Executive Summary
Online Retail Analytics — Dec 2010 to Nov 2011
Dominasi Pasar UK
84.6% revenue dari UK, namun Netherlands dan Australia punya AOV tertinggi (£121 dan £117).
Pelanggan Loyal Menopang Bisnis
337 loyal customers (>10 order) menyumbang £4.39M (47% total revenue).
Segmen Champions Paling Berharga
962 Champions dengan avg monetary £6,039 dan recency hanya 13 hari.
Tingkat Pembatalan Kritis
19.22% cancellation rate dengan kerugian £896,812 didominasi UK.
Lonjakan Q4 yang Konsisten
Revenue naik dari £760K (Jun) ke £1.51M (Nov), pola seasonal yang dapat diantisipasi.
Proyeksi Pertumbuhan Positif
Forecast quantity tumbuh dari 666K ke 827K unit (MAPE 11.5%), transaksi stabil di 2.2K-2.8K per bulan.
Model Performance (Prophet)
Cluster Summary
Dashboard Overview
Key performance indicators and high-level metrics for Dec 2010 – Nov 2011.
Monthly Revenue Trend
Unit Price Analysis
Distribution of transaction unit prices across predefined price segments.
Price Segment Distribution
Price Segment Share
Country Analysis
Geographic breakdown of revenue, transactions, and average order value.
Top 15 Countries by Revenue
Top 10 Countries by Transactions
UK vs Non-UK Revenue Share
Avg Order Value — Top 10 Countries
Customer Behavior
Customer classification based on purchasing frequency.
Customer Count by Type
Revenue by Customer Type
Monthly Trends
Time series performance across revenue, quantity, and transaction metrics.
Monthly Revenue Trend
Monthly Quantity Trend
Monthly Transactions Trend
Top Products
The ten best-selling products by total quantity sold.
Top 10 Products by Quantity Sold
RFM Segmentation
Customers segmented based on Recency, Frequency, and Monetary value.
Customers per Segment
Avg Monetary Value per Segment
Avg RFM Profile per Segment
Customer Clustering
Agglomerative clustering defining distinct customer personas.
Cluster Size Distribution
Cluster RFM Profile (Grouped)
High-Value
Highest revenue generators with frequent purchases and recent activity.
Loyal Active
Core customer base with regular purchases and moderate spend.
Occasional
Largest group, infrequent purchases with lower overall spend.
Dormant
High recency (inactive), low frequency and minimal monetary value.
Anomaly Detection
Analysis of cancelled orders and estimated revenue loss.
Monthly Cancellation Volume
Top Countries by Cancellations
Prophet Forecasting
6-Month forward projection using Facebook Prophet with 80% confidence intervals.
Revenue Forecast
Quantity Forecast
Transactions Forecast
Insights & Recommendations
Strategic takeaways translated into actionable business recommendations.
Market Basket Analysis
Association rules mining — top item combinations frequently purchased together.
Support vs Confidence (Bubble Size = Lift)
Cohort Analysis
Customer retention rates tracked by monthly acquisition cohort.
Retention Rate Heatmap (%) — Cohort × Month Index
Avg Retention Rate by Month Index
Churn Risk Scoring
Customer segmentation by churn probability based on RFM features.
Customer Count per Risk Segment
Revenue per Risk Segment
Revenue Pareto Analysis
Cumulative revenue distribution across the customer base.
Pareto Curve — Cumulative % Customers vs Cumulative % Revenue
Top 20 Customers by Revenue
Product Trend Heatmap
Monthly quantity and relative performance heatmap per product category.
Absolute Quantity per Category per Month
Relative Performance (% of Peak) per Category
Sales Funnel
Customer progression from total transactions to loyal buyers.
Customer Conversion Funnel
Funnel Stage Comparison
CLV Analysis
Customer Lifetime Value segmentation and model performance metrics.