Build Wiz AI Show

Ad Copywriting: The 4 Sales Every Ad Must Make (Derick Carneiro)

15 min · · English

Derick Carneiro explains how he writes direct-response ads for cold traffic: four sales every ad must make before the product, from 17 elements he calls Ad Psychology. Bob and Alice test his method against copywriting research, attention data and FTC rules. In this episode: - The four sales: content, hope, value, impulse - Finding what customers want to hear - Where urgency and health claims turn deceptive Sources: Derick Carneiro (@DerickCarneiro_); Katie Doll; Joanna Wiebe; Jakob Nielsen; Federal Trade Commission Big AI ideas, made clear for builders. luongnv.com/podcast

Listen on Build Wiz AI Show Download MP3 Transcript

Based on how I write ads that make strangers buy by Derick Carneiro (@DerickCarneiro_)

Transcript

Alice: A cold ad has to make four sales before it ever sells the product. Four! So what are they?

Bob: Content, hope, value and impulse, says Derick Carneiro. And is the real skill writing, or listening?

Alice: Welcome to Build Wiz AI Show. Big AI ideas, made clear for builders. I'm Alice, and today I'm the one asking the questions.

Bob: And I'm Bob! Today it's Derick Carneiro's X article, "how I write ads that make strangers buy". He breaks down his whole process for ads aimed at total strangers, step by step.

Alice: Who is he, and why should we listen?

Bob: He says he's 22 and that his ads and video sales letters sold over 300 million dollars. That's his own claim, and I think the post shows nothing we could check. So in my view, treat it as his story, not a fact.

Alice: Fair. And an ad-writing post on an AI show? Defend yourself.

Bob: Easy. Every founder shipping an AI product has to explain it to strangers. And his last step hands the writing to an AI model, which fails, he says, when people skip the thinking first.

Alice: One thing up front: his worked example is a real product, right?

Bob: Right: he picked a capsaicin supplement as his live example. We only look at how he builds the ad. We don't endorse the product, and we're not repeating or checking any of its health claims.

Alice: Good. And who's checking him today?

Bob: A guide to the classic book "Breakthrough Advertising", a copywriter's research method, and three guests: attention data, plus two documents from the US Federal Trade Commission.

Alice: Start at the beginning. Who are these strangers?

Bob: Cold traffic: people who know they have a problem, and might even know a solution or two, but still haven't solved it. He says they get more frustrated and more skeptical with every new thing that fails.

Alice: So you can't just show them a price.

Bob: Exactly. An ad that opens with price, product or offer will have a hard time at scale. His fix is an indirect ad: grab attention, hold interest, explain a new mechanism, and only then talk product, price and offer.

Alice: Give me a picture.

Bob: It's like meeting a stranger at a bus stop: you don't open with a sales pitch, you open with something they already care about. In an ad, that's their problem, long before any product shows up.

Alice: Is this his own invention?

Bob: He says himself it's nothing new and points to "Breakthrough Advertising". In Katie Doll's Shortform guide to that book, Eugene Schwartz says customers move through five levels of awareness, and your headline should cater to how much they already know.

Alice: And these strangers have seen plenty of ads already.

Bob: Which the guide covers too. When customers know many versions of a product, they're less receptive, and Schwartz suggests a new, distinct benefit. That's the seed of the "new mechanism" Derick builds his ads around.

Alice: Okay, the four sales. Walk me through them.

Bob: First, the Content Sale: grab attention and make them want to hear the ad to the end. Then the Hope Sale: show what's been stopping them, the action that fixes it, and a specific mechanism that does it.

Alice: And the last two?

Bob: The Value Sale presents a product that delivers that mechanism, at a price that feels far lower than the cost of never solving the problem. The Impulse Sale shows that not acting is the bigger risk.

Alice: So the product has to wait its turn. Rude!

Bob: Think of it as a first date before a proposal: nobody says yes to a ring on hello. The Shortform guide makes the same point: educate people on the product's value before you mention the price.

Alice: And how does he persuade?

Bob: With reasons. A sale, in his words, is a process of persuasion, and persuading is presenting an idea and giving reasons why it's true. He calls those reasons "Ad Psychology".

Alice: How many reasons are we talking about?

Bob: Seventeen elements, in four stages. The first five serve the Content Sale: format, topic, concept and two hooks. The visual hook is why someone snaps out of autopilot; the copy hook is why they decide to keep watching.

Alice: Does he write his hooks from scratch?

