
Strategy & Craft
Splitting an ad budget: the 70-20-10 rule
Seventy per cent on what works, twenty on scaling, ten on experiments. The rule is old, simple, and violated in almost every account.
Strategy & Craft
Four out of five people read only the headline. Which makes it four fifths of the work and usually the last thing written.

The headline decides whether the rest exists. It is worth more attention than any other element and usually gets the least.
Write twenty before choosing. The first five are the obvious ones everybody writes. Six to twelve are variations. Thirteen to twenty are where something unexpected appears.
It takes forty minutes and it is the highest-return forty minutes in the project.
| Pattern | Why it fails |
|---|---|
| Puns and wordplay | costs a beat of comprehension |
| The company name alone | means nothing to a stranger |
| Welcome, Hello, About us | says nothing |
| Abstract nouns: innovation, solutions | no picture forms |
| Questions with an obvious no | reader disqualifies themselves |
The last one deserves care. "Want to save money?" invites a shrug. "Paying more than £2,400 a year to heat a semi?" invites recognition.
In digital advertising the constraint is display, not attention. A search headline is cut at roughly 30 characters, a social one at around 40 on a phone. Put the load-bearing words at the front.
In print, longer headlines are fine and often better, because the reader has already committed to the page.
Run two headlines against each other in a search campaign for two weeks. It is the cheapest headline research available, and the results transfer to print, letters and the website.
Test one variable at a time. Two headlines differing in three ways teach nothing.
Where format allows, a subhead below the headline adds the qualification the headline was too short to carry: the price, the area covered, the condition. That combination lets the headline stay sharp without becoming inaccurate.
Six to twelve words in print, three to eight for a digital ad. Longer works if every word carries weight.
Only if it is known. An unknown name in the headline spends the reader's only glance on information they cannot use.
More from Strategy & Craft

Strategy & Craft
Seventy per cent on what works, twenty on scaling, ten on experiments. The rule is old, simple, and violated in almost every account.

Strategy & Craft
Percentage-of-revenue rules are widely quoted and mostly wrong. The answer comes from your own numbers in four steps.

Strategy & Craft
A bad brief produces bad work at full price, and the client pays twice. Eight questions, two pages, three hours of your time.