Merging is the combining of several SKUs into a single card: the customer sees one product with a choice of color, reviews accumulate in a shared pool, and the seller manages every variant from one place. The tool is powerful and double-edged: correct merging speeds up the launch of new items, incorrect merging sinks a strong product under a weak neighbor. Here is the mechanics and the strategy.
How merging works on WB
Grouping runs on matching SKU parameters (color first of all; sizes inside an SKU are a separate axis). What becomes shared: reviews and rating, questions, and to a large extent the behavioral history of the card. What stays specific to each variant: the article number, stock, price (which may differ), and photos of that color.
The procedure is carried out in the seller account (merging existing SKUs or creating variants inside a card) — the interface changes from time to time, so check the current WB help center for the mechanics. What matters more than the interface is deciding what to merge and why.
When merging helps
- Launching a new color. The new item starts with shared reviews and rating, skipping the cold start: this is a legitimate accelerator compared with building traction from zero.
- Scattered demand. Ten cards for the same dress in different colors compete with each other in search; one strong card collects the behavioral signals together.
- Manageability. One set of content and keywords instead of ten copies.
When merging hurts
- A weak variant drags everything down. A color with a defective batch collects negative reviews — the rating falls for the whole group, bestseller included.
- Different audiences and queries. "Black office dress" and "leopard evening dress" are different intents: inside one card they blur the keywords and the conversion rate. Different intents need different cards, each with its own semantic core.
- Different price segments. A variant twice as expensive inside the same group confuses the customer and damages conversion.
Strategy: decision rules
- Merge: one product — one audience — one intent, where variants differ only by color or print; a new variant added to a strong parent card.
- Do not merge: different intents, audiences, or price levels; a test product of doubtful quality (first test it separately, then merge the winner).
- Un-merge: when a variant steadily damages the rating or pulls conversion down. Remember: once separated, the variant loses the shared accumulated capital and effectively starts over — treat the decision as an amputation, based on the math.
Related reading: variations (aspects) on Ozon.
FAQ
"What does merging cards on WB give you?" Shared reviews and rating across every variant, unified content, and concentrated behavioral signals instead of your own cards competing in search. New colors launch with an accumulated reputation.
"Are reviews combined after merging?" Yes, the pool is shared — that is both the main strength and the main risk: negative feedback on one variant is visible to the customers of every variant.
"Can you un-merge cards afterwards?" Technically yes, through the seller account. But the separated variant loses the shared rating and history — it starts almost from scratch. Un-merging is justified when a variant systematically harms the group.
"Should you merge different models of the same product?" No — merging is meant for variants of one product (color or print). Different models with different intents inside one card blur the keywords and the conversion of both.
"Does merging affect search positions?" Yes: a merged card concentrates sales, reviews, and behavior — the signals WB ranks on. But the effect works both ways: weak variants dilute the average metrics of the group.
