Sample frameDay 277·Marketing · Copy

Nobody reads your advertising. They read one line, then decide whether the rest exists.

Marketing.

The Headline Test

5 min read·Apply by lunch

The question

Of the last five things you sent out with a line at the top — the subject line, the ad, the page header — how many of those lines ever ran against a rival version in front of real buyers? If the answer is none, you didn't pick a headline. You picked the one that sounded best in a room full of people who already knew what you meant.

The idea

John Caples wrote one of the most copied headlines in advertising history in the 1920s, at Ruthrauff & Ryan, for a mail-order piano course: 'They laughed when I sat down at the piano.' The line isn't why he's worth reading. He's worth reading because he could prove which lines worked. Mail-order advertising carried keyed coupons, so two versions of the same advertisement could run in the same publication in the same week and the replies came back sorted by version. Caples ran that test for years, first at Ruthrauff & Ryan and then across a long career at BBDO, and set the method down in Tested Advertising Methods in 1932. What it kept showing him was that changing the headline alone, with the body copy left untouched, moved response by multiples rather than by percentages. David Ogilvy, an open admirer, put a number on it: write the headline and you've spent eighty cents of your dollar. That's a rule of thumb rather than a measurement, but nobody who has run the test argues with the direction. Caples also sorted what worked — lines that offer the reader something they want, and lines that carry genuine news, beat lines that merely tease. The trap is treating the headline as the finishing touch, written last, in the ten minutes before the thing goes out.

The framework

01
A headline does two jobs, and most do only one.
It selects — telling the right reader this is for them and letting everyone else leave — and it promises, giving that reader a reason to spend the next thirty seconds. Lines that select without promising deliver the right audience and no interest. Lines that promise without selecting pull a crowd who never wanted the thing. Hold yours against both jobs before it goes anywhere: cover the rest of the page, read the line cold, and ask who it excludes and what it offers. If you can't answer both, the line isn't finished.
02
Write twenty before you judge one.
The first headline you write is nearly always a description of the product. The reader only turns up somewhere down the list. Caples' advice was to produce them in quantity, and the reason is mechanical rather than mystical — you cannot sort a list of one. Draft twenty in a single sitting, deliberately including some you find obvious, some you find faintly embarrassing, and a couple that state the plainest benefit in the flattest possible words. Plain lines win tests far more often than the people who wrote them expect.
03
What would you back to beat it?
A test needs a real rival, not a variation on a theme: two lines built on different appeals — one on the benefit, one on the news, one on the cost of doing nothing — rather than the same claim with a different verb. Run them in the same slot, to the same audience, in the same week, changing nothing but that line. Give it long enough to gather a number of responses you would actually act on. And decide beforehand what result would make you switch, because a test you argue with afterwards was never a test.
04
The specific beats the clever, almost every time.
Cleverness asks the reader to do work; specifics do the work for them. A number, a named group, a time period or a plainly stated outcome consistently outperforms wordplay, because the reader is scanning rather than reading and gives you one glance to be relevant. It is also why curiosity on its own is a weak appeal: it earns the click and then disappoints, which costs you the reader twice. Keep the pun if it survives the test. Keep the plain line when it wins — and it usually wins.

Members only · 4 principles + template + AI mentor

364 more frameworks are waiting.

The Headline Test is Day 1 of 365. One framework every morning for a year — across strategy, sales, negotiation, leadership, and more.

$1 /day

Billed $365/yr · cancel any time

365 frameworks

12 topics · templates · AI mentor

Reserve your place — $365/yr →

One payment a year · the whole library · no auto-renewal.

More Marketing frameworks

All 28 marketing frameworks →