Naver Shopping Scraper: Korea Products & Prices

Search Naver Shopping (네이버 쇼핑), Korea's largest shopping search, by keyword and get one row per product with price, lowest price, seller count, mall, brand, reviews, rating, ad flag and links.

Use it for Korean price research, competitor and brand tracking, or product lists. The smallest run is one keyword: 무선 이어폰 gives 20 products in about 15 seconds. No login, no cookies or API key from you.

Table of contents

• What it does
• How it works
• Quick start
• Use cases
• Input
• Output
• Pricing
• Limits
• API, export and integrations
• Troubleshooting
• FAQ
• Related Actors
• Support

What it does

The Actor opens the public Naver Shopping search results for each keyword, the same list a shopper sees on a phone. It reads every product card and saves one clean row per product. It can go deeper than the first page: up to 1,000 products per keyword.

• Catalog cards (price comparison pages) give the shown price, the lowest price and the number of sellers.
• Single-seller cards give the price, the list price before discount, the discount, the mall name and the purchase count.
• Sponsored listings are kept and marked with isAd: true. You can turn them off.
• Each row has the keyword, its position, the source and the time we read it.

Good fit

• Check the market price of a product in Korea, with the lowest price across sellers.
• Watch which brands, malls and ads show up for your keywords.
• Build a product list for a category keyword with reviews and ratings.

Not a fit / limits

• Search results only. Product detail pages (options, stock, full review text, seller lists) are not included.
• Naver's ranking is personal and changes during the day. Two runs can show a different order.
• This is not a drop-in replacement for scrapers that expand each catalog into all its sellers.

👉 Try it now: open the input form

How it works

1. For each keyword, the Actor opens the public Naver Shopping mobile search, the same list a shopper sees. No login: it uses only the visitor session Naver gives every shopper, and keeps it in memory for that run.
2. It reads every product card, applies your price range and ad setting, and goes deeper than the first page with narrower price ranges until it has the number of products you asked for.
3. It saves one row per product to the Dataset, in Naver's order or sorted the way you chose. A keyword with no products or a Naver error gets one free status row, and the run summary (RUN_SUMMARY) says how the run went.

👉 Try it now: open the input form

Quick start

1. Open the Input tab. Search keywords (queries) is prefilled with 무선 이어폰 (box 1). Replace it with your own keyword; Korean keywords give the best results.

2. Keep Max products per keyword (maxItems) at 20 for a first test (box 2).

3. Optional: open Filters and sorting to set a price range, a sort order or leave out ads (box 3).

4. Click Start at the bottom of the Input tab.

5. Open the Output tab. The Products table shows one row per product with the title, price, lowest price, sellers and reviews.

   

Expected result: 20 product rows for the keyword, each with title, price, reviewCount and url.

👉 Try it now: open the input form

Use cases

Buyer job	Input path	Output entity	Key fields	Next step
Price check for a product	queries + minPrice/maxPrice	product	price, lowPrice, sellerCount, mall	Export to a sheet, convert KRW
Brand and ad watch	queries, includeAds: true	product	brand, maker, isAd, position	Schedule a daily run
Category product list	queries, maxItems: 1000	product	title, reviewCount, rating, categoryId	Load the dataset by API

