Answer in brief
A listing with no sales fails at one of three points: it is never shown, it is shown and never opened, or it is opened and never bought. Each has different causes and different fixes, checked cheapest first.
The short answer: three stages, checked in order of cost
The honest answer is unglamorous. A listing fails at one of three points, and they are not interchangeable: it is never shown, it is shown but never opened, or it is opened but never bought. Until you know which one is yours, every fix you attempt is a guess.
The order of checking is set by what each check costs, not by what feels most important. Impressions sit in your seller reports and take fifteen minutes to read. Clicks sit beside them and only mean something against impressions. A refused purchase can be studied only on the listing itself.
So the diagnosis runs from cheap to expensive: first what costs nothing, then what costs an hour of attention, then what costs a photo shoot or a rewritten description. That order spares you the worst expense of all, which is paying to fix a problem you do not have.
And let us dispose of the comfortable answer early. Blaming the price is a conclusion, not a hypothesis. It can turn out to be true at the third stage, and it almost never explains zero impressions: a discount does not create impressions, it only shrinks the margin you were working with.
Stage one: is the listing being shown at all?
An impression means your listing reached a results page or a shelf where a buyer could have seen it. Zero impressions means you have not met a buyer yet, not once. Rewriting copy at this stage changes nothing at all, because there is nobody standing there to read it.
This is quick to test. Open the listing statistics for a period when you are certain the item was in stock, and look at the impression count. Then search the marketplace yourself for the exact full product name, written the way it appears inside your seller account.
If the exact name does not find your own listing, you do not have a demand problem, you have an availability problem. The item is missing from the platform index, hidden, rejected in moderation, filed under a closed or wrong category, or recorded as out of stock.
The distinction matters because the two cost very different amounts to repair. Demand problems are solved by working on the offer and the copy. Availability problems are solved administratively: restore the stock, change the category, bring the listing into line with platform rules. The second is almost always faster and cheaper.
Why impressions stay at zero: category, attributes, stock, moderation
The category is a mechanical failure that keeps recurring, and there is a structural reason for it. It is chosen once, usually in a hurry, and from then on it decides which filters and shelves the product can even appear in. A wrong choice does not stop the listing from existing, so it goes unnoticed for months.
Attributes fail just as quietly. Marketplace buyers rarely scroll a full results page; they narrow it with filters for size, colour, material, capacity or compatibility. An empty attribute field does not make your listing look any worse, it simply removes it from every selection built on that filter.
Stock and listing status sound too obvious to check, which is precisely why they should be checked by hand rather than from memory. An item with no stock, a paused card, a listing back in moderation, and a duplicate pulling history away from the original all produce the same symptom: silence in the reports.
What these four have in common is that none of them is about the product. They are fixed by correcting data, not by doing marketing. That is why this stage goes first: a week of unglamorous data work sometimes restores impressions that had been missing for half a year.
Stage two: impressions arrive, clicks do not
The second case stings more than the first, because buyers can see you and walk straight past. Here you get your first meaningful ratio, clicks against impressions. Its absolute value has nothing to be compared with, but its movement after a specific change is easy to read and hard to argue with.
It helps to be precise about what the buyer actually sees at that moment. Not your listing. A tile: one thumbnail the size of a fingernail, a truncated title, a price, sometimes a rating and a review count. Everything else you wrote is physically absent from that screen.
So the work belongs to the tile, not to the listing page. The first image has to survive a small screen: the object large in frame, no fine print, no collage of five scenes, and margins generous enough to survive whatever crop the platform decides to apply.
Titles get cut on the tile. That means the first words have to answer what this thing is, rather than opening with a brand name that means nothing to this buyer yet. Keywords parked at the end of a long title never reach anybody's eyes.
What decides the click: first image, title, and your price beside a rival
This is the screen where your price first matters, and not on its own but beside its neighbours. The buyer does not know your costs and is not comparing you with last month. They are comparing five tiles in a row and deciding which one is worth opening first.
