What are the best datasets for retail analytics?

The best dataset for retail analytics depends on the specific problem you are trying to solve—such as demand forecasting, customer segmentation, market basket analysis, dynamic pricing, or clickstream tracking.

A curated list of the top datasets widely used in industry and data science competitions, organized by analytics use case, includes:

---

1. Customer Segmentation & Cohort Analysis (RFM)

• UCI Online Retail Dataset
  • Description: A transnational dataset containing over 500,000 transactions from a UK-based online gift retailer.
  • Key Features: Invoice number, StockCode, Description, Quantity, Invoice Date, Unit Price, Customer ID, and Country.
  • Best For: Recency, Frequency, Monetary (RFM) analysis, Customer Lifetime Value (CLV), cohort retention analysis, and basic market basket analysis.
  • Where to Find: UCI Machine Learning Repository / Kaggle.

2. End-to-End E-Commerce & Logistics

• Olist Brazilian E-Commerce Public Dataset
  • Description: A real-world commercial dataset covering 100,000 orders placed on Olist (a Brazilian e-commerce platform).
  • Key Features: Multiple relational tables connecting orders, products, sellers, customers, payments, freight costs, delivery timelines, and customer review scores.
  • Best For: Supply chain efficiency, delivery bottleneck analysis, seller performance, cross-category revenue analysis, and customer satisfaction modeling.
  • Where to Find: Kaggle.

3. Market Basket Analysis & Recommendation Systems

• Instacart Market Basket Analysis

  • Description: Over 3 million anonymized grocery orders from more than 200,000 Instacart users.
  • Key Features: User ID, order ID, product ID, sequence of item additions to cart, and re-order indicators.
  • Best For: Association rule mining (Apriori / FP-Growth), repurchase timing prediction, product recommendations, and cart management optimization.
  • Where to Find: Kaggle.

• H&M Personalized Fashion Recommendations Dataset

  • Description: Rich fashion retail dataset featuring transactional history, customer demographic data, and article metadata with accompanying product images.
  • Best For: Collaborative filtering, fashion trend forecasting, dynamic item recommendation, and visual AI application in retail.
  • Where to Find: Kaggle.

4. Demand & Sales Forecasting (Time Series)

• Walmart Store Sales Forecasting

  • Description: Historical sales data across 45 Walmart stores and multiple departments per store, paired with macro-economic indicators.
  • Key Features: Weekly sales, store size, holiday indicators, temperature, fuel price, CPI, and unemployment rate.
  • Best For: Hierarchical time series forecasting, promotional impact modeling, and seasonal demand planning.
  • Where to Find: Kaggle.

• Rossmann Store Sales Dataset

  • Description: Daily sales data for over 1,100 Rossmann drugstores in Germany.
  • Key Features: Store type, assortment level, distance to nearest competitor, promotions, open state, and school/state holidays.
  • Best For: Supervised machine learning (XGBoost/LightGBM/LSTM) for store-level sales forecasting and promo-effect estimation.
  • Where to Find: Kaggle.

• Corporación Favorita Grocery Sales Forecasting

  • Description: Highly granular daily unit-sales data from Ecuador’s top grocery chain.
  • Key Features: Item-level store sales, oil prices (local economic factor), store location attributes, and promotional statuses.
  • Best For: Extreme fine-grained time series forecasting (SKU-level) and managing zero-inflated demand distributions.
  • Where to Find: Kaggle.

5. Clickstream, Conversion & User Behavior

• eCommerce Events History in Online Store (Cosmetics / Electronics)

  • Description: Event-level log data collected from multi-brand e-commerce shops.
  • Key Features: User ID, session ID, event time, event type (view, cart, remove_from_cart, purchase), price, category hierarchy, and brand.
  • Best For: Conversion funnel analysis, session duration tracking, cart abandonment analysis, and churn risk scoring.
  • Where to Find: Kaggle.

• Taobao User Behavior Dataset

  • Description: Millions of implicit feedback logs from Alibaba’s retail platform Taobao.
  • Key Features: User actions (click, buy, add-to-cart, favorite) across item categories over time.
  • Best For: High-scale clickstream modeling, real-time recommendation engine design, and intent analysis.
  • Where to Find: Kaggle / Alibaba Open Data.

6. Promotions, Customer Loyalty & CPG

• Dunnhumby "The Complete Journey"
  • Description: Household-level purchase data covering 2,500 households over two years in CPG/grocery stores.
  • Key Features: Store visits, coupon redemptions, campaign exposures, product sub-categories, and household demographics.
  • Best For: Promotion effectiveness, coupon sensitivity modeling, price elasticity analysis, and shopper loyalty analytics.
  • Where to Find: Dunnhumby Source / Kaggle.

7. Product Catalog, Reviews & Sentiment Analysis (NLP)

• Amazon Product Data & Reviews (McAuley UCSD)
  • Description: Millions of Amazon product reviews, ratings, category trees, cross-buying links, and metadata across dozens of store categories.
  • Best For: Sentiment analysis, NLP product catalog enrichment, cross-category recommendations, and price optimization.
  • Where to Find: UCSD Data Repository.

---

Quick Reference Matrix

Goal / Analytical Task	Recommended Dataset
Sales & Inventory Forecasting	Rossmann Store Sales / Walmart Sales Forecasting
Customer Segmentation (RFM)	UCI Online Retail Dataset
Cross-Selling / Association Rules	Instacart Market Basket Analysis
Full Business Operations & Delivery	Olist Brazilian E-Commerce Dataset
Campaign & Promotion Sensitivity	Dunnhumby "The Complete Journey"
Funnel & Cart Abandonment	eCommerce Events History Dataset
Visual / Fashion Recommendations	H&M Recommendation Dataset

---

Public Repositories to Explore Further

• Kaggle Datasets Search: Search tags like retail, time-series, market-basket, or e-commerce.
• UCI Machine Learning Repository: Great for benchmark tabular datasets.
• Google Dataset Search: Excellent for finding academic or government-published retail industry figures.
• Maven Analytics Data Playground: Offers cleaned, business-ready sample datasets (e.g., Toy Store Sales, Coffee Shop Sales) ideal for Power BI / Tableau dashboard practice.