👉 Try it now: open the input form

Input

Input fields

Field	Plain name	Type	Required	Default / prefill	Allowed values	Effect	Example	Recommendation
queries	Search keywords	array of strings	yes	prefill ["무선 이어폰"]	1–20 keywords, each up to 100 characters	One search per keyword	["선풍기", "립스틱"]	Use Korean words. Duplicates are run once.
maxItems	Products per keyword	integer	no	default 50, prefill 20	1–1,000	Stops each keyword after this many products	300	More products means more requests and a longer run.
minPrice	Minimum price	integer (KRW)	no	empty	0–1,000,000,000	Naver filters by price	10000	Use with maxPrice to skip accessories.
maxPrice	Maximum price	integer (KRW)	no	empty	0–1,000,000,000	Naver filters by price	300000	Must be ≥ minPrice.
sort	Sort	string	no	relevance	relevance, price_asc, price_desc, review_count, rating	Orders the collected products	price_asc	See the note below.
includeAds	Include sponsored listings	boolean	no	true	true / false	false drops ad cards	false	Keep true for ad research.
addBrandNames	Add brand and maker	boolean	no	true	true / false	Fills brand and maker from Naver's filters	true	Turn off for slightly faster runs.
deliveryConnector	Send results to	MCP connector	no	empty	your Slack, Notion or other connector	Posts a short run summary after the run	your Notion connector	See "Send results to Slack or Notion" below.
deliveryTarget	Notion page or channel	string	no	empty	Notion page link or Slack channel ID	Where the summary goes	a Notion page link	Leave empty to let the connector decide.

Empty queries or a wrong range stops the run at once with a clear message, names the field, shows an example, and charges nothing for products.

Field names from other scrapers also work: query, searchTerms or keywords are read as queries, and maxResults, limit or resultsLimit as maxItems. If both are set, queries/maxItems win. Unknown fields are ignored, logged as a warning and listed in RUN_SUMMARY.ignoredInputFields.

Send results to Slack or Notion

Optional. Pick your connector in Send results to (deliveryConnector; connect Slack or Notion once in Apify Console → MCP connectors). After the products are saved, the Actor posts a short summary there: status, product count, up to 25 product rows and links to the full dataset and the run. Put a Notion page link or a Slack channel ID in Notion page or channel (deliveryTarget): in Notion every run is added on top of that one page, so your workspace does not fill up. Without a page link, Notion creates a new private page for each run. A delivery error never fails the run and costs nothing extra; RUN_SUMMARY.delivery shows what happened. Without a connector nothing is sent.

Interacting settings

When	Input A	Input B	Effective behavior	Cost/result effect
Sort is not relevance	sort	maxItems	The Actor first collects maxItems products in Naver's order, then sorts them. price_asc gives the cheapest of those products, not the cheapest on all of Naver.	None
Ads are off	includeAds: false	maxItems	Ad cards are skipped and organic products fill the list.	Same price per product
Price range is set	minPrice / maxPrice	queries	Naver applies the range before the list is read.	Fewer products if the range is narrow

Input examples

Price check with a price range

{
  "queries": ["무선 이어폰"],
  "maxItems": 100,
  "minPrice": 10000,
  "maxPrice": 300000,
  "includeAds": false
}

Expected behavior: one search, sponsored cards removed, prices between ₩10,000 and ₩300,000. Expected output: up to 100 product rows with price, lowPrice, sellerCount, mall.

Several keywords, deep lists

{
  "queries": ["선풍기", "강아지 사료", "립스틱"],
  "maxItems": 300
}

Expected behavior: 900 rows in about 45 seconds (measured).

👉 Try it now: open the input form

Output

Output example

A real row for 무선 이어폰 (some fields trimmed):

{
  "recordType": "product",
  "status": "complete",
  "query": "무선 이어폰",
  "position": 8,
  "productId": "56796984706",
  "title": "삼성전자 갤럭시 무선이어폰 버즈3 FE SM-R420N 노이즈 캔슬링 그레이",
  "price": 143000,
  "originalPrice": null,
  "discountRate": null,
  "lowPrice": 132990,
  "currency": "KRW",
  "listingType": "catalog",
  "sellerCount": 71,
  "mall": null,
  "brand": "갤럭시",
  "reviewCount": 8467,
  "rating": 4.9,
  "purchaseCount": null,
  "isAd": false,
  "url": "https://search.shopping.naver.com/catalog/56796984706",
  "source": "naver_shopping_mobile_search",
  "sourceUrl": "https://m.search.naver.com/search.naver?where=m_shop&query=%EB%AC%B4%EC%84%A0+%EC%9D%B4%EC%96%B4%ED%8F%B0",
  "fetchedAt": "2026-10-14T12:35:23.121Z"
}

Empty values are null, never 0. A 0 means a real zero (for example, free delivery).

