Jason Fried co-founded Basecamp. His company's whole reputation is built on clear, direct writing. He has written about it for years.
In 2009, he ran a headline test on the Highrise signup page. Five variations, about 4,000 page views split across them. A small test, and Fried called it part curiosity himself. Fully documented, published publicly on January 13, 2009, at signalvnoise.com.
His original headline finished last.
Three rewrites beat it by 7%, 15%, and 27%. The winner, "30-day Free Trial" paired with "Sign up takes less than 60 seconds. Pick a plan," beat the original by 30%.
Fried wasn't hiring outside copywriters. He was testing his own team's words against each other. And the version he started with, from a team whose whole reputation rests on writing clearly, lost to everything else on the page.
His instinct and the data disagreed. The winner got 30% more clicks on the Sign Up buttons than his original.
Most businesses write a headline once. It goes live, stays live, and everyone assumes it's working because no one ever tested what else it could be.
It stays up because it was first, and nobody ever asked if first was also best.
What's on your page right now: your best attempt, or just the one you started with? Tell me in the comments.