Answer in brief
Order of operations matters more than the sum: a paid impression multiplies whatever the listing already does. What to fix before the first dollar, how to size a test, how to read the report, and what a supplier can honestly promise.
The short answer: fix conversion first, budget second
A listing that does not sell on free impressions will not start selling on paid ones. Advertising multiplies whatever the page already does, so the order of operations decides the outcome long before the size of the budget does.
Put differently, a bid buys you placement next to a shopper's query. It does not buy that shopper's agreement. The decision is made by the main image, the price, the delivery promise and the reviews already sitting on the page.
This is why the question of how much to budget is almost always asked too early. The honest first number is zero, until you have evidence that the listing converts the free traffic it is already receiving.
The rest of this article covers what to repair before the first paid impression, how to size a test you could lose entirely, how to read the report that comes back, and what a supplier can commit to without lying.
What advertising actually buys
Marketplace advertising buys impressions and position. It is a supply of attention, not a supply of persuasion. You are renting shelf space beside a query, and shelf space has never closed a sale on its own.
The first practical consequence: the same bid produces a different cost per order on a weak listing and on a strong one, because an identical number of clicks divides into a different number of purchases.
The second consequence concerns your range. Paid traffic reinforces a product that has already found its buyers, and it burns budget fastest on a product whose demand has not been confirmed by a single order.
The third concerns time. You can buy an impression today. You cannot buy a rating, a review history or a sales record, and that is exactly why paid traffic almost never substitutes for preparation.
Signs a listing is not ready for paid traffic
The first sign is traffic without orders. Shoppers are reaching the page and leaving it, and paid clicks added on top of that will simply buy you a larger number of departures.
The second is that you cannot name the query you want to appear for. Without a stated query a campaign becomes a bid on every word at once, and the report that follows is unreadable.
The third is stock that runs out. An impression paid for at a moment when nothing can be bought loses both the money and the accumulated position, which then has to be earned again from the start.
The fourth is not knowing your margin on a single unit. Without that figure every cost per order looks equally acceptable, and there is nothing to weigh a decision to stop the campaign against.
What to fix before the first dollar of budget
The title. It has to carry the word a buyer uses for the product, not the word your purchasing department uses. A quick check: ask someone outside your category to describe the item in one phrase.
The attributes. Empty specification fields keep a listing out of filters, and a filter is free qualification: an audience that has already narrowed itself down to something close to what you sell.
The main image, which settles the question of a click before any text is read. Worth knowing here: in a number of jurisdictions an image generated entirely by a model, with no human authorship, may not be protected by copyright at all, so nobody holds an exclusive right to it, including you.
The price and the delivery terms. Both are visible in results alongside the photograph, so both take part in the click decision. A bid does not compensate for losing on those two fields at once.
Setting a first budget you can afford to lose
Start with the question rather than the sum. A test budget is the price of an answer: does this product sell, for this query, at this price. That answer has value even when it comes back negative.
The working rule is a sum whose loss would not change what you buy from your supplier next month. It is a personal figure, tied to your turnover, not to any benchmark quoted in an article like this one.
The second rule is to budget in orders rather than in days. A test that ends before a campaign has collected a workable number of clicks produces the appearance of a result rather than an actual one.
The third is to hold a reserve. A first campaign usually exposes a fault in the listing or in the query set, and the reserve exists to rerun the test after the fix, not to extend a losing one.
Designing the first test: few campaigns, one question
Run fewer campaigns than you would like to. Two or three groups with cleanly separated queries can be read afterwards. Ten simultaneous experiments sharing one budget cannot be read at all.
Separate queries by intent. A branded query, a category query and a query describing the shopper's problem behave differently, and averaging them into a single report hides the one difference worth seeing.
Do not change the listing and the bid on the same day. Photographs, price and budget moved together leave you holding a result you cannot attribute to anything in particular afterwards.
Decide in advance which outcome ends the campaign. A stop condition written down before launch is the only real defence against the habit of extending a losing test by one more week.
The report: four numbers worth reading
Impressions and clicks answer a question about visibility and about the cover. Many impressions with few clicks is a conversation about the photograph, the price and the title, not about the bid.
Clicks and orders answer a question about the page itself. Clicks arriving without orders mean the content is not confirming the promise the search result made on its behalf.
Cost per order is compared with your margin, never with a market average. Only that comparison tells you whether you are buying profit or buying turnover at your own expense.
Advertising cost as a share of revenue shows durability. If it holds only during a discount, the campaign stops paying for itself when the discount ends, and that has to be checked before any scaling.
