TL;DR
- I built a free, no-registration tool to score how likely your tech articles are to be cited by AI (ChatGPT, Perplexity).
- GEO (Generative Engine Optimization) relies on answer-first structures, concrete metrics, and clear tables.
- Paste your dev.to, Medium, or Zenn URL to get a 0-100 score and an 11-point actionable breakdown instantly.
As a solo developer building in public under the handle Slipway, I write a lot of technical content. But a question kept bothering me: How do I know if AI search engines will actually cite my work?
I wanted a way to measure this. So, I extracted the scoring logic from a larger tool I'm developing called Polypost, and released it as a free, standalone entry point for anyone to use.
What is GEO?
Generative Engine Optimization (GEO)—sometimes called AEO or LLMO—is the AI-era equivalent of SEO. While traditional SEO focuses on ranking #1 on Google, GEO focuses on formatting your content so Large Language Models (LLMs) easily understand and cite it in their answers.
According to a recent study by researchers at Princeton and Georgia Tech, optimizing content with citations, clear structures, and relevant statistics can improve AI visibility by up to 40%.
| Feature | Traditional SEO | GEO (Generative Engine Optimization) |
|---|---|---|
| Primary Goal | Rank #1 on search engine result pages | Be cited as a source in AI-generated answers |
| Keyword Strategy | Exact match search volume | Semantic relevance and context |
| Content Structure | Long-form, keyword-heavy | Answer-first, TL;DRs, FAQs, Comparison Tables |
| Trust Metrics | Backlinks, Domain Authority | Concrete numbers, authoritative citations |
How the Checker Works
It's incredibly simple. You don't need an account, and you don't need a credit card. (If you want to learn more about structuring great articles before testing them, check out the dev.to tag for technical writing).
- Paste your article URL (dev.to, Medium, Zenn, or your personal blog).
- Instantly get a 0-100 GEO score.
- Review the 11-item breakdown to see exactly what to fix.
Behind the scenes, the tool evaluates your content against 11 heuristics. Here is a simplified example of the JSON data the scoring engine generates:
{
"url": "https://dev.to/your-article",
"geo_score": 85,
"breakdown": {
"answer_first_structure": true,
"has_comparison_tables": true,
"has_faq_section": false,
"concrete_numbers_count": 4
},
"recommendation": "Add an FAQ section to improve your score by 15 points."
}
Honest Disclaimer
I just launched this, and I want to be completely transparent: the score is based on my own custom heuristics. It is not a magic bullet, and a perfect score doesn't guarantee ChatGPT will cite you tomorrow. I built this to give us a measurable baseline. I welcome any harsh feedback to help improve the algorithm!
(Note: This free checker is a one-off diagnostic tool. If you want to rewrite articles in your own voice, scan all your posts in bulk, or cross-post across platforms, I offer those as paid features in Polypost with a free trial. No hard sell, just letting you know!)
FAQ
Q: Does a 100 score guarantee a ChatGPT or Perplexity citation?
A: No. AI engines use complex, ever-changing retrieval algorithms. However, hitting a high score means your formatting (TL;DRs, tables, concrete numbers) aligns perfectly with what LLMs are trained to parse and cite.
Q: Is the tool really free?
A: Yes! The single-URL diagnostic tool is 100% free and requires no registration.
Q: Can I share my score?
A: Absolutely. The tool generates a custom OG image card with your score. Share it on X or LinkedIn and ask your network: "What's your article's score?"
Check your GEO score for free -> https://polypost.dev/geo-score
Top comments (0)