Paid promotion is the single largest operational lever a foreign fashion brand has in the first six months on Russian marketplaces. It is also the single largest source of avoidable losses — first-cycle brands routinely burn 30–60% of their ad budget on auctions they shouldn’t be competing in, against ranking signals they haven’t yet built. This article walks through how the three platforms actually allocate paid placement in 2026, what the CPC dynamics look like across the year, and the operational allocation framework that compounds rather than evaporates.
TL;DR
- Ozon Performance (CPC search and product ads) runs a second-price auction. CPC climbs 2–3× in peak weeks (11.11, Black Friday) as every promo-eligible ad fights for the same inventory; uplift in buyer attention does not scale with the CPC climb.
- Wildberries WB Продвижение uses a similar auction with less transparency — bid recommendations from the platform are conservatively useful but should not be treated as ground truth.
- Yandex Market Буст продаж is the auction inside Бестселлеры — bids set as % of item price (5–15% typical) or % of total sales (8–20% typical). The compounding ranking signal pays back over months.
- For a new entrant, the working budget split is 25% S-peaks (defensive) / 40% B-tier always-on (compounding) / 20% A-tier holidays / 15% C-tier category days. The opposite of what most brands do in their first cycle (over-spending on S-peaks).
- The single most expensive first-cycle mistake is participating in Auto-Promotion / Auto-Add on Wildberries without setting a minimum price floor. Items get auto-added at deep discounts and the seller pays the ad cost on top.
1. The three engines — quick comparison
Each marketplace runs its paid promotion through a distinct engine, with different mechanics, different bid units, and different reporting depth:
| Ozon Performance | WB Продвижение | Yandex Market Буст продаж | |
|---|---|---|---|
| Bid unit | CPC (₽ per click) | CPM-equivalent + per-action | % of item price OR % of total sales |
| Auction type | Second-price | Auction (less documented) | Second-price |
| Transparency | Highest — granular metrics | Lowest — bid recommendations + outcomes | Medium — case data + auction recommendations |
| Auto-features | Optional, opt-in | Auto-promotion + Auto-add by default | Auto-participation via «Минимум для акций» |
| Compounding mechanic | Elastic Boosting (search rank from promo fulfillment) | Тематические подборки (always-on themed promos) | Бестселлеры (always-on flagship) |
| CPC range | ₽3–25 typical, peaks ₽40+ | platform doesn’t expose; ₽5–35 estimate, peaks ₽60+ | embedded in % bid |
Foreign brands tend to spend the most operational time learning Ozon Performance (most transparent), then add WB Продвижение, then YM Бустер. The recommended budget allocation does not follow that learning order.
2. Ozon Performance — auction mechanics
Ozon Performance is a CPC-based search and product placement auction. It’s the most transparent paid surface in Russian marketplaces and the most usable for foreign brands learning the system.
How the auction works
For each search query and product surface, Ozon runs a second-price auction with:
- Your bid (the maximum CPC you’d pay)
- Quality score (relevance, CTR history, fulfillment metrics on the boosted item)
- Auction depth (how many other sellers are bidding)
The winning ad pays just above the second-highest bid, not their max bid. This makes overbidding less catastrophic than in first-price auctions, but the quality score component means a low-history brand pays more for the same placement than an established competitor.
CPC dynamics across the year
CPC follows the promo calendar more than overall search demand. The pattern, indexed to a normal week:
| Week | CPC index | Why |
|---|---|---|
| Normal week (non-promo) | 100 (baseline) | |
| Pre-promo warmup (1 week before S-peak) | 130–160 | Sellers ramping budgets |
| S-peak week (11.11, Black Friday) | 200–300 | Every eligible ad fights for same inventory |
| Post-peak week | 110–130 | Settles fast |
| A-tier holiday week (Mar 8, May holidays) | 140–180 | Less competition than S-peaks |
Practical implication: budget allocation should not match this curve. If you spend more during S-peaks (as most new entrants instinct), you’re paying 2–3× CPC for buyers who would have bought from you anyway. Spend more on always-on B-tier mechanics where CPC is stable at baseline and you build ranking that compounds over months.
Elastic Boosting — Ozon’s 2026 ranking lever
Ozon introduced Elastic Boosting in 2025 — search ranking is now tied to your daily fulfillment of the promo sales share target. 50% is the floor; 100% gets the maximum search boost. This means:
- If you participate in a promo and fulfill the threshold, your organic search rank gets a boost (multi-week duration).
- If you participate but underdeliver on fulfillment, your rank takes a hit.
- The right way to use this is: participate in fewer promos but commit your inventory to hitting the threshold. Spread-thin promo participation is worse than focused participation.
Operator recommendations
- Don’t outbid for 11.11. Set Ozon Performance budget caps conservatively (50% of what you’d spend on a non-peak week). Use the savings on always-on Weekly Discounts mechanics.
- Build quality score before peak weeks. Run sustained always-on Performance for 4–8 weeks before a target peak so your quality score (CTR, conversion, ranking) is built. Last-minute bid increases without quality history are 3–4× more expensive.
- Use the bid recommendations as the floor, not the ceiling. Ozon’s recommendations are conservative — start at the recommendation and only raise after observing CTR.
3. WB Продвижение — auction mechanics
Wildberries’ paid promotion engine is less transparent than Ozon’s. The mechanics:
Auction structure
- Bids are submitted but the platform doesn’t publish detailed second-price calculations
- Bid recommendations are surfaced for each campaign but the rationale isn’t documented
- Outcomes show impressions, clicks, orders — but attribution clarity is lower than Ozon
Auto-Promotion and Auto-Add — the trap
Wildberries auto-enrolls items in promotions when their price crosses a category threshold (the «минимальная цена» / price floor). This is convenient for the platform — it ensures supply for promo windows — and expensive for the seller when:
- You don’t realize you’re auto-enrolled
- The discount required to qualify is deeper than your unit economic supports
- You pay paid promotion costs on top of the auto-applied discount
The single most expensive first-cycle mistake for foreign brands on WB is leaving auto-enrollment on without setting a minimum-price floor that protects your unit economic. We have seen brands burn 40–60% of their first quarter’s gross margin on auto-enrolled promos they didn’t actively choose.
Operator recommendations
- Set minimum-price floor on every SKU before the first listing goes live. Block auto-add to any promo that would discount below your minimum.
- WB Продвижение spend should never exceed 8–10% of GMV on basics, 10–12% on mid-market. The platform’s recommendations sometimes push 15%+; treat these as the upper bound, not the target.
- Pre-funded promo participation: when you opt into a major WB promo window, pre-load inventory and pre-fund the ad budget for the duration. Mid-promo bid changes underperform.
4. Yandex Market Бустер продаж — the auction inside Бестселлеры
Yandex Market’s paid promotion is structured differently. The Бестселлеры flagship mechanic runs always-on, and Бустер продаж is the auction layer inside it that determines placement within boost blocks.
How the bid works
You set the bid as either:
- % of item price (typical 5–15%) — pays per displayed item
- % of total sales (typical 8–20%) — pays per attributed conversion
The «ценовая стратегия» tool
Yandex Market also lets you opt into the «ценовая стратегия» (price strategy), where the platform lowers your displayed price by up to 30% using part of your bid. Items priced via this strategy can auto-qualify for Бестселлеры participation.
The math: at a 30% price-strategy reduction, you’re trading 30% of margin for higher ranking. Whether this pays back depends on the lift in conversion. From operator data on premium fashion: conversion lift is 1.5–2.5× at the 30% strategy level. For mass-market basics with already-thin margins: 0.8–1.4× lift — not worth it.
The compounding effect
Бестселлеры ranking signal accumulates over multiple waves. Items that perform well in one Бестселлеры wave (new wave every 1–2 weeks) gain ranking signal that carries forward into subsequent waves. By month 4–6, a seller with consistent Бестселлеры participation has built ranking that translates to organic traffic baseline — even between Бестселлеры waves.
This is the most compounding paid promotion mechanic on Russian marketplaces in 2026 and the reason Yandex Market is the “month-6 organic-growth experiment” we recommend in the Playbook.
Operator recommendations
- Start at 8–12% of total sales bid for Бестселлеры participation. Don’t start at the maximum bid — the platform recommendations skew high.
- Run two consecutive waves at the same bid before adjusting. Wave-to-wave performance varies materially; one wave isn’t enough signal.
- Use ценовая стратегия only on SKUs you want to build ranking on. Turn it off for premium SKUs where the 30% margin reduction outweighs the lift.
5. The budget allocation that works for new entrants
For a foreign brand launching on Russian marketplaces with a typical first-cycle budget ($10K–$50K paid promotion):
| Tier | Allocation | Where it goes |
|---|---|---|
| S-peaks (defensive) | 25% | Defensive participation in 11.11 and Black Friday — enough to defend ranking, not enough to compete with established sellers |
| B-tier always-on (compounding) | 40% | Ozon Weekly Discounts, WB Тематические подборки, YM Бестселлеры. Compounding ranking signal. |
| A-tier holiday peaks | 20% | Mar 8, Feb 23, Back to School. Less crowded than S-peaks. |
| C-tier category days | 15% | Category-specific mini-windows (Beauty Days, Sport Days, Children’s Days, etc.) |
This is the inverse of what most first-cycle brands instinct. The instinct is to over-allocate to S-peaks because they’re the highest GMV days — but:
- CPC is 2–3× peak-week vs baseline, so your ROAS is materially worse
- Established sellers’ quality scores beat yours
- Post-peak ranking decay is real — you spent the budget, got the volume, but didn’t build durable ranking
Always-on B-tier participation costs less, compounds longer, and exposes you to the buyer behavior we cover in the consumer behavior article — where mid-market basket-fill happens year-round.
