How I Got 2,400 AI Citations in 3 Weeks: A Programmatic SEO + AEO Playbook
Forget traditional SEO. The new game is getting AI to recommend you.
Traditional SEO is dying.
Google's AI Overviews now answer 30% of searches without sending traffic to any website. ChatGPT, Perplexity, and Claude are becoming the new search engines for millions of users.
But here's the twist: there's a new way to get traffic from AI — and it's easier than traditional SEO.
It's called AEO (AI Engine Optimization), and one experimenter proved it works: 2,400+ AI citations in 3 weeks on a brand new domain with zero backlinks.
I studied his process, replicated it, and broke it down into a step-by-step playbook anyone can follow.
What Is AEO (AI Engine Optimization)?
AEO is the practice of structuring your content so AI systems (ChatGPT, Perplexity, Claude, Google AI Overviews) can easily understand, index, and cite it.
Think of it this way:
- Traditional SEO = convincing Google's algorithm to rank you
- AEO = making it easy for AI to quote you
The difference? SEO takes months. AEO can show results in weeks.
The Experiment That Proved It Works
An anonymous experimenter (shared on Reddit and IndieHackers) did this:
- Registered a brand new domain — zero history, zero backlinks
- Published 200+ pages of programmatic content targeting long-tail keywords
- Added FAQ Schema markup to every page
- Structured every answer in a clear, quotable format
Results in 3 weeks:
- 2,400+ citations from ChatGPT, Perplexity, and Google AI Overviews
- ChatGPT crawled the site 320+ times in 8 hours
- The domain went from zero to measurable organic traffic
No ads. No backlinks. No social media. Just structured content.
The 5-Step Playbook
Step 1: Find Your Long-Tail Keywords
Don't compete on "best CRM" or "how to lose weight." Those are too competitive.
Instead, target questions that real people ask but nobody answers well:
How to find them:
- Search your topic on Reddit and note the exact questions people ask
- Use AnswerThePublic (free tier) to find question-based queries
- Check "People Also Ask" on Google for your topic
- Look at ChatGPT/Perplexity responses and note where they say "I don't have enough information"
Example long-tail keywords:
- "how to automate flower shop delivery scheduling"
- "best CRM for solo real estate agents under $50/month"
- "how to add row-level security to Supabase"
- "n8n vs Make.com for small e-commerce store"
These get 10-100 searches/month each. But 200 of them = 2,000-20,000 searches/month total.
Step 2: Create Programmatic Content
Don't write 200 blog posts by hand. Use templates + AI to generate them at scale.
The template structure:
# [Question] — Complete Guide [Year]
## Quick Answer
[2-3 sentence direct answer — this is what AI will quote]
## Detailed Explanation
[300-500 words of context]
## Step-by-Step Guide
1. [Step 1]
2. [Step 2]
3. [Step 3]
## Common Mistakes
- [Mistake 1]
- [Mistake 2]
## FAQ
Q: [Related question 1]
A: [Direct answer]
Q: [Related question 2]
A: [Direct answer]
Why this works: AI systems look for clear, structured, quotable answers. The "Quick Answer" section is designed to be directly quoted by ChatGPT and Perplexity.
Step 3: Add FAQ Schema Markup
This is the secret sauce. FAQ Schema tells AI systems "this page has questions and answers."
The HTML looks like this:
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "FAQPage",
"mainEntity": [
{
"@type": "Question",
"name": "What is the best CRM for solo real estate agents?",
"acceptedAnswer": {
"@type": "Answer",
"text": "For solo real estate agents under $50/month, the top options are..."
