Lead Data Analyst: Mostafizur Rahman
Data Source: Olist Public E-Commerce Dataset
This report presents the key findings from our analysis of the Olist e-commerce business data.
To conduct this analysis, we collected and organized data from multiple sources into a structured database, allowing us to accurately evaluate sales performance, customer behavior, product performance, and operational efficiency. During this process, data quality issues were identified and addressed to ensure reliable insights.
Our analysis highlights several important business findings:
- Revenue is highly concentrated among a relatively small number of products and sellers, following the common 80/20 (Pareto) pattern.
- The company maintains a healthy buffer between estimated and actual delivery times, helping ensure orders are delivered on time.
- However, delivery estimates may be more conservative than necessary, which can increase shipping expectations shown to customers during checkout.
- Longer or less competitive delivery estimates may contribute to customers abandoning their carts before completing purchases.
- The findings identify opportunities to improve customer experience, optimize delivery promises, and further increase sales performance.
Overall, this report converts raw transaction data into actionable business insights that can support better operational decisions, improve customer satisfaction, and drive future growth.
To ensure accurate financial reporting, cancelled and unavailable orders were excluded from the analysis. The platform's core performance metrics are:
- Gross Revenue: $13.49 Million
- Completed Orders: 98,199
- Average Order Value (AOV): $137.42
Although the platform offers products across 74 categories, revenue is heavily concentrated in a small number of them.
Our analysis shows that:
- 18 categories (24% of all categories) generate 81.26% of total revenue.
- These categories contribute approximately $10.96 Million in sales.
- The remaining 56 categories generate only 18.74% of revenue despite representing the majority of the catalog.
This indicates that the business relies heavily on a relatively small group of high-performing categories.
| Rank | Category | Gross Revenue ($) | Cumulative Share (%) |
|---|---|---|---|
| 1 | beleza_saude (Health & Beauty) | 1,255,695.13 | 9.31% |
| 2 | relogios_presentes (Watches & Gifts) | 1,198,185.21 | 18.18% |
| 3 | cama_mesa_banho (Bed, Bath & Table) | 1,035,964.06 | 25.86% |
| 4 | esporte_lazer (Sports & Leisure) | 979,740.92 | 33.12% |
| ... | ... | ... | ... |
| 17 | pcs (Computers / Systems) | 222,963.13 | 79.68% |
| 18 | pet_shop (Pet Supplies) [Pareto Cutoff] | 213,766.63 | 81.26% |
| - | Remaining 56 Categories | 2,530,111.90 | 18.74% |
The data clearly shows that a small number of categories are driving the majority of the company's revenue. These categories should be treated as strategic priorities because their performance has a significant impact on overall business results.
Place products from the top-performing categories closer to packing and shipping areas within fulfillment centers.
Expected Benefits:
- Faster order processing
- Reduced picker travel time
- Improved warehouse efficiency
- Faster shipment preparation
Focus marketing and promotional budgets on the highest-performing categories instead of distributing spending evenly across all categories.
Expected Benefits:
- Higher return on marketing investment (ROI)
- Increased customer acquisition in proven categories
- Stronger growth in the platform's primary revenue drivers
Just 24% of product categories generate more than 81% of total revenue. Protecting and growing these key categories should be a top business priority, while lower-performing categories should be reviewed for optimization, consolidation, or selective investment.
To evaluate delivery efficiency and customer experience, we analyzed 96,470 successfully delivered orders and compared actual delivery times with the delivery estimates shown to customers at checkout.
- Average Actual Delivery Time: 12.6 days
- Average Promised Delivery Time: 23.7 days
- Average Early Delivery: 11.2 days ahead of schedule
The platform consistently delivers orders much faster than promised. On average, customers receive their orders 11.2 days earlier than the delivery estimate displayed during checkout.
While early delivery creates a positive customer experience, displaying a delivery estimate of nearly 24 days may discourage some customers from completing their purchases. Long delivery promises can make the platform appear slower than competitors, potentially increasing cart abandonment before an order is placed.