Bob: No. To avoid unnecessary risk, he picks a hook structure first, one he found in tons of ads, such as "This is what happens to..." or "Watch what happens to...". Then he fits his concept into it.

Alice: Maya, you brought data on how fast people decide.

Maya: I did. Jakob Nielsen at Nielsen Norman Group, in "How Long Do Users Stay on Web Pages?", reports Microsoft Research data on more than 2 billion dwell times. The first 10 seconds of a page visit are critical for the decision to stay or leave.

Alice: Ten seconds? That's harsher than a first date!

Maya: It's because users are extremely skeptical after countless poorly designed pages. It's like a bouncer at a club door: most people get turned away in seconds, but whoever gets past may stay for minutes. Nielsen's advice: communicate your value proposition within 10 seconds.

Alice: Does that hold for video ads in a feed?

Maya: Not directly: this is about web pages, not ads in a feed. Read it as a hint, not proof: people spend their first moments in ruthless triage.

Alice: Next, the Hope Sale. What goes in it?

Bob: Six elements. First identification, so the viewer feels understood, close to Schwartz's advice in the Shortform guide to echo an emotion they can identify with. Then the point of contact: what they already suspect is causing their problem.

Alice: And the famous unique mechanism?

Bob: It comes in three parts. The mechanism of the problem is the missing reason that proves their suspicion right. The mechanism of the function is the action that fixes it, and why other solutions can't do it.

Alice: And then the solution.

Bob: The one solution that performs that action, plus conditions: what makes it rare or hard to get. Imagine a detective who confirms the hunch you already had, then names the one culprit nobody else caught: the hunch is the point of contact, the culprit is the mechanism.

Alice: Hang on. In his supplement example, are those health claims true?

Bob: We're not checking them, and we won't repeat them: in my view, that's not what this episode is for. But someone has a rulebook for exactly this kind of ad. Sam?

Sam: The Federal Trade Commission's "Health Products Compliance Guidance" says advertising must be truthful and not misleading. And before an ad runs, advertisers must have adequate substantiation for every objective product claim, stated or implied.

Alice: Even when the mechanism is just one ingredient?

Sam: That's one of the guide's own examples. A drink touting a "clinically tested ingredient" implies the results prove a benefit, and that the drink itself delivers it. It's like vouching for a whole cake because one egg passed inspection.

Alice: Back to the craft. What skill does he say matters most?

Bob: Picking the right reasons: the ones your audience actually wants to hear. His big insight is that your job isn't to say what you think needs to be said. Joanna Wiebe at Copyhackers puts it bluntly: you're not your prospect.

Alice: So what is the job?

Bob: To find WHAT they want to hear, how they want to hear it, and who they want to hear it from. It's like a tailor who measures before cutting: the cloth is your message, and the measurements come from the customer, not from you.

Alice: A tailor for ads. And how does he take those measurements?

Bob: Premises first: what you already know about the brand and the product. Then the target: market, then segment, then the avatar, his word for the profile of the audience.

Alice: And then lots of deep customer research?

Bob: Here's his controversial bit. He calls deep psychographic research a low-leverage activity, and says: steal other people's research. Winning ads in your niche already carry the pains, fears, desires and language.

Alice: Steal it? Bold career advice.

Bob: Wiebe goes one step further, with a twist: "Instead of writing your message, steal it. Steal it directly from your prospects." You take what they say and mirror it back to them.

Alice: Does that actually work?

Bob: In her example, she read more than 500 Amazon reviews for a client. A headline taken from one reviewer's words brought over 400 percent more clicks than the control.

Alice: So: competitors' ads versus customers' own words.

Bob: That's the difference. Derick mines what already sells, Wiebe mines what customers actually say. And he saves his own research energy for behaviour: the news, videos and topics that grab the audience's genuine interest.

Alice: Then comes the brief. What's in it?

Bob: All seventeen elements, filled with what the audience wants to hear. For the offer, he starts with conditions: why the viewer can't just get the mechanism on their own. And the selling proposition is simply the answer to those conditions.

Alice: So he doesn't invent a differentiator?

Bob: His words: "I don't need to make up any differentiator." It's like a key cut for a lock you've already shown: the conditions are the lock, and the product just has to fit them.

Alice: And features? A long ingredient list?

Bob: No catalog. Every feature has to answer one of four levers: better odds, a more tangible dream scenario, less effort or a shorter timeline. Nobody buys the ingredient, he says; they buy what it does for them.

Alice: And the close? Do we finally get the hard sell?

