TL;DR
- LLMs run web searches behind the scenes — traditional SEO still matters
- Turn keywords into natural language questions (2-3 variations per keyword)
- Identify which sources LLMs cite for your queries — then show up there
- Reddit and human-generated communities get disproportionate LLM attention
- Some startups have leapfrogged traditional SEO — but the fundamentals still apply
Guest Expert
Chirag Kulkarni — Founder & CEO of Taco, an AI-powered SEO agency. Forbes 30 Under 30 (Marketing). Former CMO at Medly Pharmacy, where he scaled from 1 to 50 pharmacies and $100M+ in revenue.
What I've Learned
There's a lot of speculation in AI search presented as confident thinking. Founders I know are collecting a patchwork of advice from experiments and conversations.
But practitioners who've been in SEO for years and gone deep on LLM optimization are starting to converge on some real patterns.
LLMs are running web searches behind the scenes. ChatGPT, Gemini, Perplexity — they're querying Google or Bing. If you don't rank there, you probably won't rank in the LLM either. Traditional SEO gives you runway.
The shift is from keywords to questions. LLMs surface content that answers natural language prompts. "Dandruff shampoo" becomes "what are the best dandruff shampoos for men?" Create 2-3 question variations per primary keyword.
Third-party sources matter more than your own site. LLMs cite Reddit threads, authoritative publications, listicles. If you're in finance, that's NerdWallet. Healthcare, Healthline. The play is showing up where LLMs already look.
Reddit gets disproportionate weight — likely because it contains human conversation, not just marketing copy.
There are exceptions. Some startups have gotten strong LLM placement without traditional SEO strength — Lenny's interview with Ethan Smith breaks down how.
Timestamps
- 0:00 — Henry's question about winning AI search
- 0:22 — Why Ben brought in an expert (Chirag from Taco)
- 0:56 — Chirag's framework: heavy overlap with traditional SEO
- 1:17 — Step 1: Turn keywords into questions
- 2:17 — Step 2: Identify LLM citation sources
- 3:12 — Step 3: Create content on third-party sites
- 4:06 — Ben's founder synthesis: three takeaways
Tools & Resources
- The Ultimate Guide to AEO - Ethan Smith (Lenny's Podcast) — goes deeper on exceptions and edge cases
- Scrunch — keyword-to-question tool
- Air Ops — AI content optimization
- Profound — LLM search analytics