JD.com Product Scraper 🇨🇳🛒

Scrape JD.com (京东) — China's #2 e-commerce platform — for product pricing, brand data, seller information, reviews, ratings, and JD自营 (self-operated) vs 第三方 (third-party) classification. Built for cross-border e-commerce sellers, supply-chain analysts, brand monitoring teams, and Chinese-market intelligence operators who need clean, structured product data without spinning up their own Chinese-internet scraping stack.

JD.com is the workhorse of Chinese e-commerce for electronics, appliances, FMCG, and premium goods. Where Taobao/Tmall is the long tail, JD is curated, fast-shipping, and largely first-party — which makes its pricing, stock signals, and review data unusually high-signal for sourcing and competitive intelligence. This actor turns search.jd.com and item.jd.com into a structured product dataset you can ingest into BigQuery, Snowflake, Postgres, an internal dashboard, or a Claude/GPT-powered analyst workflow with one HTTP call.

---

Who this is for

• Cross-border e-commerce operators sourcing products in China to resell on Amazon, Shopify, eBay, TikTok Shop, or Temu — and who need real-time JD pricing as a price-floor reference.
• Supply-chain analysts tracking JD自营 first-party listings as a leading indicator of category demand and inventory health.
• Brand monitoring & MAP enforcement teams watching how their brand appears on JD — unauthorized resellers, gray-market listings, MAP violations, and counterfeits.
• Chinese-market intelligence consultancies producing competitive landscape reports on consumer categories (mobile, appliances, beauty, FMCG, maternity, fresh).
• Investors and equity researchers tracking JD-listed brand performance, discount intensity, and review velocity as alt-data for JD.com (NASDAQ: JD) and competitor coverage.
• AI agents and LLM pipelines that need a clean tool_call against JD product data without manually managing a Playwright cluster behind the Great Firewall.

---

What you get

For every product matched, the actor returns a structured record:

Field	Type	Description
sku_id	string	JD product ID — stable, used in item.jd.com/{sku_id}.html
name	string	Full product title (Chinese, occasionally English)
brand	string	Brand parsed from the product spec sheet
category	string	Breadcrumb path (e.g. 家用电器 > 大家电 > 平板电视)
current_price_cny	float	Live price in CNY (¥)
original_price_cny	float	Pre-discount / MSRP price in CNY
discount_pct	float	Computed discount percentage
seller_type	string	JD自营 (self-operated) or 第三方 (marketplace)
seller_name	string	Shop / merchant name
rating_avg	float	Average product rating (1–5 stars)
review_count	int	Total review count
is_jd_logistics	boolean	Whether 京东物流 (JD Logistics) ships the SKU
image_url	string	Primary product image
product_url	string	Canonical item.jd.com/{id}.html link
data_source	string	Always jd.com for live records

---

Input filters

Field	Default	Notes
keywords	["手机"]	Array of search terms. Chinese keywords work best.
category	""	Optional JD category ID (e.g. 9987,653,655).
min_price	0	Minimum CNY price filter.
max_price	0	Maximum CNY price filter.
sort	default	default / sales (销量) / price (low→high).
limit	30	Total products across all keywords.

Example input:

{
  "keywords": ["空气炸锅", "扫地机器人"],
  "min_price": 200,
  "max_price": 2000,
  "sort": "sales",
  "limit": 100
}

---

Pricing

$0.01 per run (platform overhead) + $0.05 per product record.

• 30 products ≈ $1.51
• 100 products ≈ $5.01
• 1,000 products ≈ $50.01

This is volume-friendly e-commerce pricing — designed for sourcing teams running daily category sweeps and brand-monitoring teams running rolling 7-day SKU panels. You only pay for products actually returned; if JD blocks a query, you'll see a data_source: jd.com:maintenance stub record (one per failed keyword) and no overcharge.

