Naver Place Reviews + Blog — Korean Review Scraper avatar

Naver Place Reviews + Blog — Korean Review Scraper

Pricing

from $1.00 / 1,000 results

Go to Apify Store
Naver Place Reviews + Blog — Korean Review Scraper

Naver Place Reviews + Blog — Korean Review Scraper

Extract visitor reviews, ratings, keywords and visit details from Naver Place, plus blog reviews — with deepBlogSearch, also blog posts beyond Naver's ~100-per-place cap, found by place name. Place URL, ID or keyword input. From the #1 Naver data provider on Apify.

Pricing

from $1.00 / 1,000 results

Rating

0.0

(0)

Developer

OrbitData Labs

OrbitData Labs

Maintained by Community

Actor stats

2

Bookmarked

234

Total users

53

Monthly active users

a day ago

Last modified

Features

Limited permissions

Share

Extract visitor reviews and blog reviews from Naver Place — Korea's #1 local business platform. Cursor-based pagination with no server-side limit on visitor reviews, and blog posts beyond Naver's ~100-post cap with deepBlogSearch.

Features

  • Visitor reviews: cursor-based pagination, 50 per page, no upper limit on total reviews
  • Blog reviews: the posts Naver lists on the place (~100 at most, often far fewer)
  • More blog posts — deepBlogSearch: also finds posts that name the place in Naver Blog search, including for places that list no blog posts at all (Option 6 below)
  • TLS fingerprinting: curl_cffi with Chrome impersonation — bypasses bot detection
  • Structured output: voted keywords, visit counts, owner replies, star distribution
  • Two input modes: direct Place URL/ID or keyword search

What you get

Visitor review fields

FieldDescription
reviewTextFull review body (Korean)
reviewRating1-5 stars
reviewerNameAuthor nickname + profile URL + total review count
reviewDateReview creation date (YYYY-MM-DD)
visitDateActual visit date
visitCount"2번째 방문" (2nd visit)
reviewVerificationHow Naver verified the visit: 영수증 (receipt), 예약 (booking), 결제내역 (payment record); null if unverified
orderedItemMenu item or ticket the reviewer bought, when Naver links one
reviewerAverageRatingReviewer's average star rating across all their reviews; null if they don't rate
votedKeywords"음식이 맛있어요", "분위기가 좋아요", etc.
visitTagsVisit-context chips: reservation (예약 없이 이용), wait (바로 입장, 10분 이내), purpose (데이트, 일상), companions (친구, 아이); null if none
reviewPhotosPhoto/video thumbnail URLs (with includeReviewPhotos); null if none
reviewReplyTextBusiness owner's response

Blog review fields

FieldDescription
blogTitleBlog post title
blogContentsPost excerpt/contents
blogUrlLink to full blog post
blogSourceblog (Naver Blog) or cafe (Naver Café post)
blogMatchplace (Naver lists the post on the place) or name (found by deepBlogSearch)
bloggerNameBlogger name
blogDatePost date

Getting few or no blog reviews for a place? Naver lists at most ~100 posts per place, and some places list none. Turn on deepBlogSearch to add posts that name the place — see Option 6 below.

Aggregated statistics

When includeReviewStats is enabled (default), one stats row per place — a single dataset item.

Whole place (Naver's own statistics, the full review history):

  • totalVisitorReviews, averageRating — review count and average star rating
  • placeRatingDistribution — star histogram, e.g. {"4.5": 68, "5": 364}
  • placeThemes — what reviews talk about, e.g. {"맛": 782, "분위기": 184}
  • placeTopMenus — most-mentioned menu items, e.g. {"타코": 374}
  • placeKeywordDistribution — voted keywords ("음식이 맛있어요" …) with vote counts

This run (only the reviews you scraped):

  • ratingDistribution, keywordDistribution — the same histograms over this run's reviews
  • scrapedVisitorReviews, scrapedBlogReviews — how many rows this run collected

Histograms are ordered highest count first (stars ascending). Verticals Naver does not rate or tag — clinics, hospitals — return {}; null means Naver's statistics could not be fetched on this run.

Stats only, or blog reviews only

Set includeVisitorReviews to false:

  • with includeBlogReviews: true — blog reviews only;
  • with includeBlogReviews: false — just the stats row: one dataset item per place, the cheapest way to compare themes, menus and keyword votes across many places.

Input options

Option 1: Direct Place URLs or IDs

{
"placeUrls": [
"https://map.naver.com/v5/entry/place/37234567",
"37234567"
],
"maxReviewPages": 3
}

Option 2: Search by keyword

{
"searchKeywords": ["강남 맛집", "Seoul BBQ"],
"maxPlacesPerKeyword": 5,
"maxReviewPages": 1
}

Option 3: Every review since a date

{
"placeUrls": ["37234567"],
"visitedSince": "2026-09-01"
}

Visitor reviews whose visit date is on or after visitedSince. Naver lists reviews by visit date, so the run stops as soon as it reaches older visits — you only pay for the window you asked for. Without maxReviews, every review in the window is collected; with it, maxReviews still caps each place. A review written recently about an older visit is outside the window — for scheduled runs that must catch every new review, use Option 4.

