Sales Patterns and Customer Behavior of an Online Store

This project focuses on analyzing a dataset from a UK-based online retail store, covering transactions from December 1, 2010, to December 9, 2011. The analysis involved data preprocessing in Excel and exploratory data analysis using PostgreSQL. The aim was to provide insights into sales patterns, customer behavior, and product performance to guide strategic decisions.

The dataset includes transactional information with key columns such as InvoiceNo, StockCode, Description, Quantity, InvoiceDate, UnitPrice, CustomerID, and Country. It details sales transactions, customer data, and product specifics.

Data cleaning was performed to remove irrelevant entries, standardize country names, and mark refunds. New columns were added for TotalPrice, and temporal dimensions such as Invoice Day, Month, Year, and Time to facilitate deeper analysis.

Columns were reorganized for better clarity and usability.

Sales Performance: Total sales reached £8,761,066.65 from 18,402 transactions, with an average order value of £22.10. This indicates a high frequency of smaller purchases.

Top-Selling Products: Products like PAPER CRAFT, LITTLE BIRDIE, and MEDIUM CERAMIC TOP STORAGE JAR were identified as top sellers. These items should be prioritized in inventory and marketing efforts.

Customer Insights: High-value customers were identified based on their purchase frequency and total spending. Implementing loyalty programs for these customers could enhance retention and sales.

Sales Trends: Sales were highest in November 2011 at £1,142,145.77, with notable increases during the holiday season. Quarterly sales were strongest in Q4 2011, reflecting robust holiday shopping activity.

Refund Analysis: The highest refund amount occurred in December 2011 at £174,111.46, likely due to post-holiday returns. Investigating the reasons for these returns could improve product quality and customer satisfaction.

Sales Timing: Sales peaked around noon, suggesting this is an optimal time for promotions. Thursdays recorded the highest sales, whereas Sundays had the lowest. Adjusting marketing strategies to align with these patterns could enhance sales.

Geographic Sales Distribution: The United Kingdom led in total sales with £7,265,862.23, followed by the Netherlands and Ireland. Focusing on these high-revenue regions could drive further growth.

Refunds by Country: The UK also had the highest number of units returned, indicating a need for improved quality control and return management.

Inventory Management: Focus on stocking top-selling products and adjust inventory based on sales data to ensure optimal stock levels.

Marketing Strategies: Align promotions with peak sales times and high-sales days. Enhance marketing efforts in high-revenue regions and consider seasonal trends.

Customer Engagement: Develop loyalty programs for high-value customers and tailor offers based on their purchase history and preferences.

Refund Management: Address the causes of high refund rates, especially in December, and revise return policies to boost customer satisfaction.

Sales Optimization: Adjust staffing and customer service according to sales patterns to ensure efficient operations and improved customer experiences.

This analysis of the online retail dataset provides valuable insights into sales patterns, customer behavior, and product performance. Implementing the recommendations can improve inventory management, marketing strategies, and customer engagement. Ongoing analysis and updates are crucial for maintaining and enhancing business performance.