Advertising on Wildberries and Ozon in 2026: unit economics and rate management

In 2025, Wildberries and Ozon took a total of 68% of all Russian e-commerce - this is not a trend, this is a new retail reality. At the same time, most sellers spend their budget on advertising with a DRR of 40-60%, although a healthy economy is up to 15%. I'm figuring out how to fix this.

I worked with teams of sellers on marketplaces in several niches: household chemicals, children's toys, sportswear. The total turnover under management is about 80 million rubles per year. The numbers in the article are from real WB and Ozon accounts, not from marketing presentations of the platforms themselves.

There is no “marketing” in the classical sense on the marketplace. There is a unit economics and an algorithm. If the card does not convert without advertising, advertising will only speed up the drain of money, not sales.

1. How the WB and Ozon ranking algorithm works

Wildberries ranks based on three groups of factors: conversion (CTR and CR per order), seller rating (repurchase rate, delivery speed, cancellation rate), commercial conditions (discounts, participation in promotions). This is not SEO in the usual sense - there are no links and PageRank. All semantics work through a full text match of the request with the fields of the card.

Ozon adds a “localization index” to this - goods in the warehouse closest to the buyer receive a boost. Therefore, the strategy for distributing balances among regional warehouses directly affects search positions. On WB, the analogue is the FBS scheme (from its own warehouse) vs FBO (from the WB warehouse): FBO, all other things being equal, gives priority in issuing.

2. Card as a conversion asset

A card on the marketplace is a landing page. The buyer does not go to the website, does not read the blog, does not watch YouTube. He decides whether to buy or not in 8-12 seconds based on the photo and the first lines. Therefore, card optimization gives an immediate effect on conversion, which is immediately reflected in positions.

Title. Structure: Brand + Product type + Key characteristics. Example: “Whey protein Whey Optima 1 kg chocolate without sugar.” Without spamming keys separated by commas - the algorithm does not index garbage, and buyers do not read. Maximum 100 characters.

Photo. Minimum 8, optimal 12-15. The first photo is the main conversion asset: white background + product + key selling parameters in the infographic. The second slot is lifestyle, the third is large characteristics, the fourth is packaging. A 15-30 second video is required for equipment and products demonstrating use.

Rich content (WB). Infographics instead of or along with regular photos in “rich content.” According to case studies: +20-35% to CR per order when used correctly. Order from freelance designers with experience specifically on WB - they have their own specific formats.

3. Advertising on Wildberries: auction logic

WB offers several formats: automatic advertising (all placements at once), search advertising (payment for impressions in search), card advertising (displayed next to competitors), branded shelf (banner at the top of the category).

Launch strategy: first, an automatic campaign with a budget of 500-1,000 ₽/day for 14 days. The goal is not sales, but data: what queries bring clicks and orders. After that, a search campaign with manual bids only for converting queries. Non-converting queries are turned into negative keywords.

4. Unit economics: counting before launching advertising

The main mistake new sellers make is launching advertising without calculating unit economics. If the margin before advertising is less than 20%, then with a DRR of 15% you go to zero. With a DRR of 25% at the start, it’s a minus.

Formula for WB: Net margin = Price - Cost - Commission (5-25%) - Logistics (~50 ₽) - Storage - DRR advertising. The commission depends on the category: clothing 15-25%, electronics 5-10%, cosmetics 15-20%. Logistics on WB - 33-75 ₽ per shipment, plus reverse logistics for return ~50 ₽. The normal return in clothing is 50-70%, in electronics - 5-15%.

Minimum margin for working on WB with advertising: 35-40% before advertising costs. Then, with a DRR of 20%, 15-20% of the net margin remains - acceptable. If your margin is 15% before advertising, the marketplace is not your channel until you optimize the purchase price or raise the retail price.

5. Analytics and external traffic

Internal analytics of WB Seller and Ozon Seller is basic. For in-depth work, third-party services are needed: MPStats (market), Sellmonitor (competitors), Mayak (semantics + positions). Cost: 5,000-15,000 ₽/month, pays off already with a turnover of 500,000 ₽/month.

External traffic on WB: UTM parameters in the URL do not work - the marketplace cuts them off. To track, use promotional codes (created in your WB account) or analyze sales surges based on the launch time of an external campaign. On Ozon, external traffic is monitored normally - UTMs are saved, there is a separate report. Read more about UTM in the guide using UTM markup.

Build a connection between external traffic and ROAS on the marketplace through Looker Studio: data from the MP API + data from advertising accounts in one dashboard. It takes a day of setup and saves hours of manual mixing every week.

6. Four mistakes that I see in every second seller

1. Launching advertising without card optimization. CTR 0.5% with an average niche of 3% - money goes down the drain. First, bring the card up to standard using photos, text, and reviews. Only then advertising.

2. High DRR without understanding LTV. A DRR of 40% looks like a disaster, but if 30% of customers come back without advertising, CAC through LTV may be normal. Count your repeat purchases by item before you panic. More details in LTV/CAC benchmarks.

3. Participation in all WB promotions in a row. WB is actively putting pressure on sellers with promotions with 30-50% discounts. Participation is voluntary, but the algorithm is a little pessimistic for non-participants. Solution: Calculate the profitability of each stock in advance. If after the discount the margin goes negative, do not participate, despite pressure from managers.

4. Ignoring stock balances. If you run out of balance, the card goes to the end of the issue. Once items are returned to stock, they will be restored within 2-4 weeks. Plan deliveries with a margin of at least 30 days of sales at the current pace.

7. Summary

Marketplace is not about “posting a product and waiting.” This is an operating system with an algorithm that rewards conversion and penalizes weak cards. Sequence: first unit economics, then card, then organic positions, then advertising. You won’t be able to skip a step - you’ll only waste your budget.

Related materials: unit economics calculator, ROAS/DRR calculator, ROAS benchmarks for e-com, LTV/CAC benchmarks, product description for the marketplace.

If you want to analyze the economics of a specific SKU or set up advertising campaigns, write to Telegram or through form. Starting consultation - 0 ₽.