Marketplace Promotion: A Systematic Guide to Growing Visibility on WB, Ozon and Yandex Market

Break any sale on Wildberries or Ozon down into atoms and you get the same chain every time: the buyer saw the card, clicked, added it to the cart, paid, picked it up at the pickup point, left a review. Every step is a signal to the marketplace algorithm: "people buy this product for this keyword and do not return it, so it can be shown more often".

Marketplace promotion means working the entire chain at once. Not one ad account, not one "creative", not one "batch of buyouts". All the touchpoints simultaneously. Otherwise the loop never closes and growth does not stick.

This is not a "take-off in 3 days". It is project-based management of card visibility over 60–90 days, with metrics, reporting and financial documents. Below is a guide to how that work is structured: what the closed loop of visibility is, which modules the MPO approach contains, how Wildberries, Ozon and Yandex Market differ, how much promoting a product card costs, and when you do not need it at all.

The guide is written for sellers with revenue from 5M ₽/mo, marketplace agencies and in-house e-commerce teams at brands. If you are taking your first steps on a marketplace, start with the article on visibility and come back here when you want to build a system.

The standard scenario sellers come to us with:

1. 1The card is not seen. For priority keywords the product sits at position 60 and below. Buyers physically never reach it: they go to one of the first ten competitors.
2. 2Not seen means not bought. Low sales volume for that specific keyword.
3. 3The algorithm treats the card as weak. Forecast revenue is low, and it is not profitable for the marketplace to give it expensive impressions.
4. 4Ads try to compensate. ad spend share (DRR) climbs, margin shrinks. Turn the bid off and sales fall.

That is the closed loop of visibility: low visibility → low sales → even lower visibility. Next come OOS failures, attempts to catch up with competitors using promo codes, and the card slowly sliding into the "sandbox", from which it does not return without intervention.

The main mistake is trying to break the loop with a single lever. "Let's raise the bid" — ads eat the margin. "Let's run buyouts by SKU" — the card jumps for a day and rolls back. "Let's rewrite the SEO" — words were added, but nobody buys through them.

The algorithm needs evidence: that people really search these keywords, click, buy and do not return the product. That evidence is built across several signals at once — and that is the essence of systematic promotion. One module without the others gives a limited effect.

Ads only

The Wildberries and Ozon algorithms rank on more than the bid. If a card has weak intermediate conversions — low CTR in search results, few cart adds, a high bounce share — the bid will not pull positions up. Money goes on impressions to people who do not buy. DRR climbs, margin shrinks. For every cluster the marketplace has a ceiling for a weak card: ads will not push it past that point even at the maximum bid.

Buyouts by SKU only

Most buyout services work by article number or a direct link. The algorithm sees the purchase but does not tie it to a specific search query — the card rises for a day or two on "general weight" and rolls back. The effect is short-lived and growth is not locked in by organic. Money spent, no positions gained.

We covered this scenario in more detail in the "Buyouts 2.0" guide — it also explains why brand and color filters "outweigh" a purchase and reassign it to someone else's keywords.

SEO only

The description is rewritten, keywords added. Two days later indexing picks up the new words — and that is it. If no real purchases come through those words, the algorithm treats them as an informational signal rather than proven demand. The card shows up around position 80 and stays there.

SEO is a necessary condition, but not a sufficient one. The words have to be in the card, otherwise the product is not shown for them at all. But without behavioral support and purchases through the right keywords, new words do not move positions.

Reviews only

A starter pool of reviews and photos really does lift CTR in search results — by 10–25% in our measurements. But if nobody sees the card, nobody reads the reviews. Reviews strengthen conversion, but only once visibility exists. Visibility first, reviews second, not the other way around.

Bloggers and giveaways only

External traffic lands on the card bypassing search results. The algorithm sees the purchase but ties it to the traffic source rather than to a search keyword. That does little for "organic growth on the phrase 'baby crib'". Bloggers and giveaways are a separate channel; their place is as an addition to promotion, not a replacement.

