Why Your Coupang Listing Has Impressions But No Clicks (And the Korean Fix)

Impressions without clicks is a different problem than no impressions

Sellers often lump these together, but they have opposite causes. No impressions means the Coupang search engine does not think you are relevant to the query — a keyword and category problem. Impressions with no clicks means the engine already decided you are relevant enough to show, and Korean shoppers looked at your card and chose someone else.

That is good news. You have already won the hard part. What you are losing is a two-second comparison happening on a phone screen, against roughly six to ten competing cards, all in Korean. Everything the shopper uses to decide is visible before the click: thumbnail, title, price, 로켓배송 (Rocket Delivery) badge, star rating, and review count.

So the fix is not more keywords. It is rewriting the parts of the card a Korean shopper actually reads.

• No impressions → keyword relevance, category placement, option structure.
• Impressions, no clicks → thumbnail, title readability, price framing, delivery badge, review count.
• Clicks but no sales → detail page (상세페이지), shipping time, return policy, Q&A.
• Diagnosing the wrong one wastes weeks. Check which bucket you are in before changing anything.

Pull the actual search results page your buyers see

Before editing anything, look at your listing the way a shopper does. Open Coupang search on mobile, in Korean, for the exact keyword driving your impressions — not your brand name, not an English term. Screenshot the result and study the whole row, not just your card.

Most non-Korean sellers have never done this. They evaluate their listing in isolation in Wing (Coupang's seller backend) where it looks fine. On the live results page, next to competitors, the failure is usually obvious within seconds.

Track which keywords are actually delivering the impressions and where you sit for each of them. If you cannot read the results page, a rank tracker that reports Korean keyword positions in English — SeoulRank does this daily — at least tells you which query to go stare at.

• Search on mobile width; over two-thirds of Coupang browsing is phone-first.
• Note the top three cards' price, review count, and badge — that is your benchmark.
• Check whether competitors show 로켓배송 (Rocket Delivery) and you do not.
• Zoom out until cards are thumbnail-size and see if your image still reads.

Fix the thumbnail first — it is 80% of the decision

Korean marketplace thumbnails follow conventions that differ sharply from Amazon. Amazon enforces pure white background, product only. Coupang tolerates — and shoppers expect — a cleaner product shot but with far more context: the product in use, a size reference, sometimes the full bundle laid out.

The most common non-Korean seller mistake is a single floating product on white with no scale reference. It looks premium to a Western eye and looks empty to a Korean shopper who is trying to figure out how big the thing is and what comes in the box.

Second most common mistake: the product is too small in the frame. Fill 80–90% of the crop. On a phone, your thumbnail is about the size of a postage stamp.

• Show the full quantity if you sell a multipack — Korean shoppers count units in the image.
• Add a size cue: a hand, a desk, a standard object. Do not rely on the spec table.
• Avoid English-only text overlays; if you add text, it must be in Korean and legible at thumbnail size.
• Never use lifestyle clutter that hides the product outline.
• Match the color you actually ship. Color mismatch is a top cause of returns and one-star reviews.

Rewrite the Korean title for scanability, not keyword stuffing

Coupang titles get truncated aggressively on mobile — often after roughly 25–35 Korean characters on the search card. Anything after that point is invisible to a browsing shopper, even though the search engine still indexes it. So your title has two jobs with two different audiences, and the front of the title belongs to the human.

The pattern that works: 브랜드 (brand) + 핵심 키워드 (core keyword) + 차별점 (differentiator) + 수량/용량 (quantity or volume). The differentiator is where most listings fail. "무선 청소기" (cordless vacuum) tells the shopper nothing your competitors do not also say. "무선 청소기 흡입력 2만파 초경량" gives them a reason.

Machine-translated titles read wrong to Koreans in a way that is hard to describe but instantly noticeable — wrong particle usage, English word order, loanwords transliterated the long way when a shorter form is standard. If your title came straight out of a translation tool, it is probably suppressing your CTR even when the keywords are technically correct.

• Front-load the term people actually search, not the formal product-category name.
• Use the loanword form Koreans search — e.g. 텀블러 (tumbler) usually outperforms 보온병 for certain product types; check search volume rather than guessing.
• Put quantity and volume in numerals: 2입, 500ml, 3개입.
• Drop marketing adjectives like 최고의 (the best) — they read as spam and can trigger ad-copy restrictions.
• Never repeat the same keyword three times. It does not help ranking and it looks like a scam listing.

Price framing, badges, and social proof — the trust layer

Korean shoppers on Coupang are unusually delivery-sensitive. A listing without 로켓배송 (Rocket Delivery) or at least a clear fast-shipping indication loses clicks to one that has it, even at a higher price. If you are a cross-border seller shipping from outside Korea, this is your single biggest structural CTR handicap — and the honest fix is either 로켓그로스 (Rocket Growth, Coupang's fulfillment program) or brutally explicit delivery dates.

On price: avoid being the cheapest by a wide margin. In Korean marketplaces, a price far below the cluster reads as counterfeit or defective, particularly in cosmetics, electronics, and baby goods. Sitting slightly under the median with a visible discount percentage usually beats being the floor price.

Review count is the other lever. Below roughly ten reviews, your card is at a real disadvantage no matter how good the image is. Prioritize getting the first reviews on one hero SKU rather than spreading thin across a catalog.

• If you cannot offer Rocket, state a concrete arrival window in the option name where it is visible.
• Set a 할인율 (discount rate) that actually displays — small single-digit discounts often do not render as a badge.
• Consolidate variants under one product so reviews accumulate on a single listing instead of fragmenting.
• Watch for competitors undercutting on 묶음 (bundle) pricing — sometimes your unit price is fine and your pack size is wrong.

A two-week test loop you can actually run

Change one variable at a time and give it enough days to stabilize. Coupang's ranking and CTR data are noisy day to day, especially around weekends and payday cycles in Korea (late month tends to be softer).

A practical sequence: week one, swap the thumbnail only. Week two, rewrite the title front-half only. Compare CTR on the same set of keywords, not sitewide averages, because keyword mix shifts will fake a result either way.

Keep a simple log of which keyword you rank where, before and after each change. Daily position data in English makes this tractable if you cannot read the Korean SERP yourself — that is the main thing tools like SeoulRank are for. Without it you are changing things blind and attributing results to the wrong edit.

• One change per week, logged with the date.
• Compare keyword-level CTR, never account-level.
• Ignore the first 48 hours after any edit — Coupang re-evaluates the listing.
• If a change drops rank as well as CTR, revert immediately rather than stacking another fix on top.
• Once a thumbnail and title combination wins, apply the pattern across your catalog and re-test only the outliers.