Why the cost per click rises and what that tells you
A marketplace bid sits inside an auction your competitors are bidding in as well. Their behaviour is outside your control, which is precisely why nobody can promise you a stable cost per click.
A seasonal rise is ordinary. Sellers willing to pay more for the same impression enter the category, and a budget plan should assume the price of entry moves rather than holds steady.
When the cost per click rises and conversion falls at the same time, that combination usually points to a stronger competing offer rather than to a setting inside your own campaign.
The honest conclusion is that durable economics come from the gap between your listing and the one beside it, and working on that gap is cheaper than outbidding the auction indefinitely.
What a supplier can honestly promise
A supplier can commit to work and to time: what will be done, in how many working days, and how many rounds of revisions the scope includes. Those are commitments you are able to check.
A supplier cannot commit to a position in results, a number of orders or a cost per click. All of that depends on an auction, on competitors and on the platform's own ranking, meaning on third parties.
Treat phrases about getting you to the top, or a promised revenue figure, with suspicion. They cannot be verified before payment, and after payment there is nothing to argue about, because no criterion was ever fixed.
A useful test when choosing: ask what you physically receive — a document, edited listings, a report — and what happens if the conclusion about the product turns out to be negative.
Where VITON13 fits: packages, prices, and why ad spend is separate
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 part of it.
Listing Refresh is $70 and takes 2-3 working days with one round of revisions. Photography, ad spend and account management are not included in that package.
The Marketplace + SEO System is $110 across 3-5 working days with two rounds on listings and copy; photography and paid placement are excluded. Marketplace + SEO Express is $190 across 3 working days with one round.
Managed Demand Growth is $200 per month on a monthly cycle, with 30 days notice to stop. Advertising spend on the marketplace is not included in any package: you pay the platform directly, which keeps the line between work and media visible.
The unpaid half of demand: structured data and accessibility
Part of your demand arrives from search rather than from the marketplace. Structured data serves that route: the Schema.org vocabulary describes a Product type with properties such as name, image, description, brand and offers.
Markup does not buy a position, but it makes a page legible to the machine parsing it. It is inexpensive work with a long life, unlike an impression that ends the moment the budget does.
Accessibility works the same way. The W3C quick reference for WCAG 2.2 sets out requirements such as text alternatives for images and sufficient contrast, which are exactly the things that help a person and a parser at once.
One practical detail: search engines do index the text inside PDF files. A manual or a specification published as a file takes part in search, so its wording deserves the same care as a page.
A thirty-day sequence
Week one runs on no budget at all. Title, attributes, main image, price, stock availability. The same week produces the list of queries you actually want to appear for.
Week two is a small test across two or three query groups, with the stop condition written down beforehand. Leave the listing untouched during those days, or there will be nothing to interpret.
Week three is reading the report and making one change at a time. Repair the point where the buyer's path breaks first, whether that is the impression, the click or the order, and leave the rest alone.
Week four is the scaling decision. Increase budget only where the cost per order already fits inside the margin without a discount. Everything else goes back to week one.
Practical checklist
- Confirm the listing converts free impressions before launching any campaign.
- Fill every attribute and specification field so the product qualifies for filters.
- Write down the queries you want to appear for before touching any bid.
- Set a test sum whose total loss would not change next month's purchasing.
- Record the stop condition in writing before the campaign goes live.
- Compare cost per order with your per-unit margin before increasing budget.
Questions and answers
Where should a seller start with marketplace advertising?
With conversion, not with budget. If a listing already receives free impressions and produces no orders, paid clicks buy a larger number of exits rather than more purchases. Fix the title, attributes, main image, price and stock first, then write the query list, and only then set a bid.
What does the Demand Audit cost and what is included?
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 part of it: the audit works from the publicly visible demand picture and from the materials you supply.
Is advertising spend included in VITON13 packages?
No. Advertising spend on the marketplace is excluded from every package and paid by you to the platform directly. That keeps the boundary between work and media visible, and it lets you pause impressions at any moment without disturbing the agreement covering the listings themselves.
How much should a first test cost?
As much as you can lose in full without changing what you buy from your supplier next month. It is a personal figure tied to your turnover. What matters more: budget in orders rather than in days, and keep a reserve to rerun the test after the fix.
Can a supplier promise a top position?
No. Position, order volume and cost per click depend on an auction, on competitors and on the platform's ranking, meaning on third parties. What can be promised honestly is scope, a deadline in working days and a number of revision rounds, because those are verifiable.