People frequently take social media interactions as the de facto standard for “engagement” with a piece. The idea is that people who like, favorite, comment on, or share an article are likely to have consumed it entirely. While there’s some kernel of truth here, our data suggests that measuring audience attention and tallying shares are very different endeavors.

Last December, Chartbeat determined the top 20 most-read stories of 2015 — the pieces that held reader attention for the longest amount of Engaged Time. At the same time, Newswhip, a social data company, compiled the pieces that received the most Facebook interaction. Put them head-to-head and your jaw may just drop:

Clearly, the kind of content that people actually read is very different from what people interact with on social media. Measuring both of these behaviors is important, but if you’re looking to grow your audience, or improve site retention, or enhance viewability, well, the answer is simple. You’ve got to zoom in on producing engaging content — and then measuring it. It’s critical to measure what holds reader attention, not what appears to go viral.

And we see this kind of a discrepancy on Twitter as well. The graph below shows how completely people read an article (as measured by how far down the page they scrolled; all articles shown here were over 3000 pixels long), compared to how likely they were to tweet about it:

If the most engaging stories to read were the stories that were most likely to be shared, we’d expect this graph to look like a line. Instead, we see that there’s essentially no correlation between the two numbers. That doesn’t mean that social interactions are a bad way to measure engagement, but it does show that social engagement and on-site engagement are different phenomena, requiring different metrical approaches.

Eager for more data on social visitors? Don’t worry, the new Chartbeat Insider Guide has got you covered.