A financial adviser writes a genuinely useful article about preparing for retirement. She titles it “Some Thoughts on Retirement Planning” and sends it to her email list. A few people open it. Most don’t.
A month later, she sends almost the same article with a new headline: “Three Retirement Mistakes That Quietly Cost South Africans Thousands”. Opens nearly double. Several readers reply asking for an appointment.
The content hadn’t changed. Only the headline had. For any business that writes emails, blog posts, social captions or newsletters, the headline often decides whether the rest is ever read. A remarkable study of more than 100,000 real headline tests offers some useful, and slightly uncomfortable, insights into what makes people click.
One Like insight: Your headline doesn’t just introduce your content. It decides whether anyone reads it.
What 105,000 headline tests revealed
Upworthy, a viral news website that peaked in the mid-2010s, routinely tested several headlines for each story, showing different versions to different visitors and keeping the one that attracted most clicks. Its archive of these experiments was later made available to researchers.
In 2023, a team led by Claire Robertson of New York University and Stefan Feuerriegel of LMU Munich analysed over 105,000 headline variations from that archive, which together had generated more than 5.7 million clicks, and published the results in Nature Human Behaviour. Because each test compared different headlines for the same story, the researchers could isolate the effect of the wording itself.
They found that negative words increased clicks. For a headline of average length, each additional negative word increased the click-through rate by 2.3%. Each additional positive word reduced it by about 1%. Interestingly, words linked to sadness drove clicks more than those linked to anger or fear.
That sits alongside the research in Why Some Posts Get Shared and Others Disappear, which found positive, awe-inspiring content was shared more. Clicking and sharing seem to work differently: we click on things that worry us, but share things that make us look good.
The data came from one news site with a particular audience and style, and the effects per word are modest. But across many readers, small differences add up.
Better headlines, honestly
For a small business, the lesson isn’t to fill every headline with doom. Misleading or sensational headlines might win a click, but they damage trust when the content doesn’t deliver. The useful insight is that people are drawn to problems they want to avoid. You can use that honestly.
Frame around a real problem. “Three mistakes that make your geyser fail early” is both more compelling and more useful than “Geyser maintenance tips”, as long as the article genuinely helps readers avoid those mistakes.
Be specific. Numbers, names and concrete outcomes tell readers exactly what they’ll get: “How one Durban bakery cut its electricity bill by a third” rather than “Saving energy in your business”.
Keep it simple. The Upworthy researchers also found that more complex wording tended to reduce clicks. Clear, plain words work best, as we explored in Plain Language Isn’t Dumbing Down.
Deliver on the promise. The headline is a promise; the content must keep it. If readers feel tricked, they’ll stop opening your emails.
Write several, then test. Upworthy’s editors reportedly wrote many headline options for every story. Write five or ten, pick the best two and, if your email or social tools allow it, test them against each other.
[Infographic placement] Same article, different headline. Two email subject lines. Bland: “Some thoughts on retirement planning.” Specific and problem-focused: “Three retirement mistakes that quietly cost South Africans thousands.” Beneath: +2.3% clicks per negative word; −1% per positive word. Caption: Robertson et al. (2023), Nature Human Behaviour, 105,000+ Upworthy headline tests. Use honestly: the content must deliver.
A headline habit
Before sending your next email or publishing your next post, spend five minutes on the headline alone. Write at least five versions. Ask of each: does it name a specific problem or outcome the reader cares about? Is it clear and simple? Does the content genuinely deliver what it promises? Pick the strongest.
Then learn from your own results. Most email and social tools show open or click rates. Over a few months, note which headlines performed best and look for patterns. Your own audience is the best guide to what works for your business.
A few questions can help:
- Does our headline name a problem or outcome our readers care about?
- Is it specific enough that readers know exactly what they’ll get?
- Would a reader feel the content kept the headline’s promise?
- Have we written several options before choosing one?
Which brings us back to the financial adviser and her retirement article. Her original headline was polite and accurate, but gave readers no reason to open it. The new one named a specific, honest worry and promised a useful answer, which the article then delivered. That combination of a compelling headline and content that keeps its promise is how good headlines build trust rather than spending it.
Sources
- Claire E. Robertson, Nicolas Pröllochs, Kaoru Schwarzenegger, Philip Pärnamets, Jay J. Van Bavel and Stefan Feuerriegel, “Negativity drives online news consumption”, Nature Human Behaviour 7(5), 2023. Summarised by Nieman Lab.
