Interesting angle. I’ve found those human vs bot scores are useful as a first pass, but sometimes they misclassify power users or weird edge cases. When I buil

Interesting angle. I’ve found those human vs bot scores are useful as a first pass, but sometimes they misclassify power users or weird edge cases. When I built UXWizz[0], session recordings ended up being the clearest signal of real intent. Watching someone actually struggle through a form tells you way more than a score alone. Have you thought about feeding behavior patterns back into the quality score? Feels like there’s a nice combo there. PS: The linked page is quite cool, reminds me of 2005 internet and Piczo websites, lol [0] https://www.uxwizz.com https://www.uxwizz.com