Korean Reviews Bundle (Olive Young + Naver Shopping)

What does Korean Reviews Bundle do?

Korean Reviews Bundle collects product reviews from Korea's biggest beauty and e-commerce sites and returns them in one normalized schema. Feed it product URLs or IDs from Olive Young and Naver Shopping, and it writes clean, structured review items to a dataset. Ratings, review text, dates, photo flags, repurchase markers, and (where the site exposes them) helpful-vote counts and reviewer skin profiles all land in the same shape, no matter which site they came from.

Because it runs on the Apify platform, you get scheduling, API access, proxy rotation, monitoring, and easy export to your pipeline out of the box.

Supported sources

Source	Status	Target format	Notes
Olive Young	Available	goodsNo (A + 12 digits) or product page URL	Full data: helpful votes and reviewer skin profile included
Naver Shopping	Available	Smart Store / Brand Store product URL or bare product number	No helpful-vote count and no skin profile in the source data
Coupang	Not supported in v0.1	—	Coupang URLs are skipped with a reason in the run summary and are never billed

Why use Korean Reviews Bundle?

• One schema, two sites. You don't have to reconcile two different review formats. Every item follows the same field layout, so downstream analysis, dashboards, and LLM pipelines just work.
• K-beauty and Korean commerce intelligence. Track sentiment on Olive Young rankings and Naver Smart Store listings in a single run.
• Respectful by design. The Actor reads only public review data, never logs in, and never collects reviewer names, IDs, or profile images.
• Pay for results, not compute guesswork. Pay-per-event pricing means you only pay when reviews actually come back, in blocks of 50 returned reviews.

How to use Korean Reviews Bundle

1. Open the Actor in Apify Console and go to the Input tab.
2. Add your product IDs or URLs to Product IDs or URLs. Mixing sources in one run is fine.
3. Optionally set Max reviews per product and pick a Sort order.
4. Click Start and watch the dataset fill up.
5. Download the results or pull them through the Apify API.

No coding required. If you prefer code, call the Actor from the Apify API or SDK with the same JSON input shown below.

Input

Field	Type	Default	Description
productIds	array of strings	(required)	Olive Young goodsNo or product URLs, and Naver product URLs or bare product numbers. Any other site (including Coupang) is not supported in v0.1: such targets are skipped with a reason in the SUMMARY record and bill nothing.
maxReviewsPerProduct	integer	100	Stop after this many reviews per product (max 5,000).
sortType	string	USEFUL_SCORE_DESC	One of USEFUL_SCORE_DESC, LATEST, SCORE_DESC, SCORE_ASC.
proxyConfiguration	object	Apify residential proxies	Residential proxies are recommended for platform runs.

Example input:

{
    "productIds": ["A000000223414", "https://brand.naver.com/mediheal/products/11983780143"],
    "maxReviewsPerProduct": 100,
    "sortType": "USEFUL_SCORE_DESC"
}

Output

Each review is one dataset item. Example:

{
    "source": "oliveyoung",
    "productId": "A000000223414",
    "productName": "Mediheal Tea Tree Essential Mask 10 sheets",
    "optionName": "single pack",
    "reviewId": "23845123",
    "rating": 5,
    "text": "Calmed my skin overnight. Already on my third box.",
    "date": "2026-08-29",
    "helpfulCount": 41,
    "hasPhoto": true,
    "photoPaths": ["https://image.oliveyoung.co.kr/uploads/images/goodsDetail/review/abc123.jpg"],
    "isRepurchase": true,
    "reviewType": "PHOTO",
    "reviewer": { "skinType": "oily", "skinTone": "21", "skinTroubles": ["trouble"], "isTopReviewer": false },
    "scrapedAt": "2026-09-06T03:14:22.118Z"
}

You can download the dataset in various formats such as JSON, HTML, CSV, or Excel.

Data fields

Field	Description
source	oliveyoung or naver_shopping
productId	Product identifier you requested
productName	Product name as shown on the site, or null
optionName	Purchased option/variant, or null
reviewId	Site's own review ID
rating	Star rating (number), or null
text	Review body text
date	Review date, YYYY-MM-DD (KST calendar date for Naver)
helpfulCount	Helpful/useful vote count; always 0 for Naver (not exposed)
hasPhoto	Whether the review has photos
photoPaths	Photo URLs attached to the review
isRepurchase	Repurchase flag
reviewType	Site-specific review type, or null
reviewer	Skin profile (Olive Young only); Naver returns empty/null values. No names, IDs, or profile images, ever
scrapedAt	Timestamp when the item was collected

How much does it cost to scrape Korean reviews?

The Actor uses pay-per-event pricing with a single event, reviews-50, at USD 0.15 per event.

• One event covers up to 50 returned reviews, counted per run across all products. Every started block of 50 is billed as a whole event: 1–50 reviews = 1 event ($0.15), 51–100 = 2 events ($0.30), 137 reviews = 3 events ($0.45). That works out to at most USD 3.00 per 1,000 reviews.
• Billing counts reviews returned, not reviews requested. A run that returns nothing bills nothing; unsupported or unrecognized targets bill nothing.
• There is no start fee and no per-item fee on top of reviews-50. The event price covers the platform usage of the run (compute, proxy traffic, storage); you are not billed separately for it.
• You can cap the total charge of a run with the maximum cost per run option when you start it; the Actor stops between products once the cap is reached, and events above the cap are never billed.

New Apify accounts can try the Actor within the free tier before spending anything.

Tips and advanced options

• Set maxReviewsPerProduct honestly. It's the simplest way to cap both cost and run time.
• Use residential proxies on the platform. These sites rate-limit datacenter IPs aggressively.
• Batch products from the same source. A run still works with mixed targets, but keeping sources together makes logs easier to read.
• Sort by USEFUL_SCORE_DESC when you want the most informative reviews first, LATEST when you need recency for trend monitoring.

FAQ, disclaimers, and support

Is scraping these reviews legal?

The Actor reads only public review data that any visitor can see without an account. It never logs in, and it never collects reviewer names, IDs, or profile images. Normalization drops those fields deliberately, and the test suite asserts the drop. Scraping publicly available data is generally lawful, but you're responsible for complying with each site's terms of service and the privacy laws that apply to your use case.

What are the known limitations?

• Naver Shopping exposes no helpful-vote count on its review list, so helpfulCount is always 0 for Naver items. It also exposes no reviewer skin profile, and it serves 20 reviews per request, so large Naver targets need more requests than you'd expect. Olive Young fills both helpful votes and skin profile.
• Review photos come back as URLs, not downloaded files.
• Coupang is not supported in v0.1. Coupang product URLs are recognised, skipped with a reason in the SUMMARY record, and never billed.
• Sites change their markup without notice; if a source breaks, open an issue.

Feedback

Found a bug or want another Korean source added? Use the Issues tab. Custom scraping solutions and integrations are available on request.