Design only

Strong visuals lift CTR in search results. If people see the card — yes. If they do not, no design will save it. Sellers have come to us with infographics redone for 200k ₽ that got no impressions at all: the product sat at position 90 and not a single new buyer ever reached it to appreciate the new creative.

MPO stands for Marketplace Presence Optimization — systematic optimization of a product's presence on a marketplace. We use the term to separate promotion-as-a-system from "fake activity" and "buyout farms".

MPO has eight modules. They only work in combination: one without the others is exactly the "ads only / buyouts only" mistake described above.

01 · Audit — diagnostics at the entry point

Every project starts with an audit. Without one, strategy is guesswork.

• Which keywords the product is seen for and which it is not (share of results within the cluster).
• Where you and your competitors stand on rating, how many reviews there are and in what format.
• Which warehouses your competitor uses and you do not.
• Where exactly growth breaks: in SEO, in CTR, in the buyout rate, in OOS, in reviews.

The audit produces a map of keyword gaps — clusters where competitors have revenue and you have none. Without that map, launching buyouts is pointless: there is no way to know where to aim.

02 · SEO — semantics and descriptions

"First we collect every query buyers type in" sounds obvious, but few actually do it. Most copy keywords from competitors and inherit their mistakes along with them.

The SEO block of MPO:

• We collect real buyer queries, not "words off the top of our head" and not "a copy of the top 3".
• We cluster them by meaning and by search results.
• We clean out word forms, duplicates, and keywords with poor economics and low conversion.
• We prepare the description for the platform's algorithm (WB and Ozon have different rules).
• We update it regularly for seasonality and newly appearing queries.

Marketplace SEO promotion does not work on an "update once a year and forget" basis. Buyer queries shift in waves — especially ahead of seasons and holidays and when new brands appear.

03 · Visibility — managing impressions

This module answers the question "where to direct the effort". It covers:

• Share of impressions in the target clusters — where it is now, where we want it, how to get there.
• Work with search results and shelf placements (different scenarios on Ozon and Wildberries).
• The CTR / impressions / conversion balance: growing one without the others creates a skew.

04 · Behavior — behavioral signals

Behavioral signals are everything a buyer does with a card: clicks through, adds to cart, orders, picks up, saves to favorites, asks questions, leaves a review.

The algorithm multiplies these conversions together and gets a forecast of future revenue. More detail is in the article "Micro-differences, macro-results".

Within MPO we:

• Control the intermediate conversions (click → cart → order → pickup).
• Do not break them with shock waves — that is the platforms' main buyout detector.
• Close the funnel through to the target action in the required clusters.

05 · Boost — buyouts by keyword, not by SKU

The fundamental difference: the algorithm sees a classic "buyout by article number" as a purchase, but does not connect it to a search query. The purchase has to be made through a specific keyword, via search, without filters that "outweigh" the query and reassign it to another cluster.

Buyout pace goes in waves, without suspicious spikes. The "2-4-8-16-32" ladder kills a card: the marketplace sees conversions jumping up and then collapsing the moment buyouts stop. That is a clear signal that "this activity was not real".

A plateau instead of a ladder: growth, hold and a gradual taper of buyouts offset by growth in organic. The card looks steadily growing rather than "pumped up and abandoned".

06 · Trust — reviews and rating

• A starter pool of reviews for new SKUs.
• Pushing the negative mass down (not "deletion" but displacement with fresh, realistic reviews).
• Maintaining the rating over the long run.
• A natural flow with no shock waves: 100 reviews in a week on a card that normally gets one review per 100 purchases is a red flag.

Details on review safety and calculating the effective cost of a review are in the guide to safe buyouts.

07 · Ads — advertising economics

Ads in MPO are not a separate "bid inflation" service but part of the visibility strategy:

• Keywords where CPM is high but growing visibility will do more than raising the bid.
• Lowering CPM/DRR through organic growth: the higher the organic position, the smaller the bid needed to keep the paid one above it.
• Redistributing budget between campaigns on the principle of "where do we invest to get cheaper organic later".

08 · Identity — the security layer

This is about aged buyer accounts — a long life cycle, unique payment logic and realistic behavior patterns. Not disposable "SIM cards from the Sadovod market".

