Why You Don’t Want Related Content

by John Logioco

In print, the act of browsing a magazine is second only to reading the actual content. Confronting a new, unrelated story at the conclusion of another… this has been the primary way audiences interact with their favorite publishers for a long time.

To the extent online publishing began as an extension of print mentality (for better or worse), this element — the element of unpredictability — was missing from the start.

It’s not that audiences lacked the means to navigate to the next piece of content. It’s just that the options were limited and, let’s face it, kind of boring.

The easiest thing for early, custom recommendation engines to do was simply surface a story related to the one you just finished. Let’s say you just read a blog post on healthy Halloween candy alternatives. You might find a set of a options that looked like this at the conclusion of the post.

“Related” as a proxy for “relevant” content may seem like a safe, educated guess on what audiences want to read next, but is relevance really the most desired outcome? How often do we want to experience a story that’s closely connected to the one we just finished?

The average Facebook feed or Twitter stream is a jumble of content that’s tenuously connected at best. Does that make them less engaging? We know in fact the opposite is true (and so do the algorithms). The unpredictability is what keeps us interested, keeps up scrolling the feed.

Similarly, Outbrain shows that unpredictability is what keeps audiences interested on publisher sites.

For content creators, this theory has far-reaching implications. As whole companies and industries spring up to try and solve the mystery of what makes content inherently social, what’s often overlooked is what makes content interesting to audiences. One doesn’t necessarily lead to the other.

Depending on who you ask, the correlation between what we share and what we read is ambiguous, but before we do either, our curiosity has to be piqued first. According to our data, the more unexpected the encounter with the content, the better. And it starts with headlines and images — the “front door” of content.