If nobody is buying, one of four things is true: the people you are reaching do not have the problem, they have the problem but do not recognize your description of it, they recognize it but do not believe you can fix it, or they believe you and the price is wrong. From the outside all four look identical — silence — and each one has a different fix. Testing them in that order matters, because the cheapest correction sits at the top and the most expensive rebuild sits at the bottom.
Key takeaways
- One symptom, four causes. Guessing which one you have is how a quarter disappears.
- Test in order: audience, then language, then credibility, then price.
- Change one variable per test, or the result cannot be attributed to anything.
- If fewer than a few hundred qualified strangers have seen a clear claim, this is a distribution problem, not a demand problem.
- Some ideas are genuinely finished. Learning that in a month is a win, not a failure.
Silence is the worst signal a business can get, because it is the only one that is consistent with every possible explanation. A refund tells you something. An objection tells you something. A person who reads your page and quietly closes the tab tells you nothing at all, and you are left to fill in the reason yourself. Most owners fill it in with whichever explanation is cheapest emotionally, which is usually "people just have not heard about it yet."
Sometimes that is true. More often the traffic is fine and something upstream is broken. The useful move is to stop treating "no one is buying" as one problem and start treating it as four, then eliminate them in a fixed order.
The four causes, in the order to test them
The order is not arbitrary. Each cause is cheaper to test and cheaper to fix than the one below it, and — this is the part that gets missed — each one masks the ones below it. You cannot learn anything about your price from an audience that does not have the problem. You cannot learn anything about your credibility from people who never understood the offer.
| Cause | What you observe | Cheapest test |
|---|---|---|
| Wrong audience | Almost no clicks. People who do arrive bounce in seconds without scrolling. | Same claim, a narrower and more specific audience. |
| Wrong language | Clicks arrive, time on page is short, nobody asks questions. | Same audience, the problem described in their words instead of yours. |
| No credibility | People read everything, then leave. You get "looks interesting" and no action. | Add proof: a named customer, a number, a guarantee, a face. |
| Wrong price | People reach the price and stop. You hear "how much?" then nothing. | Same everything, one different price or payment shape. |
Cause one: they do not have the problem
This is the most common and the least popular explanation, because it is the only one that questions the idea rather than the execution. It does not necessarily mean nobody has the problem. Usually it means the specific group you chose to reach does not have it badly enough to do anything about it today.
What it looks like
- Impressions without clicks. People see the offer and do not react. Nothing in it is aimed at them.
- Instant bounces. Arrivals leave in under five seconds, before they could have read enough to reject it.
- Polite interest from the wrong people. Friends and peers say it sounds useful. None of them are the buyer, and none of them buy.
How to check it
Narrow the audience until you could name three real people who fit it, then run the same claim again. This feels backwards — the instinct when nothing sells is to widen the net — but a broad audience produces mediocre numbers, and mediocre numbers cannot be acted on. A test that fails against a specific group teaches you something. A test that fails against everyone teaches you nothing.
The mechanics of choosing that group are in how to choose a marketing channel for a small business, and the interview work that gives you the shortlist is in how to talk to customers before you build.
Cause two: they have the problem and your words do not match it
This is the most fixable cause and the one most often mistaken for the first. The audience is right. The problem is real. But you are describing it in the language of someone who has already solved it, and they are still living inside it.
Owners write from the far side of their own expertise. You have thought about this problem for a year, so you have a name for it, a category for it, and a vocabulary that compresses it. Your buyer has none of that. They have a bad Tuesday and a workaround they are embarrassed about.
People do not search for your solution. They search for the words they use when the problem is happening to them.
What it looks like
- Clicks without depth. People arrive, skim, and leave. The headline earned the click and the page did not confirm it.
- No questions. Nobody asks anything. A person who half-recognizes their problem asks a follow-up. Silence means no recognition.
- Explaining in conversation works. In person you can talk someone into understanding it in two minutes. That gap is the whole problem — your page does not do what your mouth does.
How to fix it
Steal the words. Go where your buyers already complain — a subreddit, a Facebook group, review sections of the tools they tolerate, your own support inbox — and copy the phrasing they use before anyone has sold them a framework. Then put those phrases in the headline instead of your category name.
This is the fastest single improvement available to most small businesses, and it costs nothing but attention. How to write ad copy for a small business covers the mechanics, and how to write a problem statement from a conversation covers turning what you heard into one testable sentence.
Cause three: they believe the problem and not you
Now the audience is right and the words land. People read the whole page. They understand exactly what you are offering. And they do not act, because nothing on the page gives them a reason to think you specifically can deliver it.
This is a credibility gap, and it is invisible to the person who has it. You know you can do the work. The reader knows nothing except that a stranger on the internet made a claim, and the base rate for strangers on the internet making claims is bad.
What closes it
- One named specific. A real customer, a real number, a real before and after. One concrete example beats five adjectives.
