Tencent Maps Scraper: China Places & Phones in 2026

The map inside WeChat

Google Maps is not where mainland China looks for a shop. People use domestic maps, and Tencent Maps (腾讯地图, also known as QQ Maps) is one of the three big ones, with Amap and Baidu Maps. It is also the map built into WeChat: when someone shares a location or looks for a restaurant in the app, Tencent’s listings are what they see.

Seen from abroad, it is the closest public directory of a Chinese city. Reading it at scale is the hard part.

What you can extract

One row per place: a business, public service or landmark in mainland China.

• Identity: stable place id, map link, and the name with its branch
• Category: name, path and the map’s 6-digit category code
• Contact: every phone number listed, landline or mobile
• Ratings: rating out of 5, spend per person in yuan, popularity score
• Location: address down to district, administrative codes, floor and parent mall
• Hours: as text and by day
• Coordinates: WGS-84, GCJ-02 and the entrance drivers are routed to
• Platform ids: Dianping, Amap, Meituan, Ctrip, Fliggy, eLong and WeChat mini program
• Place details, on demand: photos, the top 10 dishes with prices, visitor tags with counts and up to 10 reviews

The technical challenge

The map is built for domestic users:

• Offset coordinates: mainland maps use GCJ-02, which lands a few hundred meters off on GPS-based maps; Google Maps or QGIS need WGS-84.
• Search caps: about 800 places per term and city at most, and a brand name returns only the branches the map features. Wide sweeps need several category words.
• Chinese text: names, addresses and reviews stay in Chinese, so filter on codes rather than names.
• Short review samples: a place page shows 10 reviews at most.
• Mainland only: searches in Hong Kong, Macau or Taiwan come back empty.

DIY or a ready-made Actor?

Our Tencent Maps Scraper on Apify takes search terms and cities or provinces in English or Chinese (no city searches all of China, busiest cities first), plus map.qq.com links and place ids, in one run. Place details and reviews are opt-in. No Tencent account, API key or Chinese needed. Pay-per-result at about $0.0035 per place.

A practical workflow

1. Search with category words: 火锅, 咖啡 or hotel return far more places than a brand name; add related terms to widen a sweep.
2. Filter on codes: keep rows by categoryCode and adcode, and group the shops of one mall on parentPlaceId.
3. Join and enrich: match places to Dianping, Amap or Ctrip data on the id columns, then feed contacts into an automated lead generation pipeline.
4. Watch by id: schedule the place ids you follow weekly with reviews on, and diff ratings, hours and new reviews between runs.

Use cases that pay off

• Market sizing: density, rating and spend per person by district, across cities.
• Local prospecting: call lists of restaurants, hotels, clinics or hospitals.
• Hotel matching: Ctrip, Fliggy and eLong ids join the map to rate and review data.
• Online and offline footprint: a chain’s branches, grouped by brandId, beside its JD.com or Taobao listings.

The legal boundaries

• Public data only: the scraper reads what the map shows anyone and bypasses no login, paywall or CAPTCHA.
• Some personal data: small businesses sometimes list a personal mobile, and reviews carry author names, often masked; handle both under PIPL, GDPR or CCPA.
• Outreach rules: China’s rules on unsolicited marketing apply to cold calls.
• Verify before use: phones, hours and prices change.