Back in 2016, I got a job writing a bunch of political fundraising advertorials. I was helping raise funds for Hillary Clinton, for Ted Cruz, and for Donald Trump.
(I found out later that the guy who had hired me was a big-time scammer. Almost none of the money he raised was ever used for any political purpose. But it was used for the purpose of coke and hookers in Las Vegas. Which you might think is a more noble goal than furthering the political careers of any of the above faces.)
Anyways, as part of this job, I had to constantly read a bunch of news articles for research. Politico… Fox… WSJ… and of course The New York Times.
It was then that I developed my contempt for The New York Times.
I guess I still had higher expectations of the NYT. Fox News was clearly an inflammatory tabloid, but the NYT still sold itself as classy and respectable and trustworthy.
But that’s not what I saw. Not when I read each story carefully, compared it with the headline, and then compared the whole with the same story covered in other media.
Whatever. I only bring up this episode from my life because I just came across a fascinating article and a resource on the exact same topic. The article and the resource can be useful to you whether or not you support the New York Times point of view.
So:
A guy named Tom Cleveland wanted to see exactly how the NYT A/B tests its headlines. He wrote up a script to pull in the data from the NYT site, and he started looking for insights. You can head over to Tom’s Substack if you’re interested in the full story. But here’s one quick tidbit, which should be old hat if you’re interested in copywriting:
NYT headlines tend to get more dramatic through A/B testing.
Tom gives a few examples. For example, a recent story about Meghan Markle started its career with the headline:
“Saying her life was less than a fairy tale, Meghan Markle described the cruel loss of her freedom and identity”
Come on Meghan. Every angsty teenager complains of loss of freedom and identity. Sure enough, the editors at the NYT tested ways to raise the stakes. The eventual winner:
“‘I just didn’t want to be alive anymore’: Meghan Markle says life as a royal made her suicidal”
A second example, this about Trump:
“Trump, addressing conservatives, plans to claim leadership of GOP”
Trump, ok. Always a solid way to get engagement. But “addressing conservatives, plans to claim leadership”? It sounds like the overture of a long and boring opera. Compare it to the winning “Tarantino-ized” headline:
“Trump’s Republican hit list at CPAC is a warning shot to his party”
Like I said, this will be old hat to you if you’ve been writing sales copy for a while. But it’s still interesting to see when backed up with the massive data behind the New York Times… and when dealing with the supposedly sophisticated and intellectual readers of the Times.
There’s much more to Tom’s data, including stuff that’s both obvious and not so obvious for copywriters. He goes into more detail about it on his blog. But he has also made all his data available online, in real time, in a very easy-to-use format. If you’d like to see it: