NetEase Big Data Ecosystem and User Profiling Practices

Introduction – NetEase’s big‑data ecosystem spans billions of daily active accounts across entertainment, e‑commerce, education, and more, providing multi‑dimensional user behavior data for comprehensive user‑portrait construction and both internal and external commercial solutions.

1. NetEase Data Overview

Data volume exceeds hundreds of millions of users with daily active accounts in the hundred‑million range.

Complex, multi‑industry product lines covering games, education, e‑commerce, and entertainment.

High‑quality user tag coverage (>70%).

Provides thematic domain solutions such as participation, traffic, location, and relationship domains.

2. Rich Product Lines – NetEase’s platforms collect extensive user behavior data from various apps, enabling the construction of a global user portrait that serves numerous internal business scenarios and explores external productization.

3. Full‑Link Data Platform Matrix

Bottom layer: log and raw data sources.

Processing layer: data warehouse, offline/real‑time tag generation, mining algorithms, and monitoring.

Upper layer: business applications such as intelligent analysis, growth operations, recommendation, and search.

4. User‑Profile Center Classification

Basic tags (e.g., gender, age, education, location, device, membership).

Behavior tags (e.g., region, ad interaction, search, playback, click, purchase).

Preference tags (e.g., shopping, mobile, home‑decoration, education, entertainment, finance, gaming).

Predictive tags generated by algorithms (e.g., travel intent, car ownership).

5. IDMapping – Techniques for unifying multiple device IDs to a single user identity, using both engineering (SDK) and data‑layer (rule‑based and graph‑based) methods, addressing challenges such as multi‑device users, device decay, and abnormal data.

6. Regional Domain – Extracts user needs (e.g., car ownership, frequent store visits) from Wi‑Fi, IP, and device data, also supporting anti‑fraud use cases.

7. User‑Profile Management & Storage – Evolution from string concatenation to JSON‑array tag management, highlighting redundancy issues and ongoing redesign efforts.

8. Quality Assurance & Governance

Assigning tag owners for rapid issue response.

Optimizing end‑to‑end tag lifecycle processes.

Testing and monitoring tag quality before release.

Platformizing tag production, processing, and application.

9. Practical Cases

Marketing: audience selection and insight.

Recommendation & Search: data input for algorithms.

Growth Operations: support for user research and data‑driven operations.

Advertising: precise audience targeting.

Intelligent Anti‑Fraud: detecting fraudulent behaviors, improving risk identification by 6%.

10. Real‑Time Full‑Link Recommendation – Real‑time user‑portrait integration across services, enabling cold‑start handling, cross‑business data fusion, and knowledge‑graph‑enhanced user tracking.

Conclusion – NetEase’s user‑portrait platform dramatically boosts data productivity, consolidates methodology and products, empowers numerous internal scenarios, and explores external commercialization pathways.