Option 4: Scheduled runs — only new reviews

{
"placeUrls": ["37234567"],
"maxReviews": 50,
"onlyNewReviews": true
}

Each run looks at the same newest reviews it normally would, but skips the ones an earlier onlyNewReviews run already returned — they are not returned and not charged again. A re-run with nothing new returns 0 items; the run status says how many were skipped. The stats row (if enabled) gains skippedPreviouslyScraped.

  • History is kept per place in a key-value store in your Apify account (naver-place-reviews-seen by default). Use reviewHistoryName to keep separate histories for separate pipelines; delete the store or change the name to start over.
  • Only reviews that were actually stored count as delivered — if a run's cost limit cuts it short, the rest are offered again next time.
  • Don't run two onlyNewReviews runs for the same place at the same moment; both would see the same history and return the same new reviews.
  • Catching every new review — newReviewLookbackDays. By default a run only looks at the newest maxReviews, so a day with more new reviews than that, or a review written days later about an earlier visit, can slip past. Add "newReviewLookbackDays": 7 and each run instead reads every review visited from 7 days before that place's previous run up to now, returning the ones not delivered yet; maxReviews then caps the new reviews per place. The first run for a place uses the usual window and becomes the starting point. A bigger window reads more pages but costs nothing extra — only returned reviews are charged. Blog reviews keep the usual window.
  • Keeping your own database instead (n8n, Make, your backend)? Pass the IDs you already hold in excludeReviewIds — visitor reviewIds and/or blog blogUrls. They are skipped the same way (not returned, not charged), no history store needed, and it combines with onlyNewReviews.

Option 5: Only reviews with photos

{
"placeUrls": ["37234567"],
"maxReviews": 50,
"onlyPhotoReviews": true,
"includeReviewPhotos": true
}

Only visitor reviews with photos or videos. Naver filters them on its side, so text-only reviews are never fetched or charged, and maxReviews / visitedSince count photo reviews. Turn on includeReviewPhotos to get the URLs in reviewPhotos. Rows keep the place's full review count in placeTotalReviews; the stats row adds totalPhotoVisitorReviews (how many photo reviews the place has). Blog reviews are not affected. Combines with onlyNewReviews and visitedSince.

Option 6: More blog posts — search by place name

{
"placeUrls": ["37234567"],
"includeVisitorReviews": false,
"deepBlogSearch": true,
"maxDeepBlogPosts": 50
}

Naver lists only about 100 blog posts on a place, often far fewer — some places show none at all while bloggers have written about them for years. deepBlogSearch also searches Naver Blog for the place's name (newest first) and adds up to maxDeepBlogPosts posts per place, on top of the place's own.

What it added in a live check (2026-11-18):

PlacePosts listed on the placeAdded by name search
엄지척닭불고기010
핵밥 관동대점4012
갯벌의조개 여의도본점9040

Chains and franchises. Searching by branch names works the same way — each branch's name search keeps only posts that name that branch:

{
"searchKeywords": ["갯벌의조개 여의도본점", "갯벌의조개 성수역점"],
"maxPlacesPerKeyword": 1,
"maxReviews": 30,
"deepBlogSearch": true,
"maxDeepBlogPosts": 30
}
  • A post is kept only when its title or excerpt names the place: every word of the name, spacing ignored, a branch's trailing 점 optional (핵밥 관동대 counts for 핵밥 관동대점). Other branches of a chain (갯벌의조개 성수역점 for 갯벌의조개 여의도본점) and loose matches (bulgogi recipes for 엄지척닭불고기) are left out.
  • blogMatch marks every blog row place or name, so you can keep only the exact ones. Posts already listed on the place are not repeated.
  • Names shorter than 4 characters (e.g. 문의원) are skipped with a warning — searched alone they mostly match unrelated posts.
  • Works with onlyNewReviews / excludeReviewIds (keyed by blogUrl). The stats row adds scrapedNameSearchBlogReviews.