6. The three operational mistakes that burn money
From operator data on first-cycle foreign brands:
Mistake 1 — Auto-enrollment on WB without minimum-price floor
Covered in section 3. Single most expensive mistake. Set the floor before launch.
Mistake 2 — Maxing out CPC for 11.11
The instinct is: it’s the biggest day, spend the most. The reality: CPC at the peak is 2–3× baseline, quality score weights against new sellers, and post-peak ranking decay is fast. The math doesn’t work in your favor.
Spend the same budget on 8 weeks of always-on Weekly Discounts (Ozon) or Тематические подборки (WB) and your year-end position is materially better.
Mistake 3 — Bid based on platform recommendations without observation
Both Ozon and WB surface bid recommendations. Both are conservative — they’re calibrated for the platform’s revenue optimization, not yours. Treat them as the floor and observe CTR / conversion / ROAS at the recommendation before raising. About 60% of recommended bids overpay vs the optimal observed bid in our data.
7. The 90-day paid promotion calendar for a new entrant
For a brand launching on Ozon Global in late August (the FW launch window):
| Week | Activity | Budget allocation |
|---|---|---|
| Week 1–2 (launch) | Enable Performance on launch SKUs; start at conservative bids; observe CTR | 8% of monthly budget |
| Week 3–4 | Bid up on top-performing SKUs; pause underperformers; enroll in Weekly Discounts | 10% of monthly budget |
| Week 5–8 | Establish always-on rhythm; track quality score; participate in 1 category day | 12% of monthly budget |
| Week 9–12 | Pre-funded participation in 11.11 (defensive only, not max bid); maintain always-on | 25% of monthly budget |
| Week 13 (Black Friday) | Defensive participation; budget cap | 15% of monthly budget |
| Week 14+ (post-peak) | Resume always-on; analyze attribution; iterate SKU mix | 10% of monthly budget |
This calendar deliberately understates 11.11 and Black Friday spend. The brands we see succeed in their first year do this consistently.
FAQ
Is Ozon Performance worth it for a new brand? Yes — it’s the most transparent paid surface and the easiest to learn. Start at the platform’s recommended bids, observe CTR / conversion / ROAS for 2–4 weeks before adjusting. Set conservative budget caps to prevent over-allocation to peak weeks.
Why do you recommend less spend on 11.11? CPC climbs 2–3× during the week, your quality score is low as a new seller, and post-peak ranking decay is fast. The math doesn’t favor a new entrant. Spend the budget on always-on B-tier mechanics where CPC is stable and ranking compounds over months.
What is the most expensive WB Продвижение mistake? Auto-enrollment in promotions without a minimum-price floor. Items get auto-added at deep discounts and the seller pays the paid-promotion cost on top of the auto-applied discount. Set a minimum-price floor on every SKU before launch.
How does Yandex Market Бустер продаж differ from Ozon Performance? Bid unit: Бустер uses % of item price or % of total sales; Performance uses CPC. Compounding: Бустер’s ranking signal accumulates across 1-2 week Бестселлеры waves; Performance’s ranking signal is tied to Elastic Boosting (daily fulfillment share). Transparency: Performance is more granular in reporting.
When should I add Yandex Market to my marketplace mix? Month 3–6 after Ozon launch. Reason: organic-growth investments take 4–8 weeks to show measurable lift and 6+ months to compound to ranking baselines. The 3-month gap from your Ozon launch is the right time to start the 6-month Бустер компаундинг curve, so it pays off in months 9–12 when your other channels are mature.
Where this fits in the launch plan
Paid promotion is one of the four operational pillars covered in the Russia Fashion Entry Playbook 2026 alongside compliance/EAC, calendar/seasonality, and consumer behavior. The Pro tier includes the spreadsheet template for modeling paid promotion ROI against the budget allocation framework.
Get the Playbook · Read methodology · Talk to us
Sources
- Ozon Performance auction mechanics and Elastic Boosting (2025–2026) — Ozon seller documentation (seller-edu.ozon.ru).
- Wildberries WB Продвижение mechanics and auto-promotion rules — Wildberries seller portal (seller.wildberries.ru, pro.wildberries.ru).
- Yandex Market Бустер продаж and ценовая стратегия — Yandex Market partner documentation (partner.market.yandex.ru).
- CPC dynamics across promo windows — operator data, 2024–2026 cycles.
- Бестселлеры case data (+2.6× impressions, 1.5–2.5× revenue lift) — Yandex Market published cases, operator validation.
- Budget allocation framework (25/40/20/15) — see Russia Fashion Entry Playbook 2026 — Lite for full S/A/B/C tier framework derivation.