Which leads to an uncomfortable conclusion: the same price can be high and low at the same time. It is high if a visually identical product sits next to it for less. It is low and faintly suspicious if the tiles around it cost twice as much and are presented far better.
So the right move on price at this stage is explanation rather than reduction. Quantity, volume, warranty, delivery window and what the set contains should all be visible on the tile and in the first words of the title, or the buyer reads your price bare.
One honest caveat: you cannot buy review counts, and no amount of rewriting will accelerate them. A new listing has none by definition. That is not a reason to cut the price; it is a reason to stop measuring yourself against a listing that has been selling for three years.
Stage three: clicks arrive, orders do not
The third case is the most expensive to diagnose and by far the most informative. The person already arrived, spent attention, scrolled the gallery and left. So the page gave them something other than the tile had promised, or failed to give them the thing they opened it for.
The first common cause is a gap between promise and content. The tile showed a set, the page sells a single unit. The title implied a size the description never states. A gap like that kills the order more reliably than any price, because it makes the rest of your text untrustworthy.
The second is unanswered questions the buyer will never put to you. Measurements in centimetres, weight, material, compatibility, what is actually inside the box, return terms and delivery windows. A missing answer does not read as unknown; it reads as something being kept from the reader.
The third is gallery order. The early frames should answer buying questions: the whole object, its scale against a hand or a room, the detail that matters, what is included. A beautiful scene with the product out of focus spends a view that lasts only a few seconds.
Why lowering the price is a lazy diagnosis
The advice to cut the price is popular because it can be given without opening a single report. It always sounds plausible and it always produces a visible effect, just not the one you needed. A discount changes the terms of the deal before you have understood what the cause was.
Take it stage by stage. Against zero impressions a discount does nothing, because the absence had nothing to do with price. Against zero clicks it can genuinely help, but only if the problem really was comparison, and not an unreadable thumbnail or a title cut off mid-phrase.
At the third stage a discount can indeed close the sale, but it closes it instead of answering the buyer's question. You are paying money for something a line in the description would have solved. And you pay it on every sale, because the line is written once and the discount is granted repeatedly.
There is a delayed effect as well. A price dropped for a test sets an expectation the buyer returns to. Raising it back is harder than lowering it was, and the listing still carries the same unreadable first image that started the whole problem in the first place.
The second door: your own site, Product markup and external search
A marketplace has its own internal search; the open web has another. They are separate mechanisms and confusing them is expensive. The platform's search reads its own catalogue fields: category, attributes, title, stock. It does not read the markup on your website at all, in any form.
Schema.org Product markup works somewhere else, on your own pages and for external search engines. It describes the item to machines: name, image, description, identifier, and an offer carrying price and availability. It is a way of being understood, not a lever on ranking inside a marketplace.
The practical value of a second door is simple. While the listing depends on somebody else's ranking rules, the page on your own domain stays an address you control. It can send people to the marketplace, and its content does not vanish when a category changes or a card is pulled.
Separating the two also protects your budget. Fixing attributes in a seller account and marking up a product page on your site are different jobs with different speeds and different outcomes. A package that blends them into one line of an estimate usually hides what you are actually paying for.
Listing readability: contrast, text baked into images, and PDFs
Listings are read on phones, often in motion and almost never in good light. Accessibility here is not a formality, it is a sales question. The W3C quick reference for WCAG 2.2 gives workable benchmarks: text contrast, the size of tap targets, and text alternatives for images.
The most expensive mistake is baking decisive words into the picture in pale grey on white. That text cannot be read in sunlight, cannot be read on a small screen, and does not exist as text at all, neither for the platform's own search nor for a screen reader.
The rule that follows is simple: anything that must be found or read aloud belongs in the text fields of the listing, which means the title, the attributes and the description. Infographic overlays can repeat that text and help a buyer grasp it faster, but they cannot replace it.
PDFs deserve a separate note, because a persistent myth lives here. Search engines do index the text inside PDF files, so a size chart published that way is not lost to search. It is still awkward to open on a phone, which is why the measurements belong in the listing too.
