Misleading headlines & clickbait

How it works

A headline has one job in a feed: stop your thumb. The article has a different job: be defensible. When those two jobs pull apart, you get “Minister wants to ban cash” above an article in which a minister mentioned a report that discussed limits on cash payments in one sector. Every word of the article may be true. The headline is where the lie lives.

It matters because most people never see the article. In 2016, researchers at Columbia University and INRIA tracked links shared on Twitter and found that about 59% of them were never clicked, not by anyone. People shared the headline. For most readers a headline is not a summary of the news. It is the news.

The classic moves are easy to list. The question headline (“Is the new EU rule about to double your bills?”) makes a claim without making it, and the answer, per Betteridge’s law, is usually no. The verb upgrade: “suggests” becomes “shows”, “shows” becomes “proves”, “may be linked to” becomes “causes”. The violence verbs (SLAMS, DESTROYS, HUMILIATES) for what was often a polite disagreement. And pure clickbait, “you won’t believe what happened next”, which at least is honest about having nothing to say.

Research by Ullrich Ecker and colleagues found that a misleading headline shapes what people remember even when they go on to read the whole, accurate article. The first frame sticks.

Example

@daily_signal_hub, 5h STUDY PROVES: new 30 km/h zones INCREASE pollution. The green lobby’s plan backfires again.

@m_kovacs_reads, 4h I read it. One simulation study, in one city, says emissions might rise on two streets if drivers reroute. The authors say more data is needed. “Proves” is doing a lot of work here.

@civic_watch_eu, 3h Is the government hiding the real inflation numbers?

@m_kovacs_reads, 2h The article under that headline: a statistician says the methodology changed in 2019, and the change is published on the office’s own website. So… no?

Notice the pattern: twenty thousand shares on the headline, and the one person who read the article is at the bottom of the thread.

Why it works on you

• The headline is the only part you see. Feeds show the headline, the image and the share count. Reading the article costs a click, a page load and ninety seconds you had not planned to spend, so your brain quite reasonably judges on what is in front of it.
• Emotion moves before reading does. Outrage, fear and “I knew it” arrive in under a second, and sharing feels like finishing the thought. See emotional hijacking and rage bait.
• It fits what you already believe. A headline that confirms your view never gets the “wait, really?” check. See confirmation bias.

What to do

When someone shares it with you: open the article. Find the sentence that supposedly backs the headline. Very often it is softer, or it is one person’s opinion, or it is not there at all. Reply with that sentence, not with a verdict.

When you catch yourself: the rule is simple and annoying: do not share what you have not read. If the headline made you feel something strong, that is the exact moment to click rather than to share. Question headline? Assume the answer is no until the article says otherwise.

In a comment section: quote the article against its own headline. “Paragraph four says ‘may’, not ‘will’” is hard to argue with and easy to copy. If the text contradicts the headline, say so without calling anyone a liar: the person who shared it probably never saw the text either.

Don’t confuse it with

• A short headline that simplifies is normal. Headlines cannot carry caveats; the test is whether the article’s main finding survives, not whether every nuance did.
• A strong headline on a strong story is just journalism. The problem is the gap, not the volume.