A $0 Text Post Outsold a $15K Video Ad 28-to-1. Here's Why.

Why your ads don't work (even when they look great)

Beautiful banner. Polished landing page. Targeting dialed in to perfection. Result? Nothing.

Ad production is a technical beast — copy, visuals, targeting setup, keywords, optimization. You get buried in the mechanics and forget to ask the only question that matters: what are we actually saying, and to whom? When that happens, the outcome writes itself. The ad looks expensive, the targeting is tight, and the audience scrolls right past. The message is empty. There's nothing to grab onto.

The fix isn't making things prettier. The fix is stepping back before you open any design tool and thinking about the message first. A plain ad with a sharp message beats a gorgeous ad with a hollow one. Every single time.

This probably sounds obvious. But you'd be shocked how many people burn their budget — or worse, burn their time — chasing a slick format while ignoring the fundamentals.

Here's a real example. In 2024, a creator with 200,000 Instagram followers decided to launch a paid course. Hired a production crew. Shot a series of cinematic promo videos — Netflix-trailer quality. Spent $15,000 on production. Ran ads. Sales: 12. Three months later, another creator in the same niche launched a course with a text post in Stories: "I've been doing X for 4 years. Here are 3 mistakes that cost me $40,000. My course shows you how to avoid them." Production budget: zero. Sales: 340 in the first week.

The difference wasn't budget. The first one said, "Look how beautiful my course is." The second said, "Here's a specific problem I solve." Format is secondary. Message is primary.

What to do: Before you create anything, write down — in one sentence — what specific problem your product solves for a specific person. If you can't, you're not ready to advertise.

Jobs to be done — what job is the customer hiring your product for?

There's a framework that makes this systematic. Jobs to Be Done has been around for decades and has proven itself from startups to Fortune 500s. The core idea is simple: people don't buy products. They hire them to get a specific job done.

A loan is a good example. Nobody takes out money just "to have money." You hire a loan to start a business, renovate your apartment, or buy shoes. The loan is a tool. The job is whatever you hired it for. Same with a hairbrush: you don't buy it to brush your hair — you buy it so your hair looks good. The function is brushing. The job is looking good. The distinction seems subtle, but it's exactly what determines what you write in your ad.

A phone makes it even clearer. One person buys an iPhone to make calls. Another buys it to shoot content for Instagram. A third buys it to look successful in a client meeting. One product. Three different jobs. Three different messages. Write one generic ad for all three — and you miss every single one.

Every segment of your audience has its own set of "jobs" they hire your product or service for. Your task: map those jobs, form hypotheses, and test which message hits hardest. The logic is straightforward — test cheap, find what works, then scale. Not the other way around.

What to do: List 3–5 reasons clients buy your product. Not "what you do" — "why they need it." Those are your JTBD hypotheses.

Example: a marketing consultant

Say you're a marketing consultant. You help businesses grow. But "I help businesses grow" isn't a message. It's noise. Let's look at how this plays out across two real segments.

Segment 1: Startup founder. Has a product, some early users, but no idea how to scale customer acquisition. No marketing team. No budget for an agency. This person isn't hiring a consultant "for marketing" — they're hiring a consultant to stop wasting time on random experiments and get a clear plan for what to do in the first 90 days.

Message insight: "Stop guessing. Get a 90-day plan that works on a $500/month budget."

Segment 2: Freelancer who wants to move into consulting. Does hands-on project work but wants to sell expertise, not hours. They don't need a strategy for someone else's business — they need a strategy for their own. They're hiring a consultant to figure out how to package their experience into a $3,000+ service instead of a $50/hour rate.

Message insight: "Stop selling hours. Start selling outcomes. Here's how."

Same consultant. Same expertise. Two completely different messages — and if you send the first one to the second segment, it falls flat. And vice versa.

These examples are hypotheses, obviously — made up, not pulled from client interviews. The point isn't the examples themselves. It's the mechanics. You take 3–5 hypotheses per segment and test them methodically. Write a LinkedIn post from one angle — watch the reaction. Send a cold email from another — measure replies. Change the service description on your site — track how inquiries shift. When one version clearly outperforms, you scale it, sharpen it, and make it the foundation of everything you communicate.

What to do: Pick two key segments. For each, write the "job" they're "hiring" you for — one sentence. Test it in your next post or email.

Testing never stops

Found a message that works? Good. But that's not the finish line — it's the starting line.

Creatives and messages burn out. Audiences get used to them. Competitors copy your phrasing. Context shifts. In 2024, hundreds of freelancers built their positioning around "I'll do it for you by hand, high quality, on time." Then AI tools showed up and did "by hand" faster and cheaper. A message that worked for two years stopped working in two months. The ones who pivoted to "I'll help you make the right decisions so AI tools work for your business instead of generating garbage" — they won. The ones who kept running the old message? They lost clients.

Testing isn't a phase. It's an operating mode. You're constantly checking: is what I'm telling my audience still relevant? Still resonating? Or has the world moved and I haven't?

What to do: Once a month, audit your key messages. One question: "If I were writing this from scratch today, would I say it the same way?" If the answer is no — time to update.

For freelancers and consultants

Everything above might sound like advice for brands with ad budgets. But JTBD works even harder for people who sell themselves.

What doesn't work

Three mistakes that look like marketing but kill your results.

Describing services by features, not outcomes. "I do logos, banners, branding" — that's a menu, not an offer. The client doesn't know what to pick. "I help startups launch in 2 weeks with visuals they're not embarrassed to show investors" — that's a message. The first describes you. The second solves the client's problem. Reframe your service descriptions from "what you can do" to "what the client gets."

Chasing trendy formats without a strategy. Reels, Threads, podcasts — every year brings a new channel that "everyone needs to try." A creator who jumps on every platform without a clear message gets zero traction on all of them. Same mistake: beautiful format, empty message. Pick one channel. Nail one message. Get results — then expand. Message first, channel second.

"I'm for everyone" instead of clear positioning. A freelancer who takes any project from any client can't articulate a JTBD — because there's no segment. No segment means no specific message. No specific message means no conversions. The circle closes. Choosing a niche feels scary. Staying without one costs more. Pick the one segment that brings your best projects, write one JTBD message for them, and test it for a month.

The hard part of marketing is the word "many"

You might get the impression that marketing is impossibly complex. It's not. It's a lot of simple things.

The complexity starts at the word "a lot."

The JTBD exercise we just walked through? Simple. Form a hypothesis, write a message, test it, measure. Anyone can do it in an evening. The hard part is building a system where this happens continuously. Every month. For every segment. With measurements and iterations. Year after year.

It's the same difference as cooking an egg for breakfast versus running a restaurant. The egg is easy. Organizing a kitchen where 20 dishes come out at consistent quality every day, the cooks don't screw up, and they actually show up to work — that's hard. The recipe isn't hard. The process is.

Marketing works the same way. For the JTBD approach to pay off, you need to be a little bit boring and systematic: record your hypotheses, test regularly, document results, update messages when they stop performing. Not as a one-time exercise — as an operating mode.

The good news: if you're a freelancer or consultant, your "restaurant" is just you. Smaller scale, simpler process. But the discipline is exactly the same.

What to do: Create a document called "My JTBD Hypotheses." Three columns: segment, hypothesis, test result. Update it once a month. Six months from now, you'll have a system you couldn't have built in a single evening — but one that generates clients without constant effort.