How We Optimized GetCiteFlow's
Own AI Visibility
Eating our own dog food — we ran our own AI visibility scanner against ourselves and fixed what we found.
We built a scanner that tells other sites why AI doesn't cite them. Then we ran it on getciteflow.ai — and found we were missing the very things we tell our users to fix.
Added FAQ Schema, built a comprehensive llms.txt, structured our content with clear entity definitions, and optimized meta descriptions for AI snippet extraction.
AI Visibility: 75 → 92
Measured by our own scanner, verified with real LLM queries
Key Results at a Glance
| Metric | Before | After | Change |
|---|---|---|---|
| AI Visibility Score | 75/100 | 92/100 | +17 |
| FAQ Coverage | 0% (no FAQ Schema) | 85% (7 structured Q&As) | +85% |
| Entity Clarity | Weak — brand not clearly defined for LLMs | Strong — Organization + SoftwareApplication schema | Significant |
| llms.txt | Missing entirely | Complete — all pages indexed for AI crawlers | From zero |
The irony of building a visibility scanner
We spent months building GetCiteFlow — an enterprise AI brand service with a free scanner that analyzes websites and tells you exactly why AI search engines aren't citing you. We launched it. We started writing blog posts about GEO. We built landing pages targeting "AI visibility" and "how to get cited by ChatGPT."
Then one day, almost as a joke, we ran our own scanner against getciteflow.ai. The result was humbling: 75 out of 100.
Here we were, telling other companies how to optimize for AI visibility, and our own site was missing the fundamentals. No FAQ Schema. No llms.txt. Weak entity definitions. The irony wasn't lost on us.
What our own scan revealed
The scan surfaced four clear issues:
- No FAQ Schema. We had a FAQ section on the homepage, but it was plain HTML. Without JSON-LD structured data, LLMs couldn't reliably extract our Q&A pairs for citation.
- Missing llms.txt. AI crawlers like GPTBot and ClaudeBot had no structured index of our site. They were guessing which pages mattered.
- Weak entity clarity. Our brand wasn't consistently defined across pages. Search engines and LLMs both rely on clear entity definitions to understand what a site is about.
- Content not structured for AI extraction. Our blog posts were well-written for humans, but lacked the clear heading hierarchy, tables, and lists that LLMs use to extract and cite information.
What we fixed — in order of impact
We prioritized the changes based on what would move the needle fastest:
- Built a comprehensive llms.txt. This was the quickest win. We listed every important page — blog posts, landing pages, comparison pages, case studies — with clear descriptions. Now when GPTBot or ClaudeBot crawls our site, it knows exactly what's there. Time: 2 hours.
- Added FAQ Schema to the homepage. We converted our existing FAQ section into JSON-LD structured data. Seven questions covering "What is GEO," "How does GetCiteFlow work," and common user concerns. Time: 1 hour.
- Defined our entity clearly. We added Organization and SoftwareApplication JSON-LD schema to every page, consistently using the same brand name, description, and URL. This tells LLMs: "GetCiteFlow is an enterprise brand visibility service, not a generic SEO tool." Time: 3 hours.
- Restructured blog content for AI readability. We went through our blog posts and made sure every article had: clear H2/H3 hierarchy, comparison tables where applicable, numbered lists for actionable advice, and a summary section that LLMs could extract as a snippet. Time: 1 day.
The results: from 75 to 92
After implementing these changes, we re-scanned our site. The AI Visibility Score jumped from 75 to 92.
But the score is just a number. The real test was asking ChatGPT and Claude questions like "what is GEO?" and "how do I get my site cited by AI?" — and seeing GetCiteFlow appear in the answers. It did.
More importantly, we now had a credible story to tell. When potential customers ask "does this actually work?" — we can show them our own before-and-after. Not a hypothetical. Not a fabricated case study. Our own site, our own service, real results.
The lesson: build what you need, then use it yourself
Every SaaS founder should run their own product against themselves. If you're building a visibility scanner and your own site isn't optimized for AI visibility, something is wrong. Fixing our own site didn't just improve our score — it gave us firsthand experience with the exact process our customers go through. That empathy shapes every feature we build.
Key Takeaways
- Run your own scanner on yourself first. If you're selling GEO optimization and your own site scores poorly, fix that before pitching anyone else.
- llms.txt is the highest-impact, lowest-effort change. It took 2 hours and immediately gave AI crawlers a structured map of our site.
- FAQ Schema matters more than you think. LLMs frequently cite FAQ content in responses. Without structured data, your Q&A is invisible.
- Entity clarity compounds. Consistently defining your brand across all pages helps LLMs build a reliable mental model of what you do.
- Content structure is for machines, not just humans. Clear headings, tables, and lists make your content extractable by AI — which is how citations happen.