Bersyn
AI visibility scanner — see what ChatGPT says about your pro
Bersyn went live on Product Hunt this morning. The launch itself is doing fine and not much more than that. The interesting part was watching strangers run their own domains through it in real time.
A few things from today's scans, without naming anyone:
An education company came back named zero times out of twelve buyer conversations. Not low. Zero. Four models, three questions each, and not one of them mentioned the company that actually does that exact job. The models named five competitors instead.
A photo restoration company watched ChatGPT list eight tools for restoring old photos and not include them. They closed the tab without asking for the full report. I don't blame them.
A Portuguese company got asked its buyer question in Portuguese, scoped to Portugal, because that is where the domain lives. That one made me happy, it's a fix I shipped a couple of weeks ago after watching an Icelandic supermarket get compared against UK chains.
The pattern I keep seeing, and today was five for five: nobody is prepared for how completely absent they are. Not badly ranked. Absent. And every one of them had no way to know before they typed their domain in.
The launch numbers, since I'd rather post them than not: my email list outperformed Product Hunt. 17 emails to people who signed up months ago produced a real scan in 11 minutes. The launch produced one referral session. The list you build before launch day is worth more than launch day.
Someone also spent an hour on Reddit today writing a brutal teardown of my landing page and found in nine minutes that my hero copy tells visitors to go use ChatGPT instead of my product. That's tomorrow's job.
Free scan, no signup, at bersyn.com. Type your domain and see who gets named instead of you. Fair warning, it stings.
I have been scanning SaaS products across ChatGPT Claude Perplexity and Gemini for a few months now. Asking the actual buying questions people type in like "what is the best tool for X" and "compare X vs Y."
Most products are either completely invisible in AI buying conversations or described using their competitors features. And the founders have no idea.
Some patterns from scanning about 40 products:
- A product with 31K GitHub stars. ChatGPT has literally never heard of it. - A company that raised 16M. Zero mentions across all four models
- Multiple products described word for word using a competitors feature set.
The thing most people miss is that every model fails differently. Perplexity responds to new content within days.
ChatGPT uses training data from months ago. Treating them as one thing is why most efforts feel random.
I just opened this up so anyone can run one full scan for free. No credit card. Takes about 5 minutes.
Would love to hear what surprised you if you try it.
1 Comment