---

How it works under the hood

JD.com has a non-trivial anti-bot layer on plain-HTTP search XHRs. The actor uses a stealthed headless Chromium (Playwright) to fetch search.jd.com/Search pages, scrolls to trigger lazy-loaded cards, parses out SKU IDs and card-level pricing, and then enriches each SKU via the public item.jd.com/{sku_id}.html endpoint (which serves real HTML to plain HTTP clients) and the JD comment summary API for ratings.

Concurrent enrichment uses a 4-way semaphore so you won't hammer JD into rate-limiting you. If JD hard-blocks a keyword, the actor returns a maintenance stub instead of crashing — your downstream pipeline will see the failure clearly without 5xx alerts.

---

Comparison vs alternatives

Product	JD.com Product Scraper	Alibaba 1688	Shopify Markets	Sensor Tower	Sinotrust
Focus	JD.com retail SKUs, real consumer-facing prices	Wholesale / B2B 1688 listings	Your own Shopify storefronts in CN region	Mobile app intelligence	Premium consulting reports
Real-time?	Yes — live JD prices and reviews on every run	Wholesale-tier, slower refresh	Your own data, not market data	Daily	Quarterly reports
Brand monitoring	Yes — MAP, gray market, unauthorized resellers	No (wholesale only)	Your storefronts only	Apps only, no products	Custom engagements
JD自营 vs 第三方 flag	Yes	N/A	N/A	N/A	Custom
Review counts + ratings	Yes	Limited	N/A	App reviews only	Custom
Self-serve API	Apify run = HTTP call	Manual / aggregator APIs	Shopify Admin API	Yes	No
Cost	$0.05 / product, pay-per-result	Aggregator $$$, often $5k+/mo	$0 but only your data	$$$$ ($60k+/yr enterprise)	$$$$$ bespoke
Cross-border friendliness	Plug-and-play Apify run	Wholesale-oriented	Storefront tool, not intel	Apps not products	Reports, not data

If you're doing cross-border retail arbitrage, MAP enforcement, or category demand analysis, JD.com Product Scraper gives you live retail-side data that 1688 / Shopify / Sensor Tower / Sinotrust simply don't expose at this price point.

---

Use cases & playbooks

1. Cross-border arbitrage price floor

Run daily across your top 50 SKUs. Compare current_price_cny × FX × shipping × tariff against your Amazon/TikTok Shop landed cost. Auto-flag SKUs where JD price drops > 10% week-over-week — those are the ones where your retail margins are about to collapse.

2. JD自营 vs 第三方 share-of-shelf

Track the seller_type mix in your category over time. A rising 第三方 share usually signals JD ceding direct distribution to marketplace sellers — which is a sourcing opportunity but also a brand-control risk.

3. Brand-protection / MAP enforcement

For each of your brand's SKUs, monitor seller_name. Flag any non-authorized seller. Combine with current_price_cny to detect MAP violations in real time.

4. Review-velocity demand signal

Track review_count deltas weekly per SKU. SKUs with accelerating review counts are gaining demand momentum — leading indicator for restocks and category trend reports.

5. New-product launch monitoring

Run weekly with sort=default and a tight keyword list. Diff against prior runs to surface newly-listed SKUs. Great for category-launch tracking and competitive intelligence.

---

Companion actors (NexGenData fleet)

This actor is part of a Chinese-market / Asia-Pacific intelligence stack. Combine it with:

• Weibo Hot Search Tracker — 微博热搜 trending topics, viral signals, and consumer attention as a leading indicator for JD demand.
• Xiaohongshu Trends Tracker — 小红书 / RED note trends — early demand signals for beauty, fashion, lifestyle SKUs before they show up on JD.
• China Trends Tracker — cross-platform Chinese trend aggregator for omni-channel demand mapping.
• Eastmoney China Stock Screener — A-share fundamentals for the listed brands (e.g. JD-listed appliance OEMs) behind the SKUs.
• HKEX Hang Seng Stock Screener — Hong Kong-listed brand parents (JD.com itself, Tencent-backed brands, etc.).
• Finance MCP Server — drop the above into a Claude/GPT agent via MCP for natural-language e-commerce intelligence.

