Naver Search API — Shopping, Blog, News, Cafe & Trends

Search Naver through its official Search Open API, with your own free Client ID. Get Naver Shopping products with lowest and highest price, mall, brand, maker and category path; blog posts, news, cafe posts, Knowledge iN questions, images, web documents, books, encyclopedia entries and local places; and DataLab trend series for search interest and shopping clicks. Every result comes back as one clean JSON row — no HTML tags, prices as numbers, rank included — ready for a spreadsheet, a database or an AI agent.

It is an API, not a scraper: no browser, no CAPTCHAs, no blocked pages. Your key, Naver's published quota, clean data.

Start in a few minutes:

1. Create your free Naver application and copy its Client ID and Client Secret (3 steps below; a one-time setup).
2. Paste both into the Actor, keep the prefilled search words (무선 이어폰, 텀블러) or type your own — Korean or English.
3. Press Start. Seconds later the Output tab lists the products with prices, malls, brands and categories. Download JSON, CSV or Excel.

You pay per result delivered. A search word that finds nothing, a wrong input, a rejected key or a used-up daily quota cost nothing.

Use it to:

• Watch prices and offers on Naver Shopping: lowest and highest price, used or rental items filtered out, Naver Pay only, newest first or by price.
• Compare brands, malls and categories for a keyword: who sells it, under which brand, in which category path.
• Track what Korea is saying: blog, cafe and news hits for a brand or product, with summaries, dates and links.
• Find books, places and encyclopedia entries, or the images Naver shows for a keyword.
• See what is rising with DataLab: search interest for keyword groups since 2016, shopping click trends per category and per keyword since 2017.
• Feed an AI agent or a dashboard with Korean market data from the source Naver itself offers to developers.

Get your free Naver Client ID (3 steps)

1. Sign in at developers.naver.com with a Naver account (free to create) and choose Application > Register Application (애플리케이션 등록). The first time, Naver asks you to accept its terms and to verify your mobile phone number and enter a company name — a one-time step.
2. Fill in the application: any name, and under API to use (사용 API) add Search (검색); for the trend types also add DataLab (Search Trends) (데이터랩 검색어트렌드) and/or DataLab (Shopping Insight) (데이터랩 쇼핑인사이트). Under the non-login API environment (비로그인 오픈 API 서비스 환경) add a WEB setting with a web address, for example http://localhost. Click Register (등록하기).
3. Copy the keys: open Application > My Applications (내 애플리케이션), click your application, and on the Overview tab (개요) copy the Client ID and, after clicking Show (보기), the Client Secret.

Paste them into Naver Client ID and Naver Client Secret. They are stored as secret inputs, sent only to Naver's API in the request headers, and never written to the results or the log. Naver's Search API gives every application 25,000 calls a day for free (DataLab: 1,000 a day); the quota counts per Client ID, so it is yours alone.

What each search type returns

Search type	What you get	Typical fields
shop Shopping	Products from Naver Shopping	title, lprice, hprice (KRW numbers), mallName, brand, maker, category1–category4, categoryPath, productId, productGroup (general / used / discontinued / upcoming), productKind, link, image
blog Blog	Naver blog posts	title, description, link, bloggerName, bloggerLink, postDate
news News	News articles	title, description, link, originalLink, publishedAt
cafearticle Cafe	Posts in public Naver Cafes	title, description, link, cafeName, cafeUrl
kin Knowledge iN	Questions on Naver's Q&A	title, description, link
image Images	Images with size	title, link, thumbnail, sizeWidth, sizeHeight
webkr Web	Web documents	title, description, link
doc Documents	Academic and professional documents	title, description, link
book Books	Books with list and sale price	title, author, publisher, price, discount, isbn, pubDate, link, image
encyc Encyclopedia	Encyclopedia entries	title, description, link, thumbnail
local Local	Places and businesses (max 5 per search word)	title, category, address, roadAddress, mapx, mapy, longitude, latitude, link
trend-search	Search interest over time for keyword groups (DataLab)	group, keywords, data [{ period, ratio }], peakPeriod, peakRatio, lastRatio
trend-shopping-category	Shopping click trend per category (Shopping Insight)	group, categoryCodes, data, peakPeriod, peakRatio
trend-shopping-keyword	Shopping click trend of keywords inside one category	group, keywords, categoryCode, data, peakPeriod, peakRatio

Every search result row also has searchType, query (your search word), rank (1-based, in the order you chose), totalResults (how many matches Naver reports), status and scrapedAt. Trend rows have searchType, group, the period and status instead.

How it works

1. You choose What to search, list your Search words (one per line) and how many results you want per word.
2. The Actor calls Naver's official Open API with your Client ID and Secret, page by page: up to 100 results per call, up to 10 calls per search word.
3. Each result is cleaned: the <b> tags Naver puts around matched words are removed, HTML entities are decoded, prices become numbers, empty values become null, dates become ISO dates, and shopping productType is split into a group and a kind.
4. The trend types send one DataLab request per run and return one row per keyword group or category, with the whole series in data and its peak and last value ready to read. Naver reports a relative ratio: the biggest value of the answer is 100. In the dataset API, unwind=data&flatten=data turns a series into one row per data point.

If Naver rejects your key or the quota is used up, the run stops at once with one clear message in the status line and one free record in the dataset — what was already delivered is kept.