- A visible person. A name and a face. Anonymous businesses ask for more trust than they have earned.
- Risk moved off the buyer. A guarantee, a trial, a first job at cost. If they cannot lose, believing you is cheaper.
- Evidence of ordinary competence. Answered questions, a real address, replies that arrive. Most trust is built by unremarkable reliability rather than by claims.
For a new business with no customers yet, the honest version of this is doing the first few jobs nearly free in exchange for permission to describe the result. That is not a marketing trick. It is how you buy the only asset you are missing.
Cause four: everything lands and the price is wrong
Last, because it is the cause most likely to be blamed first and least likely to be the actual problem. Price is also the only one of the four where the fix can go in either direction, and where guessing wrong is expensive in a way that is hard to reverse.
The failure is rarely that the number is too high in absolute terms. It is that the number does not match the size of the problem as the buyer understands it. Someone who thinks they have a minor annoyance will not pay serious money to fix it, no matter how serious the annoyance actually is. That is a framing problem that looks exactly like a pricing problem.
Before you cut anything
- Check whether the problem was ever sized. If your page never says what the problem costs them per month, the price has nothing to be compared against, so it reads as pure expense.
- Try the shape before the number. Monthly instead of annual, per job instead of per month, deposit instead of full amount. Payment shape moves conversion more than price often does.
- Raise it as a real option. Cheap signals low stakes in categories where people are buying a fixed outcome. A higher number with a stronger promise sometimes converts better.
Cutting price to solve a comprehension problem is the most common self-inflicted wound in small business marketing. It converts a diagnostic question into a permanent margin decision.
How to run the whole diagnosis in about a week
The point of the ordering is that you can move through all four in days rather than months, provided you change exactly one thing at a time and decide in advance what each outcome will mean.
- Write the claim as one sentence. Who it is for, what problem it removes, what it costs. If you cannot fit it in a sentence, the page cannot either.
- Name the audience narrowly enough to picture three people in it. A job title and an industry, or a town and a life situation. Not "small businesses."
- Decide the threshold before launch. Write down the number of clicks or replies that will count as interest. Deciding afterward guarantees you will read the result as whatever you already believed.
- Run the claim against that audience for a fixed window. A few days, one small budget, one variable. Do not extend it because it looks promising on day two.
- Read the drop-off point, not the total. Where people stopped tells you which of the four causes you have. The overall number only tells you whether to continue.
- Change one thing and run it again. Audience first, then language, then proof, then price — in that order, because each masks the next.
The mistake that hides all four causes
Changing everything at once. It happens because it feels productive, and because after a failed week the temptation to rebuild is overwhelming. So the audience changes, the headline changes, the page gets redesigned, and the price drops — and whatever happens next is uninterpretable. If it works you do not know why, which means you cannot repeat it. If it fails you have eliminated nothing.
The second version of this mistake is running for too short a window and calling it a test. A day of spend against fifty people is a coin flip you have decided to treat as evidence. How to read ad test results covers what a readable result actually requires.
What finding out actually costs
The reason to run a paid test rather than wait is speed and cleanliness. Strangers who owe you nothing either respond or they do not. Friends, followers, and your own mailing list are all contaminated samples — they are being kind, or they already know you.
| Spend | What you can learn | What you cannot |
|---|---|---|
| Under $100 | Whether the creative is catastrophically off. Little else. | Anything about which of the four causes you have. |
| $200 – $500 | Whether one narrow audience reacts to one claim. Usually enough to separate cause one from cause two. | Reliable differences between two similar headlines. |
| $500 – $1,500 | Two or three claims against the same audience, and a readable drop-off point. | Anything about retention, repeat purchase, or long-term margin. |
| A quarter of building | That you can build the thing. | Whether anyone wanted it, which is the question you started with. |
- Demand test
- A small, fixed-window advertisement of a claim to a narrow audience of strangers, run to find out whether the problem is felt strongly enough to produce action — before the thing being advertised is fully built.
The budget guidance in more detail is in how much a small business should spend on ads, and if you have already tried paid and it went nowhere, why your Facebook ads are not working covers the specific failure modes.
When the honest answer is that the idea is finished
Sometimes you run clean tests against two or three genuinely different audiences, with clear claims, honest prices, and real proof, and nobody acts. That is not a failure of the test. That is the test working.
The cost of learning it this way is a few weeks and a few hundred dollars. The cost of not learning it is the version where you find out after eighteen months, having spent savings and goodwill on a claim no stranger ever responded to. An ad is not a company argues the other half of this: a good result does not prove a business either. Both directions of overreading are expensive.
The decision framework for stopping — as opposed to persisting through a bad month — is in when to quit a business idea.
Related reading
- Why my Facebook ads are not working — when you have already paid for traffic and nothing happened
- Traffic but no sales — when people arrive and still do not buy
- How much should a small business spend on ads — what a readable test costs
- How to get your first 100 customers — once a claim survives the test