Testing, One, Two, Three: Is it OK to Use Headline Testing for Journalism?

clickbait rainbow

By Brian Ward

In a recent profile in The New YorkerEmerson Spartz, the CEO of Spartz Inc. and “King of Clickbait,” talked about some of the methods his company uses to make its content go viral. The one that caught my attention was called headline testing. Each time Spartz’s company produces new content, they send it out to users under a varitey of headlines. After a few hours a computer algorithm replaces the competing headlines with the one that’s received the most clicks.

As a journalist, how do I feel about that method of generating headlines?

Is it ok to test for success for the news?

Spartz doesn’t claim to be a journalist and I don’t begrudge him his success. Someone under 30 making money in 2015? Kudos!

However, he is at the extreme end of web content; he’s called the King of Clickbait for a very, very good reason. Most of his content is recycled from other sites. His algorithms parse through the net to find content that’s gaining momentum and then he spiffs it up and repackages it. His business model is based on if people click, not if they read. So his practices may not be the most conducive for journalism.

It’s not a secret that some news organizations are using A/B testing of stories; tweaking page length, link placement, and photos to find which ones get more readers. The issue with A/B and headline testing is what metrics one uses to determine success. Are you trying to find the version that will give you the most clicks, attention-minutes, or some other metric?

When your business model revolves around clicks, it means interesting headlines are more important than good content. It takes no stretch of the imagination to assume that the most popular headlines chosen by algorithms like Spartz’s are the most clickbait-y. Even Facebook has realized now that clicking on something doesn’t mean people are reading it.

For journalists it’s the opposite. Headlines are meant to drive people to your story so they’ll stick around and read.

Not all things are worthy of being clickbait.

Spartz only uses content that’s already popular, and he puts a new headline on it to make it even more popular. Journalists don’t always have the option of picking the most popular stories to write about. Anyone who’s worked at a small town paper can tell you it’s a lot of selectmen’s meetings, fundraisers, and new restaurant openings. Just because a story is important doesn’t mean it can’t be boring as well (C-SPAN anyone?).

In some cases you can only jazz up a headline so much before it becomes disingenuous. True, more people might click on my article about Antarctic ice core drilling if the headline was, “You’ll never guess what secrets scientists found drilling in the Antarctic ice!” However, most people would probably ex-out when they realize the answer is atmospheric data.

Headlines do matter.

Along with convincing people to read, headlines also affect how people perceive an article as they read it. Take the title of The New Yorker’s article on Spartz: “The Virologist: How a young entrepreneur built an empire by repackaging memes.” It’s clear in the title that the focus of the story is Spartz himself. If you click on the link thinking that he’s the focus, you are primed to look for the sections on Spartz himself.

But what if the title were “How to Make a Virus: How Spartz Inc. made millions with repackaged memes?” In that version the focus is on Spartz Inc. and methods they use to make money.

Headlines can also affect a person’s retention. Broad or misleading headlines can make it harder for people to recall details in a story. Take the Ebola scare. Last year, most experts agreed that the odds of an Ebola epidemic in the U.S were extremely low, citing good American infrastructure and Ebola’s low infection rate. You probably saw several articles that said this, yet they had headlines following the theme of, “EBOLA! WHAT YOU DON’T KNOW COULD KILL YOUR CHILDREN!”

Headlines of that nature probably got more clicks than, “Experts Agree U.S. Ebola Outbreak Unlikely.”

Headlines and ethics.

My fear is that headline testing will tempt news organizations to focus more on ad revenue than content. And to be less concerned about headlines that mislead readers or change what facts they take away from a story. There’s nothing wrong with coming up with a good headline you think will attract interest, if you are not being disingenous or intentionally misleading your readers. But the main focus should be on content and what your readers take away.

Is there a metric that measures responsibility? ∞

(Thanks to Meme Generator for the image.)

About Author

1 Comment

  1. Vilan Trub

    Interesting article, however, doesn’t clickbait create a culture of mistrust? The more saturated the market becomes with clickbait the stronger the desire will be for reliable sources. People will be conditioned to identify clickbait headlines.