00:00 - Answer Engine Optimization (AEO): The Hype vs The Data
01:00 - Ethan introduced as CEO of Graphite Growth, researcher on AI and SEO trends
04:14 - Ethan's study: AI usage is larger than expected, mostly on mobile apps
05:32 - Search is not declining; macro data shows stable search volume
06:33 - The pie is growing: AI and search combined show expanding total usage
08:01 - AI usage is higher outside the US; India is now the #1 country by total usage
08:38 - Claude's explosive growth since mid-January, driven by work-related use cases
09:33 - AI introduces new use cases: prompts are longer, higher intent, more complex than searches
11:04 - The invisible long tail: most prompts are 10+ words but barely tracked by marketers
11:54 - How to find long-tail prompts: Reddit threads, sales calls, help center data
15:28 - AI summarizes consensus, not just top citations; multiple URLs per topic matter
17:31 - AI answers are not random: objective questions get stable answers, subjective ones vary
20:00 - AEO strategy: pick your prompts, track them, identify winning page types and off-site channels
21:59 - Lululemon case study: indirect content strategy to shift negative earned media narratives
24:32 - Tracking LLM conversions is nearly impossible; treat them as a dark referral network
28:56 - Lemore: tracking requires trial and error; focus on lower-funnel metrics like leads and conversions
33:57 - AI click loss from overviews is real but modest; featured snippets already had similar effect
35:37 - Mining Reddit data: using Gemini for scraping, Reddit API for scale, prompt design to avoid hallucinations
38:29 - Personalized AI in Gemini may override all SEO efforts using personal data history
40:54 - Meta AI and WhatsApp as an emerging personalized LLM channel businesses should prepare for
43:29 - Lemore and Ethan: combining proprietary data sources with Claude yields the most valuable insights
48:35 - SEO job market: low-skill tasks at risk but strategic and off-site roles are in high demand
51:52 - Ethan: SEO demand is at its highest, especially for off-site and answer engine optimization skills
57:54 - Closing recommendations: follow practitioners, refresh Python skills, embrace vibe coding tools
Why this matters:
AI-driven answer engines aren’t going anywhere and learning how to prepare your clients’ web presences to deal with this new reality can make the difference between a successful, growing agency and a failed business endeavor. Guess work in this area is not an option.
Register for this Duda webinar now to gain the practical, strategic clarity required to make data-driven decisions and drive measurable impact today
What we’ll cover:
- AI Trends: Growth & Paradigm Shift
- AI Question Research
- Earned vs Owned Answers & Optimization Strategies
- Question Types
- Answer Tracking
01:00 - Ethan introduced as CEO of Graphite Growth, researcher on AI and SEO trends
04:14 - Ethan's study: AI usage is larger than expected, mostly on mobile apps
05:32 - Search is not declining; macro data shows stable search volume
06:33 - The pie is growing: AI and search combined show expanding total usage
08:01 - AI usage is higher outside the US; India is now the #1 country by total usage
08:38 - Claude's explosive growth since mid-January, driven by work-related use cases
09:33 - AI introduces new use cases: prompts are longer, higher intent, more complex than searches
11:04 - The invisible long tail: most prompts are 10+ words but barely tracked by marketers
11:54 - How to find long-tail prompts: Reddit threads, sales calls, help center data
15:28 - AI summarizes consensus, not just top citations; multiple URLs per topic matter
17:31 - AI answers are not random: objective questions get stable answers, subjective ones vary
20:00 - AEO strategy: pick your prompts, track them, identify winning page types and off-site channels
21:59 - Lululemon case study: indirect content strategy to shift negative earned media narratives
24:32 - Tracking LLM conversions is nearly impossible; treat them as a dark referral network
28:56 - Lemore: tracking requires trial and error; focus on lower-funnel metrics like leads and conversions
33:57 - AI click loss from overviews is real but modest; featured snippets already had similar effect
35:37 - Mining Reddit data: using Gemini for scraping, Reddit API for scale, prompt design to avoid hallucinations
38:29 - Personalized AI in Gemini may override all SEO efforts using personal data history
40:54 - Meta AI and WhatsApp as an emerging personalized LLM channel businesses should prepare for
43:29 - Lemore and Ethan: combining proprietary data sources with Claude yields the most valuable insights
48:35 - SEO job market: low-skill tasks at risk but strategic and off-site roles are in high demand
51:52 - Ethan: SEO demand is at its highest, especially for off-site and answer engine optimization skills
57:54 - Closing recommendations: follow practitioners, refresh Python skills, embrace vibe coding tools
Why this matters:
AI-driven answer engines aren’t going anywhere and learning how to prepare your clients’ web presences to deal with this new reality can make the difference between a successful, growing agency and a failed business endeavor. Guess work in this area is not an option.
Register for this Duda webinar now to gain the practical, strategic clarity required to make data-driven decisions and drive measurable impact today
What we’ll cover:
- AI Trends: Growth & Paradigm Shift
- AI Question Research
- Earned vs Owned Answers & Optimization Strategies
- Question Types
- Answer Tracking
- Категория
- Бизнес
Комментариев нет.











