A common reaction to clickbait is moral — that publishers are being lazy, dishonest, or cynical. That framing is mostly wrong, or at least incomplete. Clickbait is rational behaviour given the economics newsrooms are operating in. To stop being surprised by it, you have to understand the math underneath.
Revenue per page view is small. Very small.
A typical news site earns somewhere between $1 and $15 per thousand page views from programmatic display ads. That number, the CPM, depends on the audience’s geography, the topic of the page, and the time of year. For most general-interest content, it lands closer to the bottom of that range.
That means the gap between a headline that gets 50,000 clicks and one that gets 500,000 is roughly the difference between a few hundred dollars and a few thousand. Multiply that across every article a publisher writes in a year and the click-through rate on headlines becomes the single biggest lever on revenue.
Headlines are A/B tested
Most large publishers run experiments on their own headlines. The CMS shows different titles to different segments of visitors, measures which version drives more clicks, and promotes the winner. Some sites do this manually with two or three variants; others have automated systems that test dozens.
What the experiments consistently find is that vague headlines outperform specific ones. “Apple announces new iPhone with USB-C” loses to “Apple just changed everything about the iPhone.” The first headline answers the question, so the reader doesn’t need to click. The second creates the open loop the curiosity gap exploits.
Importantly, no individual editor decided that vagueness was the strategy. The strategy emerged from the data. The system selects for it.
The aggregator effect
Even if a publisher genuinely wanted to write clearer headlines, they’re competing for placement on Apple News, Google News, Facebook, Flipboard, and a dozen aggregators that rank stories largely on engagement. A clearer headline that gets fewer clicks at the source gets less distribution downstream. The publisher who plays it straight ends up smaller than the publisher who optimises — and over a few years that gap compounds out of recoverable range.
This is why clickbait is sticky. It’s not just that individual publishers benefit from it; it’s that abstaining from it is competitively dangerous. The first publisher to write a worse-performing headline loses traffic, share, and ad revenue, and that loss eventually shows up in the newsroom budget.
Subscription publishers behave differently
The exception to all of this is publishers who make most of their money from subscriptions rather than ads. The New York Times, the Financial Times, The Atlantic, and a handful of others optimise for conversion to paid subscriber, not for raw clicks. Their headlines tend to be more specific because their funnel rewards specificity — a reader who knows what they’re getting is more likely to value the relationship and eventually subscribe.
If you find yourself frustrated by vague headlines on free sites, this is part of why paid journalism remains valuable. The incentives are pointing in different directions.
Where this leaves the reader
If clickbait is the predictable output of a system, the right response isn’t outrage — it’s practice. Treat every vague headline as a small puzzle. What is this most likely to be about? Then check, but only if you actually care. The fewer clicks the vague headlines get from you, the less you reinforce the pattern in the recommendation systems that surface them to you next time.
The game we’re building rewards this skill directly. The more often you can guess what the article says without clicking, the higher you score — and the better calibrated your sense of which headlines are worth the click in real life.
For Clicks Sake is a game built on real clickbait. The faster you can guess what the article actually says, the more you score.
Play a Quick Game →