We Query ChatGPT, Claude, and Perplexity About Your Brand. Here's the Architecture

A few months ago I had a question that didn't let me sleep. When people ask ChatGPT "what's the best tool for X", which brands actually get mentioned? And more importantly: does my brand get mentioned? Turns out this is a surprisingly hard thing to measure. So we built a product for it. It's called Be Recommended , and under the hood it's a system that parallel-queries ChatGPT, Claude, Perplexity, and Gemini, then scores how visible a brand is inside their answers. This post is about the architecture. And about the data we found, which frankly shocked us. The problem in one line Traditional SEO tells you how you rank on Google. But when someone opens ChatGPT and asks "what's the best CRM for a solo consultant", Google isn't in the room. The LLM answers directly, from its training data plus whatever live retrieval it does. If your brand isn't in that answer, you lose the customer before you even know they existed. The Gartner forecast everyone quotes says roughly a quarter of search vol