Post

How AI Agents Use Websites, Where They Fail, and What to Fix - via @giacomozecchini.bsky.social #seo

merj.comHow AI Agents Use Websites, Where They Fail, and What to FixAgent interaction monitoring shows why browser-based AI agents fail on real sites, and how accessibility engineering and task-level tests reveal breakpoints.

14/10/26, 9:48 am1 repost1 like

Matt Janaway

@mattjanaway.co.uk

Useful research. Thanks, @giacomozecchini.bsky.social. Full-screen consent and delayed newsletter pop-ups tripped up agents in MERJ's tests. eCom teams can quickly test these by running their checkout journey through an agent to see where it gets stuck.

Giacomo

@giacomozecchini.bsky.social

Thank you so much for sharing it!