Are people asking ChatGPT for recommendations instead of Googling you? Let's dissect VisibAI, the tool auditing your brand's AI visibility.
We've spent years sweating over meta tags, keyword density, and backlinks just to please the Google gods. And then, users suddenly decided they are too lazy to click links anymore. They just open ChatGPT or Perplexity and ask: "Hey AI, recommend the best budget project management tool in Europe."
Just like that, your hard-earned SEO rank vanishes into thin air. Today, let's dissect a trending Product Hunt tool called VisibAI—a platform claiming to solve this new "AEO" (AI Engine Optimization) blind spot before you lose all your organic traffic to LLM scrapers.
According to founder Francesco, who spent years selling SaaS across Europe, search behavior is rapidly shifting. Traditional search is bloated with ads, so buyers now rely on AI assistants for recommendations.
The problem? Most brands have absolutely no clue if they even exist in the minds of these LLMs. VisibAI aims to fix this with some neat features:
robots.txt rules, JSON-LD schema, llms.txt, and FAQ markups to make your site super digestible for AI bots.Developers are naturally skeptical. They don’t just buy into pretty landing pages. VisibAI's launch on Product Hunt immediately faced some hard-hitting questions.
One founder ran a free audit for his startup and received a score of 46, despite getting a wall of zeros on all actual AI queries. He raised an eyebrow: "If my mentions are zero, how am I getting a 46? Is this score generated by vibe check?"
Francesco had to jump in and admit a UI failure: "The 46 actually comes from the technical readiness half of the score (sitemap, schema, crawlability). We fused 'is your site built to be cited' and 'are you actually cited' into one number, which makes it look broken. We are splitting those into two separate scores to avoid confusion."
Another solid punch came from user @noctis06: "LLMs are non-deterministic. If you refresh ChatGPT, you get a completely different answer. Do you sample queries multiple times and average them, or is this just a single-shot lucky guess?"
And the founder kept it real: "Today, it's just a single-shot snapshot. You're right that one run isn't gospel. Multi-sampling and averaging are high on my backlog. For now, we rely on re-scanning over time to smooth the noise."
Furthermore, VisibAI accesses the models via official APIs rather than browser scraping. It's clean and stable, but it represents the API model's core knowledge, not necessarily a pixel-perfect match of what a signed-in user sees on their app dashboard.
There's no denying that the landscape is changing, and various ai tools are eating traditional search traffic. VisibAI isn't a scam; it targets a very real, growing blind spot. However, in its current state, it feels more like a glorified API testing script than a bulletproof AEO engine.
But here are the pragmatic takeaways for us code monkeys to keep our web traffic alive:
robots.txt, don't act surprised when ChatGPT pretends your business doesn't exist.llms.txt file at your root—it's becoming the new standard for LLMs to quickly grasp what your site is about.Prepare your site for the AEO wave now, or watch your traffic slowly bleed to zero.
Check out the full discussion and product launch on Product Hunt - VisibAI