How SEO Works in the AI Search Era

SEO in the AI era still depends heavily on the fundamentals of traditional search, even as users increasingly get answers from AI systems. The tutorial's presenter argues that Google continues to process billions of searches and send far more website traffic than ChatGPT, while AI search creates an additional discovery channel rather than replacing Google outright. His beginner framework therefore treats SEO and AI visibility as connected goals: build content that can rank in search and also become useful source material for generated answers.0:32

His central argument is structural: AI assistants are built on top of traditional search infrastructure. ChatGPT uses Bing's index when it searches the web, and Perplexity, though it runs its own crawler, still leans heavily on Bing underneath, and crawls the same pages that already rank in traditional search. If your site doesn't show up in Google or Bing, he says, there's a good chance it's invisible to AI assistants too. In his framing, 'SEO is not competing with AI search. It's the foundation AI search is built on.'1:17

That reframing carries a business point worth taking seriously: the mechanics of ranking haven't changed so much as the number of surfaces they now feed. The same authoritative content, clear structure, and citations that win Google rankings, he notes later, are also what AI assistants choose to cite.43:55

The tutorial's first substantive lesson is search intent: figuring out what a searcher actually wants and matching it. The presenter sorts queries into four types (informational, commercial, transactional, and navigational) and offers a disarmingly simple research method. Google your keyword and look at what ranks: the top ten results are Google telling you exactly what the searcher wants, and your job is to match that intent and do it better.5:56

The numbers make the case for specificity. The head term 'coffee' draws over 650, 000 US searches a month but produces a chaotic results page of shops, definitions, and brands. A phrase like 'best dark roast coffee beans for espresso' gets only a few hundred searches, yet the intent is crystal clear and the competition a fraction as fierce. These long-tail keywords, he explains, are where newer sites should start, building authority on specific phrases before climbing to the big ones.7:24

One wrinkle complicates this picture. AI overviews now summarize results across commercial and transactional queries too, meaning Google may answer a buyer before they click anything, what he calls zero-click marketing. His response: content must go deeper than the summary covers, because surface-level material will not earn the click anymore.9:05

On prioritization, the presenter is emphatic: start with money pages, the product, service, pricing, and feature pages that directly convert. He faults brands that chase high-volume blog topics first and then wonder why nobody buys; without compelling conversion pages, 'all the traffic in the world really doesn't matter.'19:28

His case studies show how keyword choice depends on what happens after the landing. PandaDoc ranks for 'contract template' (roughly 300, 000 monthly searches with no direct buying intent) and turns it into a pipeline: free template download, email capture, nurture sequence, product signup. Counterculture Coffee, meanwhile, targets 'pour over coffee' and related ratio queries with educational brew guides, becoming an authority in specialty coffee first and letting product sales follow. The lesson he draws: a keyword without a conversion path is vanity metrics; with one, it can be an automated pipeline.12:16

From there the plan is to build one topical cluster at a time rather than scattering posts across unrelated topics. Using Counterculture's subscription page (which ranks for 'coffee subscription' at 15, 000 searches a month) he sketches a full funnel: transactional money page, commercial comparison content linking back to it, and educational posts feeding both. Comprehensive coverage, he argues, builds the topical authority that makes every keyword in the cluster easier to rank for.20:18

On content creation, the presenter is pragmatic about AI. Let it handle structure, scaffolding, definitions, and sentence flow, he suggests, but then add what AI can't: your experience, your opinions, original data, expert quotes, and specific real-world outcomes. In a sea of AI-generated material, that human-added layer is what stands out. The scale of the problem is stark: citing an Ahrefs study of 14 billion web pages, he notes that over 96% get zero traffic from Google, because most content fails not by being terrible but by being the same as everything else.26:34

Link building he calls the hardest part of SEO, full stop, time-consuming, often like cold outreach into the void, with no shortcuts that don't eventually get you penalized. His stated belief, which he acknowledges some may mock, is that genuinely great content earns links more naturally than expected. Counterculture's brew guide is linked by WikiHow, Gear Patrol, and Breville; Huntress's ransomware posts are cited by Google Cloud's threat intelligence team and referenced on Wikipedia.35:46

For beginners, his sequencing advice is modest: don't obsess over links yet. Create 10 to 15 great pieces of content first, then start with free journalist query platforms like Source of Sources, monitoring for 15 minutes a day, while accepting it's a numbers game where a hundred submissions might yield one link.39:12

The new frontier, in his telling, is a set of answer engines rather than link lists. People ask ChatGPT, Gemini, Perplexity, and Copilot for recommendations, and these assistants synthesize answers, recommend specific brands, and cite whichever sources they judge authoritative and clear. His practical advice tracks the fundamentals: comprehensive content with direct answers, topical clusters, citations from other trusted sites, and presence in 'best of' roundups, which AI assistants especially favor.41:36

He is candid about measurement gaps. Google Search Console cannot see traffic from ChatGPT or Perplexity, and nobody outside those companies knows what users actually type there, the data lives inside OpenAI and Perplexity, and they aren't sharing it. The workable approach, he says, is monitoring whether your brand shows up in AI answers at all, using tools or manual checks, since AI recommendations shift dramatically over time.44:09

The tutorial closes with a 30-day plan: week one for Search Console setup, keyword research, and identifying money pages; week two for publishing or optimizing the single most important money page; week three for a first supporting cluster with internal links; week four for promotion and monitoring early signals like rising impressions. He tempers expectations (you probably won't see results in 30 days) but frames the outcome plainly: find what people search for, create the best content for that search, and earn trust. The fundamentals haven't changed in 20 years; what's changed is that the same work now makes you visible in Google, AI overviews, ChatGPT, and Perplexity alike.47:32