@ada.k
Bootstrapping a developer-tooling micro-SaaS since 2024. Obsessed with how ChatGPT decides what to recommend.
Three months tracking and Google AI Mode behaves differently from ChatGPT / Perplexity in ways that confuse my mental model: AI Mode pulls…
Four months tracking across all three engines for a devtool. The asymmetry is real and it's category specific. For my product a developer…
Posted this because I was confused why a smaller competitor was eating my AI citation lunch. Query: 'best CLI tool for managing local…
The plain-English body text working better than schema is counterintuitive but it makes sense when you think about it — LLMs are reading the page as text, not parsing structured data first. The structured data helps at the entity-recognition layer but the actual description the model forms comes from the body copy.
The FAQ schema thing is real. I made the same mistake — added 22 FAQ entries thinking more is better. Pruned to 4 very specific Q/A pairs that match exactly what buyers ask, and Perplexity started pulling them verbatim within 3 weeks. The 22-entry version was getting ignored entirely.
I made the exact same mistake. Worse actually — I had a WAF rule that blocked all crawlers matching 'GPT' in the user-agent string, which caught OAI-SearchBot too since the string appears there. Took me weeks to find it. Your curl diagnostic is the right first check.
Did you try asking ChatGPT directly where it learned about your brand? The 'where did you get this information' follow-up prompt sometimes surfaces 3-4 specific source clues. Helped me find the one industry directory that had the stale info and was the likely culprit.