40 people watched my landing page live. 1 understood what my app does.

I'm building Petipa, a planner that lives inside ChatGPT instead of being another separate app. ChatGPT makes the plan. Petipa helps you follow it. That's the whole idea. The plan survives between chats, it shows you one small next step instead of a wall of text, and you can pick it back up after a missed day without a streak shaming you. On October 20, my landing page got reviewed live in front of about 40 people. Scores: design 3.5/5, UX 2.5/5, trust 3.5/5. They thought it was a prompts website for ADHD. After I explained it's a planning app inside ChatGPT, someone ran a poll. 1 out of 9 had understood. One. The feedback stung because it was fair. Too much text for an ADHD audience. No immediate onboarding. And my CTAs contradicted each other, so nobody could tell if it was live or a waitlist. No real users, no team, no email capture either. Trust was 3.5 for a reason. The plugin is now listed on ChatGPT, so the "is it live?" part is finally settled. The comprehension problem isn't. My question is simple. If you land on a page and you don't get it in 5 seconds, what do you cut first: the text, the demo, or the whole hero? Brutal answers welcome, that's why I'm posting here. If you want to see the page, it's linked on my profile.