews, links, or new content. His version is a $200 sublet with a verified Google Business Profile, real staff, and signage. Full episode: www.youtube.com/watch?v=lQUL...

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

YouTube video by The Page 2 Podcast: An SEO Podcast

youtube.com

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?

Sophie Brannon's whole culture-fit screen is one question: what are you most passionate about? A work answer, or anything obviously tailored to what she wants to hear, is an immediate no. Some of her best hires said gaming. She notes it works on dates too.

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.

Distribution isn't one-size-fits-all: once on LinkedIn/Facebook, three times on X by time zone (a reader poll confirmed people want it), with different headlines and images. YouTube gets A/B-tested thumbnails and titles. How do you handle reposting the same piece?

A deliberately human-first choice: we don't automate call analysis with AI. When you automate something, you stop paying attention to it. A person listening to calls and taking notes is what sets the work apart. Where are you keeping humans in the loop on purpose?

An early SEO-testing finding for LLMs: the winning tests are about giving more information. LLMs don't get bored with long specs the way humans do — manufacturer data, product detail, and exposed info all help. It doesn't have to be UGC. What's moving the needle in your LLM tests?

AI referral traffic is sitting at 1.08%. Pat says every study keeps confirming the same thing: this was never a traffic play. What clients actually ask about is brand sentiment and visibility. He thinks the number only moves once agentic ecommerce puts clickable product cards in the conversation.

A technical SEO trap for news and evergreen content: Google often pulls the wrong date — from structured data, on-page, or CMS sections — sometimes off by years, and it kills clicks. Show one clear date, not multiple. Have you been burned by Google surfacing the wrong date?

A nuanced read on the Google clicks debate: Search Console shows clicks down, Google says up — and both can be right. At trillion-URL scale, their vetted statement reflects global data; your individual site can still contradict it. Different vantage points. Fair framing, or a cop-out?

Fascinating indexing strategy: a self-learning model, built over two years, decides which URLs are indexable using interaction and demand signals - and if a page clears the bar for the site, it usually clears Google's index too. Would you let a model gate your indexation?

Ten years before it was consensus: a team cut 80% of a client's content and rankings improved across everything that remained. That's the origin of 'quality is a site-wide metric.' Google's court docs later confirmed site authority exists. Have you seen pruning lift a whole site?

A framing worth sitting with: AI literacy is becoming like internet literacy: a skill you're economically locked out without. The way to build the muscle is to treat it like a person and pressure-test it constantly. How are you building AI literacy on your team?