This is the part we do not disclose publicly, because it is part of protecting the client from random marketplace checks. It is enough to say that by 2026 the algorithms have learned to tell real buyers from farms, and any attempt to save money on Identity ends in fines and a rollback in positions.

Wildberries promotion

Wildberries is the largest platform in Russia. The toughest competition is in FMCG, cosmetics and apparel. The key specifics:

• Huge search traffic. Share of search results decides almost everything. If a card is not on the first pages, it does not exist.
• Price decides. Most buyers filter by price and discount. A card at the niche median price has the advantage.
• A harsh reaction to OOS. Go to zero stock for 3–5 days and positions drop by dozens.
• FBO gives a boost on delivery speed and often on ranking: the marketplace prefers products it delivers itself.
• The algorithm forgets fast. If a card has no sales for a keyword within a month, the cluster is "reset".

The "turnkey Wildberries promotion" service, as we understand it, is the full set of MPO modules: audit, SEO, buyouts, reviews, conversion control, ads. Growing a card on Wildberries is not done with one tool — every attempt to "grow on buyouts alone" gives a short-term jump after which the card returns to where it was. A deep dive into WB specifics is in the hub guide "Wildberries promotion": the three-signal combination, 6 MPO modules in a WB context, a 90-day plan for a new product and 5 typical scenarios.

Ozon promotion

Ozon is the second largest platform. The differences:

• A strong review factor. More often than WB, Ozon promotes cards with photo and video reviews, especially in home goods and kids' categories.
• Promo codes under suspicion. Ozon checks buyouts with promo codes first: the marketplace loses commission on them and has every reason to find them.
• rFBS — savings with a downside. You pay less commission but attract more algorithmic attention to every purchase. For promotion it is not the best option.
• "Behavioral signals" and Trafarety. Ozon has its own amplification mechanics that, in the right hands, act as a multiplier on an MPO strategy.
• Ozon SEO is built differently. Descriptions are longer, the category tree is deeper, filters affect indexing.

Ozon SEO promotion is a separate discipline: user queries on Ozon differ from WB even within the same category, and copying keywords between platforms is a common and expensive mistake. A deep dive into Ozon mechanics is in the hub guide "Ozon promotion": the "bid ↔ organic" link, recommendation shelves as a separate channel, and a 90-day plan for cutting DRR.

Yandex Market promotion

For most of our clients Yandex Market accounts for 5–10% of turnover, and it is often ignored. That is a mistake — on Yandex Market:

• Competition is cheaper in niches where WB/Ozon are already built out.
• Yandex external SEO matters more: a product can arrive from regular Yandex search.
• Fewer buyouts are needed for a meaningful shift — the threshold for the top 10 is lower.
• Behavioral signals are more pleasant to work with: less noise, cleaner data.

For most sellers, Yandex Market promotion is the "third step". First we pull up WB, then Ozon, then we scale revenue through Yandex Market, where competition is softer.

What all three have in common

• Algorithms rank on a combination of signals, not on a single factor.
• Purchases through a keyword matter more than purchases "in general" — on any platform.
• Signal stability matters more than signal peaks: +5% organic every week beats +50% one week and -40% the next.
• OOS wipes out accumulated signals — everywhere.
• Ads work as a multiplier on visibility, not as its source.

These two services are often confused. The short distinction:

Promotion is growth in card visibility. The focus is on growing positions, CTR, conversions and revenue by keyword. It is about results.

Account management for a marketplace store / seller account is operational support: uploading cards, handling buyer questions, working with categories, supply, dealing with fines, routine actions inside the seller account. It is about process.

They complement each other. Most of our clients take promotion, but 30% take it alongside account management: either they have no in-house team for operations, or the team has no time.

What our Wildberries / Ozon account management includes:

• A dedicated manager on the project and regular calls.
• Uploading and maintaining product cards.
• Stock and supply control.
• Handling buyer questions.
• Working with platform promotions and tariffs.
• Managing fines and disputes.

If you have an in-house marketplace manager, take promotion only. If your team is small or non-existent, promotion and account management make sense as a package.

