Comment by oneeyedpigeon

I disagree. You could A/B test two good, accurate, well-written headlines and stay clear of clickbait altogether. Sure, you're still optimising for the most popular, but "clickbait" doesn't just mean "well performing", there's also an implication of duplicity.

I have a modicum of experience here. I write for another online media company and, although we produce our own headlines, we are 'strongly encouraged' to write clickbait headlines, to the extent where we are asked to remove instances of specific product names (etc.) in order to be mysterious and not give the game away too early. (Yes, in case it wasn't clear, I hate this!)