I worked on a content funnel: articles → lead form → submission. The client said: “90% of users read the articles to the end — let’s scale this format!”
You tell me the next step?)
In GA4, the scroll event triggered at 90% in most sessions. But conversions were low. Users were “reading” — yet not converting. Something felt off.
We dug in.
GA4’s scroll event is an auto-trigger that:
• Fires when a scroll depth is reached, not when content is read
• Can trigger from fast or accidental scrolling
• Doesn’t care how long users stayed on the screen
• Can activate during background rendering or auto-scroll
• Doesn’t track actual attention or intent
So: a 90% scroll doesn’t mean deep reading — just that the user got there, even by accident.
Because of this:
• The client assumed the content was engaging (No, it's not)
• Created more articles in the same “successful” format
• Scaled campaigns around content that didn’t convert
• Burned budget on a false signal
Yes, users reached the end of the page. But they didn’t read. Scroll was just noise.
GA4 worked as configured — but interpretation failed.
Now, I treat scroll events as signals, not proof. I check:
• How much time users spent after the scroll
• Whether they interacted — e.g., clicked, copied text, opened a CTA
• I add custom events like:
- “15+ seconds in content area”
- “5+ seconds at page bottom”
- “CTA block viewed”
• And I always compare scroll behavior to actual conversions
Scroll = movement. Reading = behavior. If you confuse the two, you’ll optimize for motion — not meaning.
Want to get all my top Linkedin content? I regularly upload it to one Notion doc.
Go here to download it for FREE.
