JD.com Scraper: Prices, Barcodes & Shop Data in 2026

China’s largest retailer hides its prices

JD.com (京东) is China’s largest retailer by revenue. It sells its own stock as “self-operated” listings and hosts marketplace sellers, much like Amazon does outside China, and it leads in electronics, appliances and branded groceries. For brands, distributors and analysts, JD is the reference shelf for what Chinese consumers actually pay.

Reading that shelf at scale is another matter. Signed-out visitors see prices masked as 6??9, every title is in Chinese, and one phone can exist as a dozen variant SKUs.

What you can extract

One row per product SKU:

Real price in yuan, with installment offers and subsidy badges
Brand, model and UPC/EAN barcode
Shop: name, shop ID, and whether JD sells the product itself (self-operated)
Reviews: review count, good rate and top review tags
Variants: every SKU with its color, size and spec
Stock: stock state and quantity
Delivery: delivery promise, shipping fee, return policy and warranty
Sales rank, weight, dimensions and photos

Individual review texts are out of scope: JD shows them to signed-in shoppers only.

The technical challenge

JD spreads its data across pages that behave differently:

Masked prices: the product page hides the price from signed-out visitors; search results show it in full. The price has to come from search or JD’s own price page, then be joined to the product-page fields.
Variant explosion: one product group can hold dozens of SKUs: a 50-product phone run listed 777 of them.
Duplicates: the same product turns up under several keywords and again in its brand’s shop; deduplicate on SKU or count it twice.
Chinese text: titles and delivery promises stay in Chinese, so you match on IDs, barcodes and model numbers.

DIY or a ready-made Actor?

Our JD.com Scraper on Apify takes keywords in Chinese or English, JD product, search or shop addresses, SKU ids or shop ids, in any mix, with up to 5,000 products per keyword or shop and an optional price band in yuan. Turn product pages off for fast price checks; leave them on for brand, barcode, stock and delivery. No JD account needed. Pay-per-result at about $0.0022 per product.

A practical workflow

Start broad: run a category keyword with product pages off to size the market by price, shop and review count.
Narrow to SKUs: rerun the products that matter by SKU id with details on, adding barcode, model, variants and stock.
Join on barcode: match rows to your catalogue or a distributor price file on upc and model; when only a Chinese title is available, LLM-powered extraction helps normalize it.
Schedule and diff: rerun weekly and compare on SKU id to catch repricing, launches and delistings; our e-commerce price monitoring guide covers the alerting side.

Use cases that pay off

Reseller monitoring: see who sells your brand on JD besides your official store, and at what price.
Competitor shop tracking: pull a rival’s JD shop inside a price band and diff it weekly.
Catalog matching: barcodes and model numbers join JD listings to your product master without reading Chinese.
Retail versus wholesale: compare JD shelf prices with 1688 wholesale prices for the same category before you source.

The legal boundaries

Public data only: the scraper reads what any signed-out visitor sees and bypasses no login, paywall or CAPTCHA.
No shopper data: rows carry products, shops and aggregate review figures, nothing about individual shoppers or reviewers.
Terms and local law: JD’s terms of service apply, and China’s PIPL sits alongside GDPR and CCPA wherever personal data enters your pipeline.
Verify before contracting: scraped prices are informational; check them before regulated or contractual use.