Consumer Insights Unit Applied Customer Analytics – Principles and Practice Company Overview Founded in 1966; based in Minneapolis, Minnesota $50+ billion revenue, Fortune 100 company World’s #1 Consumer Electronics retailer 4000 stores* in Americas, Europe and China 180,000 Employees * 2430 Carphone Warehouse stores Locations STORE COUNTS BEST BUY U.S. 1092 Mexico 5 UK 2 Canada 66 GEEK SQUAD Stand-Alone Stores 5 BEST BUY MOBILE Stand-Alone Stores U.S. Stand-Alone Stores Canada 98 4 FAMILY OF BRANDS The Carphone Warehouse 2430 Five Star 158 Future Shop 144 Magnolia Audio Video 6 Pacific Sales 35 AudioVisions 3 The CIU Purpose and Mission We define consumer insights as finished intelligence that drives the business. What is the value of Consumer Insight? Consumer Insight is information about consumer behaviors, attitudes and values which is derived from research, employees, and data analysis and is directly relevant for the growth of a business CIU Vision Statement Providing a clear, relevant, actionable, measurable, and trusted view of the consumer to the enterprise CIU Mission Statement To ensure locally relevant consumer insights are generated, shared, understood, and used in driving business actions in service of the consumer and growth of the global enterprise What CIU Analytics does What CIU Analytics does Happy Customers CIU Analytics CIU Analytics as a capability includes… • • • • • • • • Customer segmentation Purchase path and purchase basket analysis Descriptive and differential statistics Predictive modeling Constrained resource optimization Test and measurement Market share measurement Econometrics What it takes to provide quality analysis Translation/ Communication Analytic Talent Analytic Tools Data Accessibility Data Architecture Hardware We cannot do analytics on data we don’t have. