Wenqiong Z

Data Analyst

Shopee • Tiempo completo

Aug 2025 - Oct 2025 • 2 mos

Analyzed Shopee e-commerce user shopping data across four dimensions: user profiling, product preferences, purchase behavior, and user loyalty. Used MySQL for data cleansing, metric construction, and user segmentation; built interactive Tableau dashboards to identify high-value and at-risk churn users, supporting refined operations and marketing strategies. Cleansed user, order, product, and transaction data; constructed a consumption analysis wide table; designed 10+ core business metrics. Performed RFM-based user segmentation (High-Value / Potential / General / Churn-Risk) to identify high-value and at-risk users. Built Tableau dashboards for user profiling, sales trends, product preferences, and purchase behavior, supporting cross-analysis by time, region, category, and user tier. Analyzed regional consumption preferences and best-selling categories in cross-border e-commerce context, providing strategic recommendations for coupon distribution, membership operations, churn recovery, and category management. Outcome: Completed end-to-end analytics pipeline from data cleansing → SQL metric modeling → RFM segmentation → Tableau visualization, delivering 4 user segments and 10+ core metrics to support Shopee user growth, refined operations, and marketing conversion.

Data Scientist

Tencent Holdings • Tiempo completo

Jun 2021 - Jun 2024 • 3 yrs

Led data platform and business analytics projects, overseeing multi-source data processing, metric definition standardization, data modeling, tag system construction, data quality monitoring, and thematic analysis reporting. Processed 10+ categories of heterogeneous data using HiveSQL, Python, and Spark, cleansing and modeling tens of millions of records. Standardized 200+ field definitions and metrics, built 100+ thematic data models and tags to support user profiling, business statistics, data retrieval, and cross-project reuse. Conducted statistical analysis and machine learning modeling (clustering, decision trees, logistic regression, random forests) for user segmentation, anomaly detection, key feature analysis, and impact factor mining. Established 50+ data quality validation rules, integrating rule-based models and anomaly detection to identify missing data, duplicates, abnormal fluctuations, and definition conflicts. Delivered 30+ thematic analysis reports, metric documentation, and model documents, improving core data consistency by ~30% and reducing similar project modeling cycles by 20%–30%.