Full-service management of your business on Russian marketplaces

We rebuilt the unit economics for each of the 800 SKUs and corrected recommended prices — some items were consistently sold at a loss. We replaced the card imagery with AI-generated content and built a custom analytics system for the large assortment. Promotion ran through linked cards while keeping ad share of revenue flat.

A large in-house furniture manufacturer with stock and logistics under control. Advertising was running but producing no result — revenue had been falling for months. No analytics and no view of what the store was actually earning.

1. Rebuilt the unit economics per SKU

   Margin, commission, logistics, advertising, collection rate, turnover and net profit for every SKU. It turned out prices were held as for wholesale buyers, with no link to marketplace economics, and some items were consistently sold at a loss while the store overall stayed positive. We corrected those prices and started from there.

2. Compared the funnels with category leaders

   Detailed analysis showed the funnels were not competitive. Minimal infographics, studio photography that was dated and did not sell.

3. Introduced a new card imagery standard

   Content moved to AI, which made replacement cheaper and faster. We tested several funnel variants, picked the winner and rolled it out across all 800 SKUs.

4. Built custom analytics for the account

   With 800 cards, item-by-item control is impossible. We built our own system so that changes affected the whole account rather than single cards.

5. Promotion through linked cards

   We picked the strongest cards, re-linked them and promoted the best performers — they fed traffic to every card in the group. Once a batch peaked we moved to the next. Sales grew several times over with ad share of revenue kept flat.

6. Launched a new category — TV stands

   Two months of promotion made it the category leader on WB, and it became the second largest category by revenue in the store.

7. Built the client an org structure

   The Mebelson team came to us asking to be shown how this work should be done. We handed over our playbook for structuring a marketplace department.

A launch from zero in an overheated niche. We checked and corrected the card imagery, built up reviews, worked out in detail where competitors take their traffic from, and derived target figures for ranking, traffic and funnel — then promoted against those.

An in-house bar manufacturer with individual terms from Wildberries and product cards ready. Sales had not started yet — we begin from zero.

1. A correct start: funnels and reviews

   We checked and corrected the card imagery, then worked on reviews — points for reviews and undercutting competitors, which the individual terms allow. By month three there were enough reviews for the funnels to match the category leaders.

2. The best analytics we have done

   The goal was to lead a large, competitive niche, so we broke down in detail where competitors source their traffic. From that we derived target figures for ranking, traffic and funnel — and promoted against them.

3. In-platform promotion

   We still apply our strongest advertising playbook here. By the end of 2026 we plan to double again on the tools already in place.

We cut the assortment from 100 to 40 SKUs, dropping the items that were draining budget from the strong cards. We added seasonal shoots and showed each garment on five models of different sizes — the collection rate rose along with the margin.

One of the category leaders with solid sales, but an average margin of 15% and constant stock problems: not enough cash for goods because of a low collection rate. Over 100 SKUs in the assortment. The client stayed with us through 2023 and 2024: back then we grew from 30 to 80 m ₽/mo in three months and deliberately pulled back to 60 m — an eight-day turnover was risky for stock.

1. Reworked the assortment matrix

   Low-margin, low-collection items were draining budget from the profitable ones. The decision was blunt: drop the unprofitable positions — they still produced margin, but there was not enough funding for them. We cut gradually: 100 SKUs in 2023, 40 in 2024. That solved the stock problem, lifted revenue and, most importantly, brought margin to 25%.

2. Set up the card imagery

   Before us the account used only factory photos. We added seasonal shoots: New Year, summer street, studio for the rest of the year. The collection rate improved because each garment was shown on five models of different sizes.

3. In-platform promotion using our own playbook

   The key driver of sales. The account is still under our management and the advertising strategy changes as the platform rules change.

4. Different prices per size

   In early 2024 we were among the first to use this: the buyer sees the lowest price across sizes. CTR rose three to four times.

We picked the niche and the product ourselves: men’s polo shirts in the upper-mid segment with a 40% collection rate. Cards, SEO and the whole image funnel were done entirely with AI — not a single photo shoot.

No product, a store full of dead stock, a wish to find a new niche and lead it, and the money to do it. The short answer to “can you still enter apparel in 2026” is yes — if you keep the collection rate sane.

1. Product sourcing

   One of our strongest areas. We proposed products across several categories; the choice fell on men’s polo shirts: a huge niche where traffic is still growing while the number of sellers is shrinking. We took the upper-mid segment — poor quality never gives a sane collection rate. When the manufacturer showed 40%, we knew it would work.

2. Store launch

   Card creation, SEO, the best possible image funnels entirely on AI — without a single shoot. First reviews collected and promotion started. Production timelines were bumpy, but we almost made the season.

3. In-platform promotion

   When we pick the niche ourselves, promotion is never the problem. We took our place in the upper-mid segment quickly, built traffic and keep growing month on month.

We found the growth points inside the client’s own range: hair supplements, face serums and keratin instead of the overheated masks and shampoos. Promotion in the narrow niches took the brand to the top.

An in-house cosmetics manufacturer, third year on marketplaces, best result 1 m in revenue at the peak of the season. Stuck, with no idea where the growth points were.

1. Found the gems in their own range

   Only 40 SKUs, hard to get lost in — but production was betting on entirely the wrong items: masks, shampoos, conditioners, which are huge niches with equally huge competition. We saw something else: hair supplements, face serums, keratin. Biohacking is in fashion — we asked production to expand in that direction and took on promoting the interesting items that already existed.

2. Promotion in narrow niches

   As soon as we started promoting the chosen items in the less “popular” niches, they became leaders there and picked up revenue. One problem remains — stock: production is short of components inside Russia, and we are still solving it.

We created the cards with main-image testing, built reviews from zero and wrote our own competitor price monitor. Traffic came from low-frequency queries and in-card recommendations, where the manufacturers are not present.

Contracts with the manufacturers as their official distributor. Recommended prices across the whole range that we are obliged to hold — while the manufacturers themselves keep breaking them. Sales at zero, we start from scratch.

1. Created the cards, SEO and image funnels

   All to standard, but with testing: first the main image — if it worked we built the funnel in that style, then tested the funnel itself and rolled the winner out to the rest of the range.

2. Reviews from zero

   The main problem of a new store is having no reviews. We used every tool: points for reviews, self-purchases. Within six months that made us competitive on funnel metrics in a very crowded niche.

3. Our own price monitoring service

   It tracks competitor prices and adjusts ours when the manufacturers sell below their own recommended price. Periodically it lets us be cheaper than the manufacturer itself.

4. Advertising around the manufacturers

   The detailed WB ad settings let us find narrow gaps — low-frequency queries and in-card recommendations where competitors are not visible. That is how we took our share of traffic and of the market.

The manufacturers entered the platform themselves and dropped prices by 30% — fighting on price was not an option. We grew traffic on low-frequency queries and brought in the large-assortment analytics from the Mebelson case. The fall was stopped.

In 2024 the store did 160 m ₽ for the year — before the manufacturers came onto the marketplace themselves. In 2025 it was already 45 m and falling every month; in November, at the peak of the season, revenue was 2.3 m. The cause: the manufacturers went around their own recommended prices and started selling 30% below market, which for our client meant deep losses. On top of that the manufacturers have individual terms from the platform and we do not.

1. First we worked out where the sales even came from

   If we are 30% more expensive than the manufacturers, why is the store still selling? The analysis showed the sales were incidental. There is traffic, conversion is worse but it exists, and across 600 SKUs that randomness added up to 45 m over the year. So the task was not to fight on price but to grow traffic and increase that randomness.

2. Brought in solutions from other cases

   The large-assortment analytics came from Mebelson, as did working with linked cards and promoting the conversion leaders first.

3. Collected traffic on low-frequency queries

   As in the distribution case: picking up crumbs in low-frequency queries, growing traffic without going head to head with the manufacturers.