Or maybe you will – I’ll let you decide.
It’s hard to imagine an online world without clickbait headlines, but we may soon have to.
While it’s hard to establish when exactly clickbait headlines began, the term itself was first popularised in 2006 giving the sensational, emotive and often humorous style of writing an official title.
This style of writing then gained momentum, due to sites such as Buzzfeed – and with good reason. For a while, these types of headlines were good at driving website traffic and encouraged social sharing.
However, it seems the general public has begun to tire of the ‘over promise, under deliver’ nature of these headlines.
Research carried out by Penn State’s Media Effects Laboratory and Institute of Computational and Data Sciences has found that Clickbait headlines (defined as relying on ‘linguistic gimmicks’ such as superlatives and curiosity gaps) are not any more effective in driving website traffic than traditional headlines.
In fact, it found that the Clickbait headlines actually performed worse on occasion.
The researchers carried out two studies, and each time they presented a group of 150-250 subjects one of eight headlines, featuring both Clickbait style writing and more traditional headlines.
The headlines presented to the subjects were classed as Clickbait if they included the following traits:
- Lists
- Positive Superlatives (eg Best, Greatest)
- Negative Superlatives (eg Worst, Least)
- Modals (eg Could, Should)
- Demonstrative adjectives (eg This, That)
- ‘Wh’ Words (eg What, When, Who)
They then monitored to see if the subjects would follow through and read the article or share it on social media.
The study concluded that Clickbait headlines “did not dramatically outperform” headlines written in the traditional style. Whilst they could only speculate on the reason, they suggested that because are so common these days, they no longer stand out as intriguing.
Interestingly, for the first study, the researchers used AI to identify which headlines were traditional and which were clickbait headlines. This AI was similar to the type used by social media platforms.
They found that the AI systems frequently disagreed on what was classed as a Clickbait headline – in fact, they only agreed 47% of the time
This may not seem a huge deal if you’re not planning to research Clickbait with AI, however, it does pose a problem for social media platforms that are trying to clamp down on the spread of misinformation & fake news.
It is widely believed that due to its exaggerated nature, and social sharing potential that Clickbait headlines can help misinformation spread – especially as the topics tend to be trending and easily skimmable.
Therefore, many a social media platform has assumed that by reducing the reach of these articles that are shared, the less misinformation will circulate on their platform.
As such, they were relying on AI to scan content and accurately determine which content was trustworthy and which were clickbait. If the AI struggles to do so, it could see some Clickbait titles slip through the net and feature on more people’s newsfeeds.
It also opens the door for Clickbait to evolve. If, for example, Clickbait writers notice that their stories get more reach when including questions rather than lists it is only logical that the number of Clickbait headlines including lists will reduce, whilst the number including questions will increase.
In turn, this will mean that the AI itself has to evolve to match, eventually leading to a cat and mouse style contest.
So what does this all mean for us?
Well two things, firstly if Clickbait headlines aren’t your style you don’t need to worry. Writing in a traditional style won’t hinder your website’s traffic & may even improve it.
It also means that websites that intend to spread misinformation, may have to evolve constantly, to keep ahead of the AI trying to stop them. This may make Clickbait headlines harder for users to identify, making it even hard for digital novices to identify a trusty source