Images: what a shoot gives you and what generation does not
The first image is the most valuable real estate you own on a marketplace, and it is exactly where sellers economise. They post the supplier frame that already appears in a hundred other listings, or a scene where the product occupies a quarter of the frame and disappears inside the tile.
The temptation to replace a shoot with generation is understandable, and it carries a cost that is easy to miss. Copyright protection rests on human authorship. An image produced entirely by a model may leave nobody holding an exclusive right, which means you have no instrument against a copycat either.
There is a commercial side as well. The buyer came for a specific object with specific seams, ports and shade. A generated image that does not match what arrives in the box works as a promise you cannot back, and it comes back to you in the form of returns.
This is why our packages keep photography outside the price, plainly. Listing Refresh at $70 does not include photography, ad spend or account management; the Marketplace + SEO System at $110 does not include photography or paid placement. Work on text and data does not replace a camera.
How to run the diagnosis yourself in a week
Start with a sheet of three columns: impressions, clicks, orders. Fill it in for every listing over one identical period. A single look at that sheet is usually enough to see where each product stops, and how many of your products stop in exactly the same place.
Then act on the first zero from the left. Zero impressions sends you to category, attributes, status and stock. Zero clicks sends you into the results page with a buyer's eyes, looking at your tile among its neighbours. Zero orders sends you through the whole page as somebody about to pay.
Change one thing at a time and write down the date. If you rewrite the title, swap the first photo and fill in ten attributes on the same day, you will get a result but not a reason, and you will not be able to repeat that success on the next product.
Finally, let the change work. Listing edits do not answer instantly: impressions accumulate, the platform recalculates its data, and buyers arrive at different hours of the day. Judging a week by its first two hours is a dependable way to cancel the thing that had just started working.
What VITON13 does at each stage, and what it costs
When the stage is unknown, everything starts with diagnosis. The Demand Audit costs $50, takes 1-2 working days and includes one round of revisions on the findings document. Access to your seller account is not part of it, so we work with the data you choose to show us.
When the stage is known and it lives in the tile or on the page, Listing Refresh runs $70 over 2-3 working days with one round of revisions. If both the listings and the copy on your own site need work, the Marketplace + SEO System is $110 over 3-5 working days with two rounds on listings and copy.
When the calendar is tight there is Marketplace + SEO Express: $190, 3 working days, one round of revisions, with photography and ad spend excluded. When the work is not a one-off there is Managed Demand Growth at $200/mo on a monthly cycle with 30 days notice to stop; advertising spend on the marketplace is not included.
What each package contains, how long it takes and what it leaves out is listed on /services/marketplaces. The page is worth opening after the three-column sheet is filled in: then you are buying a fix for a named stage rather than a general improvement to something unspecified.
Practical checklist
- Export impressions, clicks and orders for every listing over one identical period.
- Search the marketplace for your listing using its exact full title from the account.
- Check the category, required attributes, stock level and moderation status.
- Look at your own tile in the results page beside five neighbouring ones.
- Answer sizing, contents, delivery windows and return terms inside the description.
- Change one element at a time and record the date of each edit.
Questions and answers
Where do I start if there are no sales at all?
With the impressions report. Until you know whether the listing was shown, you do not know what to fix. Zero impressions and zero orders after hundreds of impressions are different problems with different costs.
What does it cost to find out which stage is losing the sales?
The Demand Audit is $50 and takes 1-2 working days, with one round of revisions on the findings document. Access to your seller account is not included in it.
Is it true that price is usually the reason?
Price explains part of the loss on the results screen, where you are compared with your neighbours. It does not explain zero impressions, and it never answers a buyer's question about sizing or contents.
Will Schema.org markup lift my listing in marketplace search?
No. A platform's internal search reads its own catalogue fields. Product markup lives on your own website and addresses external search engines.
What if the result is needed quickly?
Marketplace + SEO Express runs $190 over 3 working days with one round of revisions, photography and ad spend excluded. It pays off best when the stage is already identified and the rewrite has a defined target.