Rewriting is the cheap, obvious response when nothing sells, and it rarely changes anything. A business can spend months on headlines while the actual blocker sits one level further back, in what is being sold rather than how it is described.
This article covers the two failures that cause most of it, how they show up differently for a service business and a product business, and a diagnostic that takes about twenty minutes on your own page.
Is this a positioning problem or an offer problem?
A positioning problem is about who you are for and how you sound, and its symptom is that nobody can tell you apart from four competitors. An offer problem is about the deal itself, and its symptom is that people understand you perfectly, say so, and still do not buy.
They are different problems and they fail differently, so it is worth separating them before changing anything. A positioning problem is a message problem, and rewriting genuinely fixes it. An offer problem is about what is included, what the buyer ends up with, what happens if it goes wrong, and how much of it they have to work out for themselves.
The tell is the quality of the attention. Silence usually means a positioning or a distribution problem. Engagement without purchase, people reading properly, replying warmly, saying it was useful, and then not buying, usually means an offer problem. That second pattern is the one this article is about.
The first failure: describing the work instead of the result
Most sellers describe what they do, because it is what they know best and what they spend their days on.
For a service business that reads as a list of inputs: the number of sessions, the call cadence, the deliverables, the revision rounds, the length of the engagement. Every line is accurate and every line is about the supplier's process.
For a product business it reads as a specification: materials, dimensions, weight, capacity, what is in the box, how long it lasts. Again accurate, again about the object rather than the person buying it.
In both cases the buyer wants something on the other side of the purchase. Not twelve sessions, but a plan they can run without help. Not a 45 hour burn time, but an evening that feels a particular way. Between the description and the wanted outcome sits a gap, and the buyer is the one filling it in. They have to reason from your inputs to their result, using less information about your work than you have.
Some will do that and reach yes. Most will not attempt it, because the effort is real and their interest is provisional. The loss is invisible, which is what makes it persistent: nobody writes in to say they could not work out what your service would get them.
The fix is to make the translation yourself. State what the buyer has after the purchase that they do not have now, in their language, and let the inputs sit underneath as supporting detail rather than as the headline. The specification is not deleted. It stops being the pitch.
The second failure: leaving all of the risk with the buyer
Moving risk back to the seller's side does not have to be expensive. Make the first step small and keepable, name who the offer is not for, remove the uncertainty about what happens next, and show it working for somebody who is not you. None of those requires a money-back guarantee or a free trial.
Look at a standard offer from the buyer's side. They pay first. They find out afterwards whether it worked. If it does not, they lose the money, the time, and whatever standing they spent recommending it internally. The seller loses nothing.
This is not a fairness argument, it is a conversion one. The person with the least information is being asked to absorb all of the downside. A buyer who is unsure will resolve that by not deciding, which looks identical to not being interested.
The common response is a discount, and it is the wrong instrument. Price and risk are separate objections. Reducing the price of something a buyer is unsure about produces a cheaper thing they are unsure about. The uncertainty is untouched.
The four moves, in both shapes
Make the first step small, real, and keepable. For a service, something the buyer retains whether or not they continue: a written plan, an audit, a recording. A discovery call does not count, because it is worth more to the seller than the buyer. For a product, a single unit, a sample, or a sizing and suitability guide honest enough to talk the wrong buyer out of the purchase.
Name who the offer is not for. Exclusion is a strong trust signal precisely because it is costly to fake, and the customers it turns away are the ones who would have been unhappy.
Remove the uncertainty about what happens next. For a service, describe day one, week one and week four. For a product, state when it ships, what arrives, and exactly what happens if it is wrong. Ambiguity is a cost the buyer pays in advance.
Show it working for somebody who is not you. One real customer photograph, or one sentence in a customer's own words, carries more weight than another product shot or another paragraph about your methodology.
What does the rebuild actually look like?
Two worked examples. Both are constructed illustrations rather than client work, so read them as demonstrations of the change, not as evidence of a result.
A service, before: "Six one-hour coaching sessions, £900. Includes email support and a workbook."
After: "In six weeks you will have a marketing plan you can run on your own, and you will have run it once with me watching. If after the first session you decide it is not for you, we stop and you keep the plan."
A product, before: "Handmade soy candle, 220g, 45 hour burn time, cotton wick, £28."
After: "A candle that makes the room smell like you meant it, for about six weeks of evenings. If the scent is not right, send it back opened and we will swap it once."
In both, the work is unchanged, the price is unchanged, and the person or object delivering it is unchanged. Two things moved. The description now names the outcome, so the buyer does no translation. And the seller absorbed a defined, affordable slice of the risk: one session, or one swap on an opened item.
How do you tell which failure you have?
Open the page you sell from and read it as somebody who has never encountered your business. Then answer three questions. What does the buyer have afterwards that they do not have now? What happens if it does not work, or it arrives and it is wrong? And what does the buyer have to guess at?
If the honest answer to the first is a count of sessions, hours, files, grams or millimetres, the offer describes the work or the object rather than the result. That is failure one.
If the answer to the second is that the buyer loses their money, all of the risk sits with them. That is failure two.
The third question is the one that produces the week's work. List what the buyer has to guess at. Every item on that list is a reason somebody read the page, understood it, and closed the tab. Whatever appears there is the thing to fix, and it is almost never the headline.
How long does this take to fix?
The rewrite itself is short, usually a paragraph, and it costs nothing. The part that takes time is deciding what risk you are willing to carry, because that is a commercial decision rather than a copy decision, and it should be made deliberately rather than in the middle of an edit.
A practical sequence for a business with no budget for outside help: rewrite one sentence this week, the one describing what the buyer gets. Watch what changes. Then, separately, decide on one risk-reducing element you can afford to honour every time, and add it. Two changes, made one at a time, are easier to attribute than a full rewrite.
When is it genuinely the marketing?
Not every conversion problem is an offer problem, and it is worth naming the cases where this diagnosis does not apply.
If the traffic is unqualified, no offer will fix it. If nobody is reading at all, the problem is distribution or positioning, not the deal. If people convert well from one channel and not from another, that points at the channel and its audience rather than at the offer, which is constant across both. And if the price is genuinely wrong for the market, that is a pricing question, which this diagnostic deliberately does not touch.
The offer diagnosis applies to one specific pattern: qualified people, paying real attention, understanding what is on the table, and declining anyway.
↳ Frequently asked
01How do I know whether my problem is the words or the offer?
Look at the quality of the attention, not the volume. If nobody engages at all, that points at distribution or positioning. If people read properly, understand it, tell you it was useful, and still do not buy, that points at the offer.
02Does this apply if I sell products rather than services?
Yes, and it shows up differently. A service business lists inputs like sessions and deliverables. A product business lists a specification like materials, dimensions and capacity. Both are accurate, and both leave the buyer to work out what they actually end up with.
03Will discounting fix a conversion problem?
Not if the hesitation is about uncertainty rather than price. Price and risk are separate objections, and reducing the price of something a buyer is unsure about produces a cheaper thing they are unsure about.
04Do I need a money-back guarantee to reduce the risk?
No. A keepable first step, naming who the offer is not for, stating exactly what happens after payment, and showing one real customer outcome all move risk off the buyer without a guarantee and without costing real money.