Output fields

Field	Meaning	Type	Can be empty?	Example / enum
recordType	Row type	string	no	product, query
status	Row quality	string	no	complete, partial (some fields missing); query rows: not_found, error
query	Your keyword	string	no	무선 이어폰
position	Place in the result list, starting at 1	integer	no	8
productId	Naver product ID (nvMid), stable	string	no	56796984706
title	Product name, HTML removed	string	no	필립스 ENC 블루투스 무선 오픈형 이어폰 TAQ4000
price	Price the shopper sees first, KRW	integer	yes	59000
originalPrice	Price before discount	integer	yes	115000
discountRate	Discount in %	number	yes	48
lowPrice	Lowest price shown by Naver; for catalogs, across sellers	integer	yes	132990
currency	Always KRW	string	no	KRW
listingType	Catalog (price comparison) or one seller	string	no	catalog, single_seller
sellerCount	Sellers in the catalog; 1 for single-seller cards	integer	yes	71
mall	Store name (single-seller cards)	string	yes	필립스 오디오 공식몰
brand / maker	Brand or manufacturer, from Naver's filters	string	yes	삼성전자
categoryId / categoryIds	Naver category IDs, leaf and path	string / array	yes	50002327
reviewCount	Number of reviews (catalogs: total across sellers)	integer	yes	8467
rating	Average rating 0–5	number	yes	4.9
purchaseCount	Purchases shown by Naver (single-seller only)	integer	yes	1200
wishlistCount	Wishlist (찜) count	integer	yes	340
deliveryFee	Delivery fee, KRW; 0 = free	integer	yes	0
isAd	Sponsored listing	boolean	no	true
isSmartStore / isBrandStore / isOverseas	Store flags	boolean	yes	true
smartStoreProductId	Smart Store product number	string	yes	11223344556
url / mobileUrl	Product or catalog link	string	yes	https://search.shopping.naver.com/catalog/...
image	Main image	string	yes	https://shopping-phinf.pstatic.net/...
missingFields	Which key fields were missing	array	no	["mall"]
source	Where the row came from	string	no	naver_shopping_mobile_search
sourceUrl	Search page for the keyword (without your filters)	string	no	https://m.search.naver.com/...
fetchedAt	When we read it (UTC)	string	no	2026-10-14T12:35:23.121Z
message / errorCode	Query rows only: what happened	string	yes	Naver Shopping has no products for this keyword and filters

Output entities and surfaces

Situation	Where	Result shape	What to do
Normal product	Dataset, view Products	one product row per product	Export or use by API
Keyword with no products	Dataset	one free query row, status: not_found	Check the spelling or the price range
Keyword failed (block, timeout)	Dataset + RUN_SUMMARY	one free query row, status: error, errorCode	Run that keyword again later
Run summary	Key-value store RUN_SUMMARY	status, counts, per-keyword depth, warnings	Check status and message

RUN_SUMMARY → status tells you the result: COMPLETE, NO_RESULTS, PARTIAL or CAPPED (your spending limit was reached). When a run stops early (PARTIAL or CAPPED), the run status message in Console also says why.

👉 Try it now: open the input form

Pricing

Pricing unit

Pay per event. You pay for each product row saved, plus a small start fee per run. not_found and error rows are free. Platform usage is included: you do not pay for compute.

Event	Free plan	Bronze	Silver	Gold and higher
Product (one row)	$0.0015	$0.0014	$0.00135	$0.0013
Run start	$0.004	$0.004	$0.004	$0.004

A run with no products (all keywords not found, or a price range with no matches) costs only the start fee, $0.004. A partial run pays only for the rows it saved.