}
}
]
}
</script>
How to add it:
- WordPress: Use Yoast SEO or RankMath (free)
- Static sites: Add the JSON-LD to each page template
- Custom sites: Add it to the
<head>section
Step 4: Optimize for AI Crawling
AI systems need to be able to read your content. Here's how to make it easy:
Do:
- Use clear heading hierarchy (H1 → H2 → H3)
- Include the question in the H1 tag
- Put the direct answer in the first paragraph
- Use bullet points and numbered lists
- Include the year in titles (AI prefers current information)
Don't:
- Hide content behind JavaScript rendering
- Use images for text
- Require login to see content
- Use infinite scroll (AI can't scroll)
- Block AI crawlers in robots.txt
Step 5: Monitor and Iterate
Track your AI citations using:
- Perplexity Analytics (if available)
- Manual testing: Search your topics on ChatGPT and Perplexity, see if you're cited
- Google Search Console: Track AI Overview appearances
- Server logs: Look for AI crawler user agents (GPTBot, PerplexityBot, ClaudeBot)
What to iterate on:
- Pages that get cited → create more like them
- Pages that don't get cited → check structure, add FAQ Schema, improve the "Quick Answer"
- New questions that appear → add pages for them
Real Results from Real Implementations
Case 1: Travel Affiliate Site
- Strategy: 500+ pages targeting "best [activity] in [city]" queries
- Results: $6.8K in brand deals + $2K/month in affiliate revenue
- Time: 6 months to first significant revenue
- Traffic source: 20% from Pinterest, 80% from organic search + AI citations
Case 2: Legal Lead Generation
- Strategy: 100+ pages targeting "[practice area] lawyer in [city]"
- Results: 3 leads in month 1, 6% Google conversion rate
- Time: 1 month to first lead
- Business model: Rent the leads to local lawyers ($500-2,000/month)
Case 3: Local Business Directory
- Strategy: 1,000+ pages targeting "best [service] near [location]"
- Results: 10,000+ monthly visitors by month 4
- Time: 4 months
- Monetization: Google Ads + affiliate links
The Math That Makes This Work
| Metric | Conservative | Aggressive |
|---|---|---|
| Pages published | 200 | 1,000 |
| Avg. searches/page/month | 50 | 50 |
| Total monthly searches | 10,000 | 50,000 |
| Click-through rate | 3% | 5% |
| Monthly visitors | 300 | 2,500 |
| Revenue per 1,000 visitors | $10 | $10 |
| Monthly revenue | $3 | $25 |
Wait — $3/month? That's nothing.
But here's the thing: You build it once, and it compounds. Month 2 adds another 200 pages. Month 3 adds another 200. By month 6, you have 1,200 pages generating $36-300/month on autopilot.
And that's the conservative estimate. The travel affiliate case generated $2K/month from 500 pages.
Tools You Need (All Free or Cheap)
| Tool | Purpose | Cost |
|---|---|---|
| Next.js or Astro | Static site generator | Free |
| Vercel or Netlify | Hosting | Free tier |
| AnswerThePublic | Keyword research | Free tier |
| ChatGPT/Claude | Content generation | $20/month |
| Yoast SEO | FAQ Schema (WordPress) | Free |
| Google Search Console | Monitoring | Free |
Total startup cost: $20/month (for AI content generation)
Common Mistakes
Don't create thin content. Each page needs genuine value, not just a template filled with keywords. AI systems are getting better at detecting low-quality content.
Don't ignore freshness. Update your content quarterly. AI prefers current information.
Don't skip the FAQ Schema. It's the single most important technical element. Without it, AI systems may not recognize your Q&A format.
Don't expect overnight results. The 2,400-citation experiment was unusual. Typical results take 4-8 weeks to show.
Don't forget to monitor. If you're not tracking citations, you're flying blind.
Your Next Step
- Pick a niche you know something about
- Find 50 long-tail questions in that niche
- Create a template page with the structure above
- Generate 10 test pages using AI
- Add FAQ Schema
- Publish and wait 2 weeks
- Check if AI systems are citing you
If they are → scale to 200+ pages.
If they're not → adjust the structure and try again.
Have you tried AEO or programmatic SEO? What results are you seeing? I'd love to hear your experience.
About the author: I study how businesses get traffic from AI systems. No fluff, just data and playbooks. Follow for more.
Top comments (0)