Every promotion project we run follows a three-phase plan. It is not "magic" but normal project work, like with any contractor on a large task.

Phase 1 — preparation (T+0 → T+2 weeks)

T+0. A free card review or mini-audit. Aligning on goals: what exactly we are trying to do (grow positions? cut DRR? recover after OOS? launch a new product?). The rest of the strategy follows from the goal.

T+1 week. A deep MPO audit: the keyword-gap map, a list of priority clusters, an understanding of what is blocking growth right now. The output is a plan for the next 90 days with priorities and expected effects.

T+2 weeks. A finished SEO description built for the marketplace algorithm. A buyout strategy: what we do on low-frequency queries, what on mid-frequency ones, and when we bring in the high-frequency ones. Initial review work: either a starter pool or displacement of negatives.

Phase 2 — going live (T+3 → T+30 weeks)

T+3–4 weeks. Launch:

• Buyouts through priority keywords and shelves.
• The behavioral layer: CTR in search results, cart adds, favorites.
• A starter pool of reviews or displacement of negatives.
• The first status call with the client's team: what is done, what is in progress.

T+30 days. First signals:

• Position dynamics in priority clusters.
• Control of CTR and intermediate conversions.
• Adjustment: which cluster we push next, which we postpone, which we drop.

By this point the card usually moves from 60+ to position 25–40 in the priority clusters. That is not the "top 10" yet, but it is the first sales through new keywords.

Phase 3 — consolidation (T+30 → T+90 days)

T+60 days. We check the main question — has organic picked up. This is the key checkpoint: if organic purchases in the priority clusters happen without buyout support, the effect has locked in. If not, we adjust the strategy or change priorities.

T+90 days. Steady-state mode:

• A stable flow of reviews, behavioral layer and buyouts.
• Reduced advertising load thanks to organic.
• Regular reporting on DRR, revenue and organic share.
• A decision on what comes next: more SKUs, new clusters, new platforms.

Promotion makes no sense over less than 3 months. Anything shorter is "giving it a try"; in that time the algorithm does not have time to trust the card, and the effect does not lock in.

A direct answer to the popular queries "how much does Wildberries promotion cost" and "the cost of Ozon promotion" — it depends on three parameters: the product price, the search volume of the priority keywords, and the current position.

Individual SKUs

The budget for the whole project

It is calculated as the cost of one buyout × the number required + the cost of goods + SEO and analytics services.

Simple logic: if a competitor in the top 10 for a keyword does X revenue per month, we need a comparable number of purchases through that same keyword for the algorithm to believe in our product.

Roughly how many buyouts are needed: 5–15 on a low-frequency keyword, 30–80 on a mid-frequency one, 100–300+ on a high-frequency one. This is very rough and depends on the niche, the product price and the current position.

Free entry points

If it is unclear where to start, we have free formats:

• A free consultation on your card.
• A free card review based on an AI audit.
• Keyword selection.
• An SKU mini-audit.
• A growth map for a product line.

After those it is clear whether you need full-service promotion or targeted work on a few clusters is enough.

Case 1. Cosmetics: 100k ₽/mo → 1M ₽/mo

Before. An SKU in a mid-sized WB category. Revenue 100k ₽/mo. DRR around 25%. Positions for priority keywords: 60+. Several failed promotion attempts through cheap buyout services: a fast jump and a rollback within 3 days.

Hypothesis. The wrong keywords were chosen (taken from the top 3 competitors along with their mistakes). Buyouts ran by SKU with no cluster targeting. The behavioral layer was never connected.

What we did.

• A full MPO audit — we found 12 clusters where competitors had revenue and this card had zero impressions.
• Rewrote the SEO around real demand, removed 30+ duplicates and keywords with poor economics.
• Launched buyouts in low- and mid-frequency waves with a behavioral layer (CTR, cart, favorites).
• Covered 11 negative first-page reviews with 8 fresh, realistic ones.
• In week three we added shelf buyouts.

Result after 60 days. Positions in the priority clusters 100+ → 25th. Revenue 100k ₽ → 1M ₽/mo. Ad spend at the same level.

Consolidation. By day 90 organic had picked up. Paid sales account for 30% of turnover, the rest comes from organic.