The analysis also shows a strong relationship between delivery speed and customer satisfaction.
| Review Score | Orders | Avg. Delivery Time (Days) | Avg. Days Early | Customer Sentiment |
|---|---|---|---|---|
| ⭐⭐⭐⭐⭐ | 56,810 | 10.7 | 12.7 | Excellent |
| ⭐⭐⭐⭐ | 18,943 | 12.3 | 11.7 | Very Good |
| ⭐⭐⭐ | 7,942 | 14.3 | 10.1 | Neutral |
| ⭐⭐ | 2,938 | 16.7 | 7.9 | Dissatisfied |
| ⭐ | 9,380 | 21.3 | 3.4 | Highly Dissatisfied |
Customer satisfaction remains strong when deliveries are completed within approximately 14 days. However, ratings decline significantly as delivery times increase beyond this point.
Notably:
- Average delivery time for 5-star reviews is only 10.7 days.
- Average delivery time for 1-star reviews rises to 21.3 days.
- The decline in satisfaction accelerates after the 14-day mark, indicating that delivery speed is a major driver of customer perception.
Although most orders arrive earlier than promised, customers primarily judge their experience based on how long they actually wait, not how much earlier the order arrives compared to the estimate.
This means that deliveries taking more than two weeks present a higher risk of negative reviews, lower customer satisfaction, and reduced repeat purchases.
Consider reducing checkout delivery estimates from the current average of 23.7 days to a more competitive range based on actual delivery performance and regional delivery history.
Expected Benefits:
- Improved conversion rates
- Reduced cart abandonment
- Stronger competitive positioning
- More accurate customer expectations
Create automated alerts for orders approaching 14 days in transit so customer service teams can proactively engage affected customers.
Expected Benefits:
- Reduced negative reviews
- Better customer communication
- Higher customer trust
- Improved retention and repeat purchases
The logistics network is performing well, delivering orders in an average of 12.6 days. However, customer satisfaction declines sharply when delivery times exceed 14 days. Improving delivery promises and proactively managing delayed orders represent significant opportunities to increase customer satisfaction, protect brand reputation, and improve conversion rates.
Ensuring data accuracy was a critical part of this analysis. During the project, several data quality issues were identified and addressed to improve the reliability of reporting and business insights.
We identified 8 orders that were marked as successfully delivered but did not contain a customer delivery confirmation date.
Further investigation showed that most of these records occurred during the same period (June–July 2018), suggesting a system synchronization or data integration issue rather than manual data entry errors.
Action Taken:
- These records were excluded from delivery-time calculations to prevent inaccurate performance metrics.
- The issue was documented for future system monitoring and investigation.
We identified 610 products that did not have an assigned product category.
Without correction, sales generated by these products would not appear in category-level performance reports, leading to incomplete business analysis.
Action Taken:
- A data quality rule was implemented to automatically classify uncategorized products as "Unassigned".
- This ensured that all product sales remained visible in reporting and analysis.
Business Impact:
- Recovered visibility for $178,572.55 in revenue that would otherwise have been excluded from category-based reporting.
Reliable business decisions depend on reliable data. Addressing these issues improved the accuracy of revenue analysis, category performance reporting, and delivery performance metrics.
The overall data quality was strong, with only a small number of records requiring correction. By resolving these issues, we improved reporting accuracy and ensured that approximately $178,000+ in sales remained visible in business performance analyses. This strengthens confidence in the insights and recommendations presented throughout this report.
Phase 1 focused on establishing key business performance metrics and operational insights. The following areas are recommended for Phase 2 analysis:
Data Source: olist_order_payments
Analyze how payment methods and installment plans affect customer spending and Average Order Value (AOV).
Opportunity: Evaluate whether Buy Now, Pay Later (BNPL) or expanded installment options could increase sales and order values.
Data Source: olist_sellers
Investigate seller-level fulfillment performance to identify the root causes of delivery delays.
Opportunity: Determine whether delays originate from seller operations or logistics infrastructure.
Data Sources: olist_customers, olist_geolocation
Analyze customer demand across regions and identify high-growth markets.
Opportunity: Support decisions on future fulfillment center locations to reduce delivery times and shipping costs.
Phase 2 should focus on customer spending behavior, seller performance, and geographic expansion opportunities to drive revenue growth, improve operational efficiency, and strengthen customer experience.
© 2026 Mostafizur Rahman