Which places finished — RUN_SUMMARY

Every run saves a free RUN_SUMMARY record in its default key-value store (not a dataset item, never charged). It is meant for pipelines that need to know whether to run again — for example when the run's maximum cost per run stops it partway through a long keyword list:

{
"status": "incomplete",
"stoppedByCostLimit": true,
"placeCounts": {"complete": 38, "no_data": 0, "cut_by_cost_limit": 1, "skipped_cost_limit": 26, "failed": 0},
"retryKeywords": ["핵밥 원주단계점", "..."],
"retryPlaceIds": ["1613941555", "..."],
"retryExcludeReviewIds": ["6915a0c2e4b0...", "https://blog.naver.com/...", "..."],
"unmatchedKeywords": [],
"unusablePlaceUrls": [],
"places": [{"placeId": "1613941555", "placeName": "...", "searchKeyword": "핵밥 원주단계점",
"status": "cut_by_cost_limit", "visitorReviews": 30, "blogReviews": 1, "declinedByCostLimit": 9}]
}
  • status per place: complete, no_data (nothing on Naver for it), cut_by_cost_limit (started, but the budget ran out), skipped_cost_limit (never started), failed (with reason).
  • retryKeywords / retryPlaceIds are exactly the places to run again. With onlyNewReviews on, a re-run returns only what is still missing from a cut place.
  • retryExcludeReviewIds lists what those places already delivered in this run. Pass it as excludeReviewIds in the re-run so a place that was cut partway is not charged again for the rows you already have.
  • API: GET https://api.apify.com/v2/key-value-stores/{defaultKeyValueStoreId}/records/RUN_SUMMARY.

Full input example

{
"placeUrls": ["37234567"],
"maxReviewPages": 10,
"visitedSince": "2026-09-01",
"includeVisitorReviews": true,
"includeBlogReviews": true,
"maxBlogReviewPages": 10,
"deepBlogSearch": false,
"includeReviewPhotos": false,
"onlyPhotoReviews": false,
"includeReviewStats": true,
"onlyNewReviews": false
}

Pagination guide

maxReviewPagesReviews collected
1~50
3~150
10~500
50~2,500
100~5,000

Output format

Visitor review

{
"reviewType": "visitor",
"placeId": "37234567",
"placeName": "맛있는 식당",
"placeCategory": "음식점",
"placeAddress": "서울특별시 강남구 ...",
"placeRating": 4.52,
"placeTotalReviews": 3162,
"placeUrl": "https://pcmap.place.naver.com/place/37234567",
"reviewId": "abc123",
"reviewText": "정말 맛있었어요! 분위기도 좋고 서비스도 훌륭합니다.",
"reviewRating": 5,
"reviewerName": "맛집탐험가",
"reviewerProfileUrl": "https://m.place.naver.com/my/...",
"reviewerReviewCount": 42,
"reviewerAverageRating": 4.9,
"reviewVerification": "영수증",
"orderedItem": "치즈닭갈비",
"reviewDate": "2026-05-15",
"visitDate": "2026-05-10",
"visitCount": "2번째 방문",
"votedKeywords": ["음식이 맛있어요", "분위기가 좋아요"],
"visitTags": ["예약 없이 이용", "바로 입장", "데이트", "연인・배우자"],
"reviewPhotos": null,
"reviewReplyText": "감사합니다! 또 방문해주세요.",
"reviewReplyDate": "2026-05-16",
"searchKeyword": null,
"language": "ko",
"scrapedAt": "2026-05-17T12:00:00.000Z"
}

Blog review

{
"reviewType": "blog",
"placeId": "37234567",
"placeName": "맛있는 식당",
"blogTitle": "강남 맛집 탐방기",
"blogContents": "오늘은 강남에 위치한...",
"blogUrl": "https://blog.naver.com/...",
"blogSource": "blog",
"bloggerName": "맛집블로거",
"bloggerHome": "https://blog.naver.com/blogger123",
"blogDate": "2026-05-10",
"blogThumbnailUrl": "https://...",
"isRepresentative": false,
"blogMatch": "place",
"scrapedAt": "2026-05-17T12:00:00.000Z"
}

Technical details

  • HTTP client: curl_cffi with TLS fingerprinting (impersonate="chrome")
  • Visitor review pagination: cursor-based via after parameter (not page-based)
  • Blog review pagination: offset-based (page + display=10)
  • Rate limiting: automatic retry with exponential backoff on HTTP 429
  • Deduplication: reviewId-based dedup for visitor reviews, id-based for blog reviews

Use cases

  • Korean market entry research — what do locals say about competitors?
  • Restaurant/cafe competitive intelligence
  • Customer sentiment analysis (NLP-ready structured data)
  • Brand reputation monitoring in Korea
  • Franchise and multi-branch monitoring — every branch's reviews, plus blog posts that name the branch (deepBlogSearch)
  • Academic research on Korean consumer behavior

Korea Data Suite by huggable_quote

ActorDescription
Naver MapBusiness listings & details
Naver BlogBlog posts & content
Naver ShoppingProduct data & prices
Naver NewsNews articles & headlines
Naver Place ReviewsThis actor
KRX StockKorean stock market data