Bob: He says the close isn't about squeezing anyone. It's about showing that the safest decision is buying now: a guarantee framed as plenty of time to try it, plus what you gain now and lose by waiting.

Alice: Why frame a guarantee as time to try?

Bob: The Shortform guide names the lever: loss aversion, worrying more about what you might lose than hoping for what you might gain. Its antidote is a risk-free way to try, like a trial period.

Alice: And what makes waiting feel like a loss?

Bob: Timing. The ad runs during Black Friday, and the script says the sale stock is running out and the price goes back up once new stock arrives.

Alice: Theo, this is where you get nervous.

Theo: I do. The Federal Trade Commission's report "Bringing Dark Patterns to Light" was announced in an FTC press release. It flags countdown timers that make people believe an offer is time-limited when it really isn't.

Alice: So scarcity is fine only if it's true?

Theo: That's the line. The same report flags ads designed to look like independent, editorial content. And it warns that companies experiment with dark patterns to exploit the most effective ones.

Alice: Okay, last step: actually writing. Finally!

Bob: He calls it the fastest step, because the hard part, the psychology, is already done. You break a proven reference ad into mini-blocks, like hook, belief break, mechanism reveal and product presentation, then fill them with your brief.

Alice: By hand, or with AI?

Bob: Either. Here's the key point: the brief decides WHAT gets said, and the reference only decides the order and the rhythm. Picture the reference as a song's beat: you still write your own lyrics over it.

Alice: So why do so many AI-written ads flop?

Bob: His answer: people skip the brief and expect the model to guess the psychology for them. I'd bet anyone who's prompted a model with one vague line knows that feeling.

Alice: Guilty as charged.

Bob: And he spreads risk across a batch: a few conservative ads, a lot of moderate ones, some aggressive ones. His rule, like a poker player's: minimize risk where leverage is low, raise the bet where leverage is high.

Alice: Time for pushback. Theo's disguised ads made me think: his very first element is a format where the viewer doesn't even suspect a sale!

Bob: That's the tension in one sentence. I think a native format can be honest storytelling or a disguise. The test is whether the viewer would still feel fine once they know it's an ad.

Alice: Sam, what about testimonials? These ads love them.

Sam: The guide says an honest testimonial still isn't enough: the advertiser needs scientific evidence that the product will work for buyers as it did for the endorser. And "Results not typical" doesn't cure the deception.

Alice: Does that land on the copywriter, or only the brand?

Sam: Both can be on the hook. The FTC has acted against ad agencies, infomercial producers and expert endorsers, not just product marketers. And it judges the net impression of the whole ad, images included.

Alice: Bob, so where does his method cross the line?

Bob: I think the structure itself is neutral. Finding what customers want to hear is research; making up a mechanism, a result or a deadline is manipulation. And his own results are his report, not audited numbers.

Alice: Maya, any caveat on your attention numbers?

Maya: One. Nielsen describes the median curve across a huge dataset of web pages, not your audience in your feed. So my take: test your own first seconds instead of borrowing anyone's.

Alice: Time for Builder's Corner. Bob, one thing to try this week?

Bob: Before you write your next landing page hook, collect real customer phrases from reviews, support tickets and sales calls, a full page of them. Wiebe's rule: their words become yours.

Alice: And then what do I do with the page?

Bob: Look for the phrase that keeps coming back, and make it your first line, so a stranger gets the value fast. Nielsen's 10 seconds is your budget. Then reread the page as your most skeptical customer would.

Alice: Let's land it. What should people remember?

Bob: A cold ad makes four sales before the product: content, hope, value and impulse. Each one is made with reasons the audience already wants to hear. And the brief, not the prose, does the heavy lifting.

Alice: And from our guests: hooks get seconds, claims need proof, and urgency has to be real.

Bob: My take: use his structure to find message-market fit, and keep every reason in it true.

Alice: What's still open? For me: how much of his results would survive an independent audit?

Bob: And mine: does a mechanism-first ad even fit software, where you can just show the product working? My guess is a live demo can be the mechanism.

Alice: Plus, if everyone steals the same winning ads, how long does a winner stay a winner?

Bob: That's our look at how direct-response ads are built. Thanks to Derick Carneiro for his X article.

Alice: Thanks also to Katie Doll's Shortform guide to Breakthrough Advertising, Joanna Wiebe at Copyhackers, and Jakob Nielsen at Nielsen Norman Group. And the Federal Trade Commission's health products guidance and dark patterns report.

Bob: That's all for today on Build Wiz AI Show. Thanks for listening, and see you next time.