Sign in to view source links and access this dataset
Description
Over 2.5 million customer records, including 457,298 who made purchases, and 31,868 SKUs from China's largest retailer JD.com for the month of March 2018. The dataset was provided to support the 2020 MSOM Data Driven Research Challenge, offering transaction-level data for econometric and data-driven modeling.
Use Cases
Analyze purchase frequency and basket composition across the 31,868 SKUs to understand consumer behavior patterns.
Model customer lifetime value by segmenting the over 2.5 million customers, focusing on the 457,298 purchasing cohort.
Develop demand forecasting models for specific SKUs using transactional timestamps and customer purchase data.
Strengths
Large-scale data covering over 2.5 million customers and 31,868 distinct products (SKUs).
Provides a specific temporal snapshot of retail activity for the entire month of March 2018.
Focuses on a well-defined purchasing cohort of 457,298 customers within the larger user base.
Limitations
Data is from a single month (March 2018), limiting analysis of seasonal trends or long-term patterns.
Specific column names and data schema are not provided, complicating immediate analysis.
The dataset is several years old, which may reduce its relevance for modeling current retail dynamics.
Provenance
Source
JD.com, China's largest retailer.
Collection Method
Provided transaction-level data for a research challenge.
Time Range
March 2018.
Freshness
Last updated on the platform in 2025, but the underlying data is from March 2018.
Geography
China.
The full dataset description and specific data schema are hosted externally; users must visit the provided URL for complete details.