The full stack lets you trace consumer attention (Weibo / Xiaohongshu) → product demand (JD) → brand financials (Eastmoney / HKEX) in a single agent workflow.

---

Operational notes

• Anti-bot: JD's defenses are real but not draconian. Expect ~85–95% success on normal keywords. Niche or restricted categories (alcohol, pharma) sometimes return shorter result sets.
• Geography: all prices are CNY, China-region pricing. Cross-border ¥-USD conversion is your job downstream.
• Compliance: this actor accesses only publicly-served pages on JD.com. Use responsibly and in accordance with JD's terms and your jurisdiction's data-protection regime.
• Throttling: the actor self-throttles enrichment to 4 concurrent requests. Don't run more than ~5 parallel actor runs against JD or you risk seeing maintenance stubs.

---

Roadmap

• Multi-category sweep mode (auto-walk a category tree)
• Stock-availability flag (in stock / 预售 / 缺货)
• Historical price tracking via JD's price-history endpoint
• Review-text scraping (premium add-on)

---

Try it now

Click Run with the default input, get 30 real JD.com products in under 60 seconds, and pipe them into your existing data pipeline. Smoke-tested daily; pricing is transparent and pay-per-result.

Run the JD.com Product Scraper → and start your free trial today. Sign up via the link, run any NexGenData actor, and the free $5 of Apify platform credit covers your first 100 JD products — risk-free. Build your Chinese e-commerce intelligence stack with NexGenData.

---

Built by NexGenData. Questions or custom data needs? Reach out via the Apify Store contact form.

---

Why this beats Tmall, 1688, Pinduoduo, and Crunchbase enrichment alone

Capability	NexGenData JD.com Scraper	1688 Alibaba Browse	Pinduoduo App	Tmall Open Platform	Crunchbase / PitchBook enrichment
JD.com SKU-level price + discount	✅ CNY current + original	❌ wholesale only	❌ different catalog	❌ different storefront	❌ company-level only
JD自营 vs 第三方 seller-type flag	✅	❌	❌	❌	❌
JD Logistics flag (is_jd_logistics)	✅	❌	❌	❌	❌
Review count + average rating	✅	❌ public	⚠️ limited	⚠️ closed API	❌
Bulk export (JSON / CSV / Parquet)	✅ via Apify dataset	❌ HTML only	❌ app only	⚠️ closed merchant API	✅ but at $1500-3000/seat/mo
Cross-border friendly (no Chinese phone / 营业执照)	✅ public endpoints	⚠️ partial	❌ app-only auth	❌ requires Tmall merchant account	n/a
Pricing	$0.05 / SKU, no subscription	Manual browse	n/a	Enterprise contract	$1500-3000/seat/mo flat

For competitive-intel teams comparing brand presence on China's #2 e-commerce platform, this gives you the same SKU-level depth Bloomberg China Consumer Intel and SimilarWeb Shopper Intelligence charge five-figure annual fees for — without the contract.

---

About NexGenData

NexGenData publishes 200+ buyer-intent Apify actors covering SEC filings (Form 4 insider buys, Form D, 13F holdings, 8-K material events, Schedule 13D/G activist tracker), YC alumni, Delaware DOC, lead generation, competitive intelligence, stock fundamentals across 30+ exchanges (SGX, ASX, TSX, B3, TWSE, KOSPI, HKEX, BSE Beijing, STAR Market, NSE India, HOSE Vietnam, Saudi Tadawul, JSE), property & macro data (Singapore HDB / URA, US Treasury yields, FX rates, commodity futures), and Chinese-market intelligence (JD, Douyin, Weibo, Xiaohongshu).

All actors are pay-per-result — you only pay for rows you keep. No subscription, no seat licence, no annual contract.

Browse the full catalog and start your free trial: https://apify.com/nexgendata?fpr=2ayu9b

Sign up via that link and the free Apify platform credit covers your first hundred-plus rows on every actor — risk-free evaluation.