Negative keywords are the main precision tool in Wildberries advertising: they cut off impressions for queries that drain your budget without bringing orders. Here is where to look for exclusion candidates, what to add to the list, and how not to overdo it and choke your reach.
Why negative keywords matter
An automatic campaign pulls queries from the card's content, and even with a well-built semantic core some impressions go to junk: other brands, homonyms, off-target intent. Every such impression is paid for through the shared auction — and it damages campaign statistics: the algorithm sees impressions without clicks or orders and marks the card as weak. Exclusions solve both problems at once: they save money and clean up the signals.
Where to find candidates
The only source is the search query report in the "WB Promotion" dashboard. Look at queries that have impressions and spend but no carts or orders. The working order:
- Before launch — a draft list: competitor brands, obvious off-topic queries, attributes your product doesn't have (wrong size, color, material).
- Days 3–7 — first cleanup based on the real report: queries with spend and zero carts.
- Weekly — regular cleanup with the same filter.
- Once a month — audit the negative keyword list: check whether you have cut off a query that started to grow.
What to exclude first
- Other brands. Impressions on those queries get clicks from people loyal to another brand — conversion is near zero.
- Homonyms and different products. A sports "suit" catches impressions for costume queries; men's "sneakers" catch kids' queries.
- Off-target intent. "How to sew," "DIY," "sewing pattern," "repair" — this person is not going to buy a finished product.
- Nonexistent variations. Sizes, colors, and configurations your card doesn't offer: there will be a click, but no order.
Exact and phrase match exclusions
WB changes the exclusion modes and interface from time to time — check the dashboard for current options. The principle stays the same: a phrase exclusion cuts the whole cluster of queries containing that phrase, an exact exclusion removes only the specific query. Phrase mode is more powerful and more dangerous: along with the junk, it easily cuts the targeted long tail. When in doubt, exclude exact matches, one query at a time.
Negative keywords are part of a system, not a magic button
Cleaning up semantics reduces spend, but it doesn't fix a weak card: if the main image gets no clicks and the price is above market, negative keywords only slow down the bleed. The full budget-management logic is in the guide on how not to waste your WB ad budget; the big picture of ad formats is in the Wildberries advertising hub.
FAQ
"What are negative keywords on Wildberries?" Queries you block your ads from showing on. They are set in the campaign settings and remove impressions on irrelevant queries: other brands, homonyms, off-target intent.
"How do I know which queries to exclude?" Use the search query report: candidates are queries with impressions and spend but no carts or orders. Orders lag behind carts, so watch carts first.
"Can I add negative keywords to an automatic campaign?" Yes — exclusions are the main control tool for an auto campaign: you can't pick queries directly there, so you manage the bid, placement zones, and negative keywords.
"How often should I clean negative keywords?" First cleanup 3–7 days after launch, once statistics accumulate. After that, weekly based on the report, plus a monthly audit of the negative keyword list itself.
"Why did impressions and sales drop after adding negative keywords?" Most likely a phrase exclusion cut a targeted query cluster. Restore the most recently added negative keywords one at a time and find the one that was carrying your reach.