The start fee depends on run memory: it is charged once for runs with up to 1 GB (the default is 256 MB) and once more for each extra GB if you raise the memory in the run options.

Cost examples (Free plan)

Scenario	Input scope	Products	Base charge	Optional charge	Stop/cap behavior
First test	1 keyword × 20	20	$0.034	none	stops at 20 products
Normal	3 keywords × 300	900	$1.354	none	stops at 300 per keyword
Large	5 keywords × 1,000	5,000	$7.504	none	stops at 1,000 per keyword or at your cost limit
No results	20 unknown keywords	0	$0.004	none	start fee only

If you set a maximum cost per run, the Actor stops before it goes over, keeps the rows it saved and ends with status CAPPED.

If you set Max items in the run options (the platform limit on paid results for the whole run), the Actor saves at most that many products across all keywords, ends as SUCCEEDED with limitReached: "MAX_ITEMS" in RUN_SUMMARY, and lists the keywords it did not reach.

👉 Try it now: open the input form

Limits

• Up to 20 keywords and 1,000 products per keyword per run.
• Speed: about 1,000 products in 20 seconds. Each keyword takes at least 3–15 seconds, even for a few products, because brand names need extra requests (20 keywords × 1 product took about 3 minutes). Set addBrandNames to false for faster small runs. 20 keywords × 1,000 products took 350 seconds. The default timeout is 300 seconds, so for more than about 15,000 products set a longer timeout (for example 600 seconds) in the run options. If time runs out, the run keeps what it saved, ends as PARTIAL and lists the keywords it did not start.
• A few keywords show a short list on Naver (for example hotel or travel words). Then you may get fewer products than maxItems.
• If Naver fails for 3 keywords in a row, the run stops instead of waiting for the timeout. It keeps the products it saved and lists the keywords it did not start.
• If one keyword fails but others work, the run still succeeds: it ends as PARTIAL, and that keyword gets a free error row.
• Keywords are up to 100 characters. Prices in minPrice and maxPrice are 0 to 1,000,000,000 KRW.
• One run makes at most 900 requests to Naver and stops after 25 minutes even with a longer timeout. One keyword uses at most 40 requests. For 20 keywords × 1,000 products, split them over two runs if the run stops early.
• If you set a maximum total charge for the run, the run stops when it is reached and ends as CAPPED.
• Max items in the run options limits products for the whole run, across all keywords. When it is reached, the run ends as succeeded, RUN_SUMMARY shows COMPLETE with limitReached: "MAX_ITEMS", and the keywords it did not reach are listed as skipped and not charged.
• Each request to Naver waits up to 25 seconds and is tried up to 3 times; one run waits at most 30 seconds in total between retries. A Naver answer larger than 6 MB is rejected for that keyword. The run timeout must be longer than 30 seconds.

👉 Try it now: open the input form

API, export and integrations

• Export: download the dataset as JSON, CSV, Excel or HTML from the Output tab.
• API: start runs and read the dataset with the Apify API or the Apify client.
• Scheduling: run the same keywords every day with Apify schedules and compare price and position over time.
• No-code: connect the dataset to Google Sheets, Make or Zapier through Apify integrations.

Use with AI agents (MCP)

• MCP server URL with only this Actor: https://mcp.apify.com?tools=cliqtomedia/naver-shopping-scraper
• Smallest useful input:

{ "queries": ["무선 이어폰"], "maxItems": 20 }

• Key output fields: title (product name), price and lowPrice (KRW), reviewCount and rating, isAd, url.
• A 20-product call takes about 15 seconds and costs $0.034 on the Free plan. Keep to 20 keywords and 1,000 products per keyword.

👉 Try it now: open the input form

Troubleshooting

The run stopped before all keywords

Symptom: the last keywords are skipped with "Not processed: run reached the Max items limit". Likely reason: Max items in the run options is lower than keywords × maxItems. Check: RUN_SUMMARY.limitReached is MAX_ITEMS. Action: raise or clear Max items in the run options, or lower maxItems per keyword.

