Hwahae Scraper — K-Beauty Rankings, Ratings & Prices

Scrape Korean beauty product rankings from Hwahae (화해) — Korea's #1 cosmetics review app, built on 10M+ real user reviews. Point it at Hwahae's global English rankings and get back clean, structured JSON for every ranked product: name, brand, category, rank, star rating, review count, USD price, discount rate, and review-topic keywords.

Hwahae is the K-beauty consumer authority in Korea — its rankings reflect what Korean users actually rate highest, a review-driven signal distinct from what a store is merely pushing. If you need to know which K-beauty products are winning in Korea right now, this is the source, in English and USD.

Try it free. Apify's free plan includes $5 of monthly platform credit — roughly 2,500 results from this actor, no credit card required. Set your input, click Start, and export JSON/CSV/Excel.

Privacy-first. No personal data is ever collected — no reviewer names, IDs, or profiles. Product, brand, ranking, rating, and price data only.

---

What it does

• Reads Hwahae's public, server-rendered global rankings (English) — works from most IPs, including datacenter.
• Extracts each ranked product from the page's embedded __NEXT_DATA__ blob.
• Emits one clean record per product against a stable schema.

Input

Field	Type	Default	Description
categoryThemeIds	string[]	["5102"]	Hwahae ranking category theme_ids. Empty = default "All Categories" trending list. E.g. 5102 All Categories, 5103 Skincare, 5177 Cleansers, 5296 Suncare, 5338 Face Makeup.
maxProducts	integer	40	Max products to scrape across all requested rankings.
proxyConfiguration	object	RESIDENTIAL	Proxy settings (optional for ranking data).

Input example

{
  "categoryThemeIds": ["5102", "5103"],
  "maxProducts": 40
}

Output

One dataset item per product. This is real output from a live run of the global rankings (see samples/sample-output.json for 39 real records):

{
  "source": "hwahae",
  "product_id": "1918760",
  "name": "MUNG BEAN pH-BALANCED CLEANSING FOAM",
  "brand": "beplain",
  "category": "All Categories",
  "rank": 1,
  "rating": 4.63,
  "review_count": 49461,
  "price_usd": 9.96,
  "currency": "USD",
  "capacity": "80mL ≈ 2.71 fl. oz.",
  "discount_rate": 40,
  "keywords": ["알갱이있는", "각질제거잘되는", "미끌거리지않는"],
  "url": "https://www.hwahae.com/en/products/MUNG-BEAN-pH-BALANCED-CLEANSING-FOAM/1918760",
  "image_url": "https://img.hwahae.co.kr/products/1918760/1918760_20230628115919.jpg",
  "scraped_at": "2026-07-09T12:00:00+09:00"
}

Field notes

• product_id — Hwahae's stable product identifier.
• price_usd / currency — Hwahae's live exchange-rate price for the global store (USD). discount_rate is the current percent discount when present.
• rating — 1–5 star average from real Hwahae reviews; review_count is the number of reviews behind it. Together these are the core "what Korea actually rates highly" signal.
• keywords — Hwahae's top review "topics" for the product (the aspects reviewers mention most), as native Korean topic labels — a compact, ready-made input for aspect/sentiment analysis.
• rank — 1-based position within its ranking list.

Use cases

• Market research for K-beauty market entry — pull the current top-rated products by category with real ratings, review counts, USD prices and review-topic keywords before launching a competing SKU.
• Trend tracking — snapshot the rankings on a schedule to see what's rising and falling in Korean consumer favor over time.
• Pricing & dropshipping — track top K-beauty products, their USD price and discount rate.
• Aspect/sentiment signal — thousands of privacy-clean review-topic keywords feed straight into an NLP/LLM pipeline. No PII to scrub.

Proxy & rate limits

The ranking data — the product this actor sells — reads from Hwahae's server-rendered __NEXT_DATA__ and works from most IPs, including datacenter (the 39-record sample here was produced with no proxy). A residential proxy is only worth enabling for high-volume runs. The actor runs politely: effectively concurrency 1 with ≥600 ms between requests, and if a proxy tunnel hangs it retries directly so a bad tunnel never hangs the run.

Note on category coverage. Only "All Categories" trending views embed their product list in server-rendered HTML; narrow leaf-category rankings load client-side and aren't in the SSR blob. The actor scrapes every ranking view that embeds products and skips the rest.

Known limitation — ingredients & review text (not currently available)

Hwahae's flagship ingredient lists and full review text live only on product detail pages (hwahae.com/en/products/<slug>/<id>), which are protected by a JavaScript bot-challenge that returns HTTP 202 with an empty body — even through residential proxies (verified). There is an experimental includeIngredients input, but it is off by default and normally returns empty ingredient/review fields; the actor falls back to the ranking record. Delivering ingredient/review-text data reliably would require full browser automation to clear the challenge — a possible future enhancement, not a current feature. This actor is sold on its ranking, rating, review-count and price data, which are reliable.

Legality & data ethics

• Only public ranking data is collected — no login, no private endpoints.
• No personal data. Reviewer names, IDs, and profiles are never read or stored.
• You are responsible for using the output in compliance with Hwahae's terms and applicable law. Because no personal data is emitted, output is safe to feed directly into analytics and LLM pipelines.

❓ FAQ

Is it legal to scrape this data? This actor collects only public, non-personal ranking data — the same rankings any visitor sees on Hwahae's global site without logging in. No personal data is collected (no reviewer names, IDs, or profiles). You are responsible for how you use the data; see the legality note above.

What does it cost in practice? $2.00 per 1,000 results (launch pricing) + a few cents of platform usage. Example: a 500-result daily pull costs about $1.00/day. Apify's free $5 monthly credit covers ~2,500 results.

Do I need to configure proxies? No — the default settings work out of the box, and the ranking data works from most IPs, including datacenter. Optionally, enable a residential proxy for high-volume runs.

How fresh is the data? Every run scrapes the live site at run time. Schedule the actor for recurring snapshots.

What is NOT included? Ingredient lists and full review text are not currently available — they live on bot-challenge-protected detail pages (see the known limitation above). Also, only ranking views that embed products in server-rendered HTML are scraped; narrow leaf-category rankings that load client-side are skipped.

🇰🇷 More Korean data actors

This actor is part of a suite of Korean-platform scrapers by the same maintainer:

• oliveyoung-scraper — Olive Young K-beauty bestsellers, prices and reviews.
• glowpick-scraper — Glowpick Korean cosmetics award rankings and ratings.

Browse all: apify.com/kdatafactory

Local development

npm install
# put input in ./storage/key_value_stores/default/INPUT.json, then:
npm start   # results land in ./storage/datasets/default

# polite lite test against the live site (<=2 pages, ~40 products):
HWAHAE_SCRAPER_LITE_TEST=1 npm start

---

If this actor saves you time, a rating on the Store page helps a solo maintainer a lot. Found an issue? Open it in the Issues tab — I respond fast.