Case 2. A line of 5 SKUs: linked set 2.2M ₽ → 7M ₽

Before. A line of related products in one niche, revenue across the set 2.2M ₽/mo. One SKU carried it, the rest lagged. Ads were spread across all of them — overall DRR was climbing.

What we did.

• Split the semantics between the SKUs: each got its own priority cluster so the cards would not cannibalize each other in search results.
• Coordinated buyouts and reviews across the whole line: a cascade strategy — first the "anchor" SKU, then pulling the rest up into the freed-up impressions.
• Redistributed the ads: bids removed from overlapping keywords, added on the long tail.

Result after 90 days. Revenue across the set 2.2M ₽ → 7M ₽/mo. The line stopped "competing with itself" for the same queries.

Case 3. Cutting DRR without losing revenue

Before. An Ozon seller with DRR above 30%. Ads kept sales "on a drip" — turn the campaigns off and sales collapse to zero.

What we did.

• Analyzed the ad campaign structure — found 7 clusters where CPM was high but organic was close by (positions 11–20).
• Launched buyouts and a behavioral layer in those clusters.
• Two months later organic in 5 of the 7 clusters reached the first page — where only paid ads used to hold position.

Result. The ad budget stayed the same. Revenue grew 40% thanks to organic. DRR fell from 30% to 18%.

The formula of an honest case: visibility was low → we found the gaps in the semantics → we strengthened the right clusters → positions rose → organic picked up → revenue grew. Not "we ran buyouts and here is the result".

What we measure — and what we show the client.

Visibility

• Share of search results in the priority clusters — the main indicator.
• Position dynamics for the priority keywords.
• Impression volume by cluster per week.

Behavioral signals

• CTR in search results.
• Click → cart conversion.
• Cart → order conversion.
• Order → pickup conversion.
• Share of adds to favorites.

Economics

• Organic revenue vs paid revenue.
• DRR, CPM.
• Organic share of total turnover.
• Revenue gain vs project cost.

Reputation

• Card rating.
• Review growth by format (text / photo / video).
• Share of negative reviews on the first page.

What the client sees

• A dashboard with the key metrics — updated several times a day.
• Weekly status calls focused on trends.
• Monthly reports: what was done, what produced an effect, what is in progress.
• Purchase receipts, review IDs and shipment IDs (SRIDs) — for the finance and compliance side.
• Analytics "through the buyer's eyes": what an ordinary buyer sees when they type in a keyword.

"Analytics through the buyer's eyes several times a day" is, in our experience, the main difference from agencies that send a report at the end of the month. A marketplace is an environment where search results, competitor prices, stock and discounts change within a week. Once a month is too late.

Trusty does not sell rule-breaking. We sell management of signal credibility.

How we reduce risk:

• Aged buyer accounts — a long life cycle, realistic scenarios and unique payment logic.
• No shock waves — pace in waves, without suspicious conversion spikes.
• No promo codes and no cheap schemes.
• Distribution across pickup points and regions — no concentration on the same locations.
• Deviation analytics — we watch the risk indicators several times a day.

What we do not do:

• We do not use disposable accounts.
• We do not work with suspicious payment patterns.
• We do not run templated review waves.
• We do not promise to "trick the algorithm".
• We do not disclose the technical details of the Identity layer publicly — that is part of protecting the client, not a marketing secret.

We covered how to do this safely in more detail in the guide to safe buyouts — it also covers the effective cost of a review and protection from dishonest contractors.

"It does not work — we tried it." Most often the problem is not the tool but the clusters, the goal and the distribution. If you hit the SKU without SEO and without controlling intermediate conversions, growth does not lock into organic. Show us the project history and we will tell you where the mistake was.

"It is risky, there are fines." The risk comes from primitive schemes: weak disposable accounts, templated scenarios, concentration on the same pickup points, shock waves of conversions. Trusty works on a white-hat model — aged buyer accounts, realistic scenarios, no suspicious patterns. In 7 years of work we have had no systematic fines on client accounts.