Limits (set by Naver)

• 1,000 results per search word. Naver lets an application page through the first 1,000 results of a search (100 per call). Need more? Search with more specific words. Local search returns at most 5 places per search word.
• 25,000 Search API calls a day per Client ID (DataLab: 1,000 a day). A 1,000-result search word uses 10 calls. The quota resets every day.
• Shopping search returns the list data above. Naver's Search API does not include reviews, seller offer lists, stock or option details.
• Web, document and encyclopedia search have no sort order; local search has two (relevance, most reviewed).

Input

Field	What it does
clientId, clientSecret	Your Naver application's keys (secret inputs).
searchType	What to search: shop (default), blog, news, cafearticle, kin, image, webkr, doc, book, encyc, local or a trend type.
queries	Search words, one per line, Korean or English. For trend types, one group per line: Group name=word1,word2.
maxResultsPerQuery	Results per search word, 1–1,000 (default 100; local: up to 5).
sort	sim relevance (default), date newest first, asc / dsc price low→high / high→low (shopping), point rating (Knowledge iN), comment most reviewed (local).
excludeUsed, excludeRental, excludeCrossBorder, naverPayOnly	Shopping filters Naver applies itself: leave out used items, rentals, overseas direct purchase; keep only Naver Pay products.
imageSize	Image search: all, large, medium, small.
shoppingCategory	Trend type "shopping keyword": the cat_id number of a Naver Shopping category, for example 50000000.
trendStartDate, trendEndDate, trendTimeUnit	Trend period (yyyy-mm-dd; default: the last year up to yesterday) and date / week (default) / month.
trendDevice, trendGender, trendAges	Trend filters: pc / mo, m / f, age codes (see the input form).
maxConcurrency	Search words worked on at the same time (default 2).

Output

An example shopping row (the values are illustrative):

{
  "searchType": "shop",
  "query": "무선 이어폰",
  "rank": 2,
  "title": "삼성전자 갤럭시 버즈 무선 이어폰 5.3 \"프로\" 케이스 & 파우치",
  "productId": "82452688468",
  "lprice": 29900,
  "hprice": 159000,
  "currency": "KRW",
  "mallName": "네이버",
  "brand": "삼성전자",
  "maker": "삼성전자",
  "category1": "디지털/가전",
  "category2": "음향가전",
  "category3": "이어폰",
  "category4": "블루투스이어폰",
  "categoryPath": "디지털/가전 > 음향가전 > 이어폰 > 블루투스이어폰",
  "productType": 1,
  "productGroup": "general",
  "productKind": "price-comparison",
  "link": "https://search.shopping.naver.com/gate.nhn?id=82452688468",
  "image": "https://shopping-phinf.pstatic.net/main_8245268/82452688468.jpg",
  "totalResults": 1234567,
  "status": "OK",
  "scrapedAt": "2026-10-31T12:00:00.000Z"
}

An example trend row (search interest for one keyword group):

{
  "searchType": "trend-search",
  "group": "한글",
  "keywords": ["한글", "korean"],
  "startDate": "2017-01-01",
  "endDate": "2017-04-30",
  "timeUnit": "month",
  "points": 4,
  "peakPeriod": "2017-03-01",
  "peakRatio": 100,
  "lastPeriod": "2017-04-01",
  "lastRatio": 85.32,
  "data": [
    { "period": "2017-01-01", "ratio": 47 },
    { "period": "2017-02-01", "ratio": 53.23 },
    { "period": "2017-03-01", "ratio": 100 },
    { "period": "2017-04-01", "ratio": 85.32 }
  ],
  "status": "OK"
}

A search word that gave nothing is one free record with status NO_RESULTS (or FAILED, PARTIAL, INVALID_INPUT) and the reason in error.

FAQ

Is this an official Naver product? The data comes from Naver's official Search Open API and DataLab APIs, called with your own application's keys. The Actor itself is an independent tool and is not made by, or affiliated with, Naver. Your use of Naver's APIs is subject to Naver's API terms, which you accept when you register your application.

Why do I need my own Client ID? Naver issues API access per application: the 25,000 daily calls belong to the Client ID that makes them. With your own key the quota is yours, the usage shows in your own Naver developer console, and nobody shares a limit with you.

What does the Naver side cost? Nothing within the free quota (25,000 Search API calls a day, 1,000 DataLab calls a day). You pay only for the results this Actor delivers.

My run says "Naver rejected your credentials". What now? Check three things: the Client ID and Client Secret are copied exactly (no spaces); the application on developers.naver.com has Search added under its API settings tab (and DataLab for the trend types); the application is the one you copied the keys from. Then run again.

The status says HTTP 429. Naver's daily call quota for your Client ID is used up (or calls came too fast). It resets every day — run again later, or ask for fewer results per search word. The results delivered before the stop are kept.

Can I get product reviews, seller offers or stock? Not from this API: Naver's Search API returns the list data in the table above. It is the right tool for prices, malls, brands, categories, ranks, blog, cafe and news coverage, and trends.

Why are lprice or hprice empty? Naver sends 0 or an empty value when it has no price information; the Actor turns that into null instead of a misleading 0. For a product without price comparison, lprice is the product's own price.

Which language are the results in? Korean, as Naver sends them. Search words can be Korean or English.

Can I schedule it? Yes — use Apify schedules or the API to run the same search words daily and compare prices or ranks over time.

Like it? A short review on the Store page helps other people find this Actor. Something missing or broken? Tell us on the Issues tab — we read every one.

Ready? Get your free Client ID, paste it in, and press Start — your first Naver Shopping table is minutes away.