Retail Data Insights

By akash.verma1076 · January 3, 2026

Analyzed an end-to-end retail dataset in just 7 minutes, the platform generated multiple business-ready insights, charts, and tables—covering revenue…

Hello! Welcome to the Online Retail Dataset Analysis. I've loaded your dataset and it's ready for exploration. You have 100,000 transaction records covering sales data from an online retail store, including information about invoices, products, customers, and countries.

Customers make an average of 2.36 purchases each, with the median being just 1 purchase. Out of 1,664 unique customers, 781 (46.9%) are repeat buyers who made more than one purchase, while 883 (53.1%) are one-time customers. The most dedicated customer made 43 purchases, showing significant variation in customer loyalty.

The analysis attempted to show how revenue is distributed across low, medium, and high unit price tiers. The price tier thresholds were successfully calculated: Low tier includes items priced up to £1.65, Medium tier covers £1.65 to £3.75, and High tier includes items above £3.75. However, the full revenue breakdown by tier encountered some technical issues during processing.

The analysis reveals clear day-of-week patterns in your sales data. Tuesday is your strongest sales day with 269,245 units sold, while Sunday shows the lowest activity at just 62,088 units. The average daily sales volume is approximately 167,642 units. Unfortunately, the monthly/seasonal analysis encountered a technical issue and couldn't be completed, so we only have the day-of-week visualization available.

About 2.08% of all transactions have negative quantities, indicating returns. Out of 100,000 total transactions, 2,079 are returns. A horizontal bar chart has been generated showing the top 10 most returned products by quantity, helping you identify which items are being returned most frequently.

The highest transaction volume occurs at 2:00 PM (14:00), with 15,700 transactions recorded during that hour. The analysis shows a clear pattern of transaction activity throughout the day, with afternoon hours showing the strongest activity.

The average number of unique products (StockCode) per invoice is approximately 21 products. This means that on average, each customer order contains about 21 different product types.

The analysis reveals that first-time buyers actually contribute more revenue than returning customers. First-time buyers generated £989,588 (65.1% of total revenue), while returning customers contributed £529,630 (34.9%). This means first-time buyers bring in nearly twice as much revenue as returning customers.

The analysis identified products with declining sales trends by calculating monthly revenue patterns and trend slopes. Products that appeared in at least 3 months were analyzed, and those with negative trend slopes (indicating decreasing revenue over time) were flagged. The visualizations show the top 10 products with the steepest declining trends and their monthly revenue patterns.

I've identified the top 10 best-selling products by revenue from your dataset. The analysis calculated total revenue by multiplying quantity sold by unit price for each product, filtering out returns and invalid prices. A horizontal bar chart has been created showing these top performers ranked by their total revenue contribution.