"It is expensive." What you should count is the price of the result, not the price of a single action. A cheap contractor often costs more in practice: a wasted budget, lost visibility, sanction risk, and recovery from the mistakes. A free consultation will help you work it out in advance.

"We do giveaways and bloggers — why would we need anything else?" External traffic is a separate channel, and it does not always convert into visibility for search keywords inside the marketplace. Giveaways and bloggers are an addition to promotion, not a replacement.

"Can we do buyouts only?" You can, but without SEO and control of intermediate conversions the effect rarely locks into organic. At a minimum we recommend keyword selection before the start, so that buyouts run through the right queries.

"How long until results?" First signals on day 30. Consolidation by day 60–90. A minimum of 3 months. If someone promises "top 10 in 2 weeks", it is either a buyout farm (it will roll back) or a lie.

"Do you guarantee growth on a specific SKU?" Guaranteeing specific numbers on someone else's product is dishonest. What we do guarantee is clean execution, transparent reporting, purchases tied to the right clusters, regular strategy adjustments and a return to the conversation if something is not working, instead of "service delivered, goodbye".

"What if our product or our batch is bad?" We will tell you honestly at the consultation. A bad product cannot be saved by SEO, behavioral signals or buyouts. In those cases we fix the batch problem before launching promotion.

"How do you work with agencies?" Often as a white-label subcontractor. The agency owns the client relationship, we own execution. Transparent reports, receipts, SLAs, contracts. The agency gets a result it is not embarrassed to show its client or its director.

"Are we a fit for you?" The minimum is turnover from 2M ₽/mo. Ideally from 5M ₽/mo and up. Below that the project economics do not add up: on small revenue the cost of buyouts and services eats the margin faster than we can accelerate the card.

"Which niches do you work with?" FMCG, household chemicals, cosmetics, apparel, footwear, accessories, electronics, small appliances, DIY, kids' products. Adult and regulated categories are handled case by case and discussed separately.

"What is the difference between your WB and Ozon promotion?" Semantics are collected separately for each platform — queries and buyer behavior differ. SEO descriptions are written for each platform's rules. The buyout strategy accounts for Ozon being stricter about promo codes and rFBS, while WB is more sensitive to OOS.

"How much does full-service marketplace promotion cost?" It is calculated from the purchase budget plus the cost of services. For a small project, from 150–200k ₽ for the first 30 days. For a large enterprise project, from 500k ₽/mo. At the start we walk you through the project economics at the consultation: what we invest and what return we expect.

A few cases where we honestly say "not now" or "not us":

• A card with 100% visibility on the right keywords. There is little room for buyouts to add growth. Better to work on unit economics and range expansion.
• A product with systemic problems: poor quality, a defective batch, constant returns. Fix the product first, promote later.
• Turnover below 2M ₽/mo. The project economics do not add up.
• You need a one-off delivery of "20 reviews" with no strategy. That is not our story — take a simpler service.
• A fully regulated category (medicine, supplements, weapons, special machinery). Discussed separately — sometimes we take it on, sometimes we honestly decline.
• You sell only through external traffic (bloggers, direct links, marketing traffic). MPO is not for you — you have no need to grow inside marketplace search.

• "Micro-differences, macro-results: what marketplace algorithms hide" — on how visibility works and on behavioral signals. The foundation without which nothing else works.
• "Buyouts 2.0" — a full guide to buyout strategies on WB, Ozon and Yandex Market. Stages, budgets and the "plateau" pattern.
• "Safe buyouts" — on marketplace detection markers, aged buyer accounts, the effective cost of a review and protection from dishonest contractors.

If the task is only taking shape, take a free card review. We will show your share of search results in the priority clusters, find the keyword gaps and tell you whether the growth target is realistic. No obligations: roughly a third of these reviews end with "you do not need this right now".

If the strategy is clear and you need someone to execute, take a look at how turnkey card promotion works and get in touch for a consultation. At the start we ask about turnover, niches, current DRR and the type of partnership (brand / agency / seller). The output of the consultation is a 90-day promotion plan with economics and priorities.

If you are choosing a contractor, we have compiled a 2026 ranking of marketplace agencies with the evaluation methodology.