I got fewer products than maxItems

Symptom: the keyword has fewer rows than you asked for. Likely reason: Naver shows a short list for this keyword or price range. Check: RUN_SUMMARY → the keyword's depth, and the price range. Action: widen the price range or use a broader keyword.

The run ended as PARTIAL with "reached the run time limit"

Symptom: some keywords are missing and the message says to give the run more time. Likely reason: many keywords × 1,000 products need more than the 300-second default. Action: set the run timeout to 600 seconds, or split the keywords over several runs.

A keyword has a row with status: error

Symptom: one query row with errorCode such as RATE_LIMITED or SOURCE_UNAVAILABLE. Likely reason: Naver did not answer in time or blocked the request. This is not an empty result. Check: the run itself ends as PARTIAL (not red) when other keywords worked. Action: run that keyword again later. Error rows are free.

Error codes: SOURCE_TIMEOUT — Naver did not answer in 25 seconds; SOURCE_UNAVAILABLE — Naver answered with a server error; RATE_LIMITED — Naver asked to slow down; BLOCKED — Naver refused the request; SOURCE_SCHEMA_CHANGED — Naver's page or answer looks different from what the Actor reads (or is larger than 6 MB).

The run stopped with "keywords in a row failed"

Symptom: the run is FAILED and the message says 3 keywords in a row failed and some keywords were not processed. Likely reason: Naver was unavailable or blocked requests for a while. The Actor stops early so you do not wait for the timeout. Action: run the same input again later. You pay only for products already saved.

The run status is FAILED

Symptom: the run is red and the message says how many products were saved. Likely reason: every keyword failed, or Naver stopped answering for 3 keywords in a row. Check: the dataset still has the products saved before the failure, and you pay only for those rows (plus the start fee). RUN_SUMMARY lists which keywords failed. Action: run the failed keywords again later.

brand and maker are empty

Likely reason: for some keywords (for example fashion), Naver shows no brand filter, so there is nothing to read. Action: none needed; the rest of the row is complete.

👉 Try it now: open the input form

FAQ

Do I need a Naver account or API key?

No. The Actor reads public search pages. No login, and you don't provide cookies.

Why is a catalog price sometimes very low?

For catalog cards, Naver shows the lowest offer across all sellers. Sometimes that offer is a rental, a used item or an accessory (we saw rentals at ₩1). The Actor keeps Naver's number as shown; check sellerCount and listingType, or set minPrice.

Is the data the same as on Naver?

Yes, it is the order and prices that Naver's mobile search shows at fetchedAt. Naver personalises ranking and updates prices all the time, so runs at different times can differ. Reviews on catalog cards are Naver's total across sellers.

Can I use the data?

The Actor reads public pages. Naver's terms and robots.txt limit automated access, and you are responsible for how you use the data. If you need an official channel, Naver offers a Shopping Search API with your own developer key and a daily quota. This Actor is independent and not affiliated with Naver.

👉 Try it now: open the input form

Related Actors

• AliExpress Scraper: Products, Prices & Reviews, No Login — compare the same products with AliExpress prices and reviews.
• Idealo Price Scraper: All Shop Offers, 6 Countries, No Login — all shop offers for a product in six European countries.
• Trendyol Scraper: Products, Prices & Reviews (No login) — products, prices and reviews from Turkey's largest marketplace.
• Historical Exchange Rates & Currency Converter (ECB) — convert KRW prices to USD or EUR for a date.
• Google Autocomplete Scraper: Keyword Suggestions — find more keyword ideas to search.

👉 Try it now: open the input form

Support

Use the Issues tab of this Actor. Add the run ID, your keywords and the status or errorCode you saw. Do not send API tokens or private data.

⭐ Found it useful? Rate this Actor. Something wrong? Report a problem — we fix reported problems.

👉 Try it now: open the input form