Hey, Mark here.

Last Thursday I asked Claude for campaign concepts for a home-security client.

Twenty of them, big swings, go.

It gave me twenty. Clever ones. A couple made me smile.

Not one of them was about the product. They were about fear, and family, and peace of mind, the same four notes every security ad has hit since the first alarm company bought its first billboard.

Twenty concepts, and every one of them could have run for any competitor in the category without changing a word.

That's when I closed the laptop and pulled Ogilvy off the shelf.

Because he wrote the book on exactly this failure, and he did it before anyone had a laptop to close.

David Ogilvy built the most respected agency of his era, and the ad people still study most is a 1958 spread for Rolls-Royce. You know the headline even if you don't know you know it.

"At 60 miles an hour the loudest noise in this new Rolls-Royce comes from the electric clock."

Ogilvy didn't dream that line up, and that's the part the swipe files leave out. He spent three weeks, twenty-one days, reading everything he could find about the Silver Cloud.

The line about the clock was a fact he surfaced from a technical write-up on the car.

His job was to read until the product handed him the headline. Then he got out of its way.

The ad roughly doubled Rolls-Royce sales in the U.S. And the man who wrote it always told the story as a research win, not a creative one.

That wasn't a one-off. Before Ogilvy wrote a line of copy for anyone, he'd run audience research for George Gallup. He came into advertising already believing the answer lived in the reading, not in his own head.

His line for it:

Advertising people who ignore research are as dangerous as generals who ignore intercepted signals. He meant it literally.

The facts are the intelligence. You don't go in without them.

The research was half of it. The other half was a standard.

Ogilvy's test for any campaign was whether it carried a big idea. His words: "It takes a big idea to attract the attention of consumers and get them to buy your product. Unless your advertising contains a big idea, it will pass like a ship in the night."

He was blunt about how rare that was. Most advertising, his own included on his off days, never found one.

And the big idea, for him, was never decoration. "If it doesn't sell, it isn't creative."

The clever concept that moves no product wasn't a near miss to Ogilvy. It was a failure with better art direction.

Put those two beliefs together and you get his whole method. Read the product until it gives you something true and specific and a little surprising. Then judge every concept against a single bar: is there a real idea here, or is it just well-dressed.

The electric clock cleared the bar because it was concrete, provable, and it dramatized the one thing a luxury buyer wanted to believe. My twenty security concepts cleared nothing, because they started from a mood instead of a fact.

He proved the same thing with a prop. The Hathaway shirt campaign that put his agency on the map ran on a man in an eyepatch, a detail that cost about a dollar and gave the ad a story the reader's eye couldn't leave alone.

No new budget, no effects. One concrete choice, and the shirts sold out.

The idea lived in the detail, the way it always did with him.

Under all of it sits a belief about the reader that's easy to quote and hard to practice. "The consumer isn't a moron," he wrote. "She is your wife."

A specific, provable fact treats her like someone who can tell a real claim from a nice feeling. Ogilvy bet his whole method on the idea that she could.

Run his method forward sixty years and drop Claude into it, and you get a strange split.

Ogilvy would have loved the reading. He spent twenty-one days on one car because that was how long the reading took a human being. Claude reads twenty-one days of material in about a minute.

The single most expensive part of his process, the part that made the Rolls-Royce ad possible, is now nearly free. He would have been the first person in the building to see what that was worth.

He'd have seen the second thing faster than most of us did, too. He could only ever read one product deeply. Claude can read the whole category in the same minute, every competitor's page, every one-star review, every spec that doesn't match the marketing.

Ogilvy found one electric clock by reading one car. The same method now runs across a whole shelf, which means the fact you build on can be one no competitor bothered to find.

Then he'd have watched how most people actually use it, and he'd have groaned.

Because we skip the reading. We hand Claude a product name and a vibe and ask it to be clever, which is the exact move Ogilvy spent his career arguing against.

A model given nothing true about the product does the only thing it can. It reaches for the category's stock emotions and writes the ad everyone else already wrote.

My twenty concepts weren't a Claude failure. They were a briefing failure. I gave it fear and family, so it gave me fear and family back.

The electric-clock line could never come from a model that never read the report. The fact wasn't in the prompt, so it couldn't be in the copy.

That is the whole lesson, and it hasn't changed since 1958. The copy can only be as specific as what you put in front of the writer, and now the writer can hold more than any human ever could.

Before you ask Claude for a single concept, you make it read. This is the closest thing I've found to running Ogilvy's twenty-one days in an afternoon.

Feed it everything real about the offer:

Then you don't ask for ideas yet. You ask for facts.

That last instruction is the guardrail. Left alone, Claude will happily jump to the clever angle, because the clever angle is easier than the reading. You're forcing it to stay in Ogilvy's twenty-one days a little longer, in the part where the facts are still just facts.

What comes back is a menu of real things. Most are dull. One or two, if the source material had anything in it, will make you sit up the way the clock line did.

That's your electric clock. Now you have something to build on that a competitor can't run without lying.

A fact is raw material. The idea is the promise that fact lets you make with a straight face.

Under a minute to dispatch is a fact. "Faster than the fire alarm you already trust" is the idea it earns.

Which is where my security client comes back in. I ran the fact hunt on their actual operations notes instead of my mood, and one line came up that I'd walked past a dozen times.

Their monitoring center dispatches in under a minute. The response time everyone in the category quotes, the scary one on the billboards, is the police, not the alarm.

That gap was the concept, a number where I'd been reaching for a mood. Twenty clever swings never found it, because it was never in the vibe. It was in the file.

The fact hunt gives you raw material. The second move is the bar.

Once you've built concepts, from the facts, run them back through Ogilvy's standard instead of your own affection for them.

The "be willing to tell me none of them do" line matters more than it looks. The whole point of a bar is that things fail to clear it. If every concept passes, you built a bar out of nothing.

When it comes back and says none of them clear the bar, the instinct is to ask for twenty more concepts. Resist it. More concepts off the same thin facts just gives you more clichés.

The problem is upstream, in the reading. Go back to the fact hunt, feed it more real material, and find a better thing to stand on.

You can't filter your way to a big idea you never dug up.

Ogilvy did brand and direct response both, and he never pretended they were the same job. Worth borrowing that honesty here.

Claude is genuinely strong on the half of Ogilvy's method that was always labor. The reading, the extraction, the sorting of fact from fluff. That work was expensive because it took a person days, and it is now cheap.

Use it without guilt. Read more than you ever could before.

The other half, the leap from a true fact to a big idea, is still yours. Recognizing that the clock line was the headline, and not one of a hundred other true and boring facts about the car, was judgment. Taste. A sense of what the buyer secretly wants to believe.

The model can hand you the clock. It can't yet know that the clock is the one that matters.

That call is the job, and it's the part worth getting good at, because it's the part that isn't getting automated this year.

Everyone quotes Ogilvy's headlines. Almost no one copies the twenty-one days that produced them.

The tool that finally makes the reading cheap is sitting on your desk, and most people are using it to skip the reading.

Load the research first. Hunt the fact.

Then hold what comes back to the bar that a ship in the night never clears.

More clicks, cash, and clients,
Mark Masters