Most "great copywriting" you see online is confidence with no evidence. Someone tweets "the best ad I've ever seen." But who measured it? Compared to what? Did it actually convert?
There are exactly two forms of proof in copy:
Everything else is taste.
So how do you tell the difference in 30 seconds without becoming a research librarian? The framework below.
From the founder of Marketing Examples — voted the #1 marketing newsletter, 130k subscribers. For every sentence of copy, ask:
Close your eyes. If you can't see it, the copy is abstract. Abstract copy doesn't get remembered. Abstract copy doesn't get felt. If you can't drop it on your foot, it isn't real.
"Couch to 5K" is the most-downloaded fitness program of all time. Made by a Boston TV producer — not Nike, not Adidas. The name was the entire pitch.
A sentence that's true-or-false makes the reader sit up. Vague claims slide past. Galileo got 10 years house arrest for "the earth spins around the sun." If he had said "the earth has a harmonious connection with a celestial object," they'd have told him to go to the pub.
Setting up a friend on a date? "He's smart, hilarious, kind" → ignored. "Reads on the tube. 6'2". Looks like Ryan Gosling." → believable.
Don't talk. Point.
If your competitor could sign the same line, throw it out. Real differentiation forces specificity.
Your car has five numbers on the speedometer. Volvo has six. One could get the impression that the people who made your car lack a little confidence. — Volkswagen on Volvo. Visual ✓ Falsifiable ✓ Uniquely theirs ✓
Three yeses → you're onto something.
Three nos → you've written rubbish.
Sometimes you know more in 2 seconds than you know in 2 hours. — Harry Dry
Show your copy to someone. Count: "One Mississippi, two Mississippi." Do they get it? If yes — it works. If no — it doesn't matter how clever it is.
This is why Athletic Greens' homepage works: "You're going to need a smaller cabinet." Two seconds. You instantly understand the positioning (replaces all your morning vitamins).
If you want to study copy that's been validated by markets — not by personalities — start with these five. Each ran as a control for years. Each made millions.
| Letter | Author | Validation |
|---|---|---|
| "Tale of Two Young Men" | Martin Conroy (WSJ) | $2 BILLION over 28 years |
| Coat of Arms Letter | Gary Halbert | 600M mailed, $300k/day, 30+ years |
| "They Laughed When I Sat Down at the Piano" | John Caples | Ran for decades, most-copied headline ever |
| "Lazy Man's Way to Riches" | Joe Karbo | Tens of millions in sales, 1973+ |
| "Read 300 Magazines in 30 Min" | Eugene Schwartz (Boardroom) | Multi-decade control |
You can study every one of them on swiped.co. They're tagged "Classic" or "Control" — meaning real working copywriters reverse-engineered them and confirmed the runtime.
There are exactly two questions that matter when evaluating any piece of copy:
Everything else is opinion.
When in doubt: default to multi-decade controls. They were validated by markets that paid real money for years — not by tweets.
The more specific your proof, the more believable your claims. — Gary Bencivenga (the world's greatest copywriter)
24 files. 150 references. 27 pieces of copy with documented proof. 30 drills. 32 templates. The full learning path from Beginner to Expert — plus an Action Hub that routes by intent ("I need a cold email in 5 min" → exact file + template + example).
Browse the Vault →