Why Your Coupang Listing Has Impressions but No Clicks (and the Korean Fix)

Impressions without clicks is a tile problem, not a ranking problem

Coupang search works in two stages. First the algorithm decides whether to show you — that is driven by keyword relevance, sales velocity, reviews, and fulfillment signals. Then a Korean shopper decides whether to tap you — and that decision is made entirely inside a small product tile: thumbnail, title, price, delivery badge, review count and star rating.

If you have impressions, stage one is working. You are relevant enough to be served. A near-zero click-through rate (클릭률, CTR) means stage two is failing, and stage two is where non-Korean sellers lose the most money, because every element the shopper reads is in a language you cannot audit.

Worse, low CTR feeds back into ranking. Coupang's system interprets "shown often, never tapped" as a weak match for that query and gradually demotes you. So a tile problem becomes a ranking problem within a few weeks if you ignore it.

• Diagnose first: high impressions + low CTR = tile issue. Low impressions = keyword or relevance issue. Good CTR + no orders = detail page or price issue.
• Check CTR per keyword, not in aggregate. You can look fine overall while bleeding on your highest-volume term.
• Do not touch keywords until the tile converts. Fixing CTR lifts every keyword at once.

Reason 1: your title is grammatically foreign

The single most common cause is a machine-translated title. Korean shoppers skim listings at speed and they are unusually sensitive to awkward phrasing, because on Coupang it is a reliable signal of a low-effort overseas reseller. A title that reads as translated triggers a silent trust penalty before price is even considered.

Native Coupang titles follow a loose but real convention: brand → product noun → key spec → variant → quantity. Particles (은/는/이/가) are dropped. Words are separated by spaces, not commas. English brand names stay in Latin script; almost everything else is Hangul.

Two specific translation traps: a literal rendering of the product category noun, and adjective order. "무선 휴대용 미니 선풍기" (wireless portable mini fan) reads native. Reverse those modifiers and it reads like an import.

• Use the category noun Korean shoppers actually type — 텀블러 (tumbler) not 물병, 무선청소기 (cordless vacuum) not 선없는 청소기.
• Put the highest-intent keyword in the first 20–25 characters; mobile truncates the rest.
• Drop marketing adjectives like 최고의 (the best) — they read as spam and are often restricted.
• Include the spec Korean shoppers filter on: capacity (500ml), count (30매 = 30 sheets), size (사이즈), wattage.
• Never repeat the same keyword three times. Coupang deduplicates and shoppers notice.

Reason 2: your thumbnail loses the 3-item scan

Coupang's mobile grid shows roughly two to three products per screen. Your thumbnail is competing against Korean sellers who have iterated on theirs for years, and the local style is specific: high-key white background, product filling most of the frame, and — critically — the product shown in use or in context rather than floating alone.

Coupang's main image policy restricts overlaid text and promotional badges on the representative image, so you cannot win with a Korean-style sales graphic there. The lever you do control is crop, brightness, and angle. Overseas listings consistently under-crop: the product sits small in the middle with dead white space, which at thumbnail size reads as an empty tile.

Fix it by exporting your main image at the recommended square size, cropping so the product occupies most of the frame, and raising exposure slightly. Then use your additional images and the detail page for the Korean-language text callouts.

• Crop tight — product should fill roughly 80–90% of the square.
• Test on a phone at actual thumbnail size, not on your desktop monitor.
• Avoid lifestyle shots with non-Korean models for domestic-feel categories like kitchen, baby, and home.
• Keep the color palette clean; heavy shadows and gray backgrounds disappear in the grid.
• Put your Korean feature callouts in images 2–5, where overlay rules are looser.

Reason 3: the delivery badge gap

Korean shoppers filter and scan for 로켓배송 (Rocket Delivery, Coupang's next-day fulfilled service) almost reflexively. If the tiles above and below you carry that badge and yours says a 5–10 day overseas shipping window, your CTR will be low no matter how good your title is. This is not a copy problem you can write your way out of.

If you are shipping cross-border, your realistic options are to move inventory into a Coupang fulfillment program, or to compete in categories and price bands where delivery speed matters less — bulky niche goods, collectibles, specialty ingredients, hobby items where nothing comparable is domestically stocked.

If you must stay on longer lead times, make the delivery expectation explicit and specific in Korean rather than vague. A stated arrival window reads as competent; an undefined one reads as risky.

• Compare your tile against the top 10 results for your main keyword and count how many have 로켓배송.
• If it is 8+, that keyword is effectively closed to slow shipping — shift targeting.
• Look for long-tail keywords where the Rocket density is low; those are your entry points.
• Mention 해외직구 (overseas direct purchase) clearly — shoppers who search it expect longer shipping and self-select.

Reason 4: you are priced against the wrong shopper

Coupang's default sort mixes relevance with price competitiveness, and the tile shows a unit price breakdown for multipacks. Overseas sellers frequently price a single unit against Korean sellers' 3-packs, so the shopper sees your ₩12,900 next to an effective ₩4,300/unit and never taps.

Also check how your price renders. Korean listings almost always show a coupon-applied or discounted figure with a strikethrough original. A flat single price with no discount framing looks stale in the grid even at the same final cost.

And check the review count on the tiles around you. Under roughly 10 reviews, Korean shoppers heavily discount everything else on your tile. Early review accumulation is a CTR investment, not just a conversion one.

• Bundle into 2-packs or 3-packs so your per-unit display price is competitive.
• Set a modest list price and run a standing coupon so the tile shows a discount percentage.
• Avoid free-shipping-inclusive pricing that pushes your headline number above the visible cluster.
• Prioritize getting past the first 10–20 reviews on one hero SKU rather than spreading thin.

How to run the diagnosis weekly

Treat CTR as a per-keyword metric with a weekly review. Pull your top 10–15 impression-driving keywords, note the CTR on each, then search each term yourself on mobile and screenshot the grid your tile lands in. Ninety percent of the diagnosis is just looking at your tile next to the competition at real size.

The hard part for non-Korean sellers is knowing what the competing tiles actually say and whether your own title reads native. This is where translated keyword and rank tooling earns its keep — SeoulRank shows Korean keyword data and daily Coupang positions in English, so you can see which query you are being served on and what rank you occupy before you start rewriting.

Change one variable at a time and give each test 7–14 days. Title and thumbnail changes propagate through Coupang's index gradually, and stacking two edits at once means you learn nothing.

• Week 1: rewrite the title using native keyword order. Measure.
• Week 2: re-crop and brighten the main thumbnail. Measure.
• Week 3: adjust bundle size or add a coupon. Measure.
• Week 4: prune keywords where Rocket density makes you uncompetitive, and redirect ad spend to long-tail terms.
• Keep a simple log of before/after CTR per keyword — it becomes your playbook for every SKU you launch after this one.