A photo wallpaper and wall art manufacturer from Ivanovo cut its account-wide ad-to-sales ratio (DRR) from 10% to 3.5%, while one competitive SKU generated about 17 rubles in orders for every ruble spent on ads. The project launched on Wildberries and Ozon right after Easter, during a seasonal dip, while competitors were selling similar products at half the price.
With photo wallpapers averaging over 7,000 ₽ per order, matching a price war kills margin right away: workshop costs, logistics, and platform commissions stay put. Instead of dumping prices, the team rebuilt traffic, improved listing conversion, and launched fresh designs.
Weekly Funnel of the Rebuilt Ad Campaigns
Instead of vague promises, let's look at one week of stats for a single competitive SKU after cleaning up the ad campaigns.
| Metric | Value |
|---|---|
| Impressions / Clicks | 96,859 / 4,557 (CTR 4.70%) |
| Add to Cart | 699 (15.34% of clicks) |
| Orders | 28 units for 216,018 ₽ (avg check ≈ 7,715 ₽) |
| Ad Spend | 12,191 ₽ |
| DRR to Orders / ROAS | 5.64% / 17.74 |
The listing reliably turns clicks into cart additions, and some drop-off at checkout is normal: people rarely buy large decor on impulse, as they measure walls and check with family first. Still, the high average check and clean traffic deliver strong payback. Across the whole account, DRR ended up even lower than on this single SKU thanks to organic sales and cross-browsing inside the store.
Five Levers Instead of Price Dumping
You cannot cut ad spend by two-thirds just by trimming bids; it took a five-step combination.
First, the team completely rebuilt campaigns and cut out irrelevant impressions that were burning budget. In parallel, they ran A/B tests on main photos for top listings to squeeze more clicks out of the same search results.
The third step was a line of lead-magnet SKUs for assisted traffic. You put a catchy or more affordable item on the storefront and send ads there: shoppers enter the brand, browse similar options, and buy a different, higher-margin size or design. Ads on one listing feed sales to neighboring ones.
As the fourth lever, the seller's team received a barter traffic system. In home decor niches, instead of paying cash, you send the product itself to bloggers and renovation channels: working capital stays intact, and the listing gets warm traffic with real apartment photos. Finally, having their own production allowed them to release new wallpaper designs regularly, so buyers choose a specific image rather than the lowest price in search.
The Video Cover Trap in A/B Tests
A technical detail from this case shows how easily a small setting can ruin your test data. On Wildberries, the main-photo testing tool runs under the 'Jam' subscription, and the system aborted the test with the status 'Test canceled: Disable video autoplay'.
If a listing has video cover autoplay turned on, shoppers in search see the video instead of the test photo, so the algorithm cannot compare click-through rates fairly. You don't have to delete the video from your gallery: just uncheck autoplay in the video cover settings, and the clip will still play inside the listing when clicked.
Even before cancellation, each variant gathered over 1,500 impressions.
| Metric | Variant A (300×280) | Variant B |
|---|---|---|
| Impressions | 1,629 | 1,594 |
| CTR | 7% | 9.6% |
| Add to Cart | no data | 8 (5.23% of 153 clicks) |
| Orders | 0% | 1 (12.5% of carts) |
Variant B beat Variant A in CTR by more than a third, which immediately lowers your cost per click at the same bid. However, it is too early to judge purchase conversion from this table: a single order out of 1,500 impressions is too small for conclusions, so you pick the cover based on CTR.
Reading the Numbers Without Illusions
The weekly stats above cover one SKU and count orders, not actual buyouts. If shoppers buy out, say, 80% of orders, the real DRR for this SKU rises from 5.64% to 7.05%, and ROAS drops from 17.74 to around 14.2. That is still a strong result, just more modest than the headline number. For the full account, the case study only shares the DRR drop, without total revenue or net profit.
The model also depends on your resources. Launching fresh designs and running steady blogger barter is much easier for a manufacturer with a team and a visual product. If you resell a standard item from China identical to your neighbors', you can't refresh the lineup in a week. You will have to lean harder on rebuilding ads, A/B testing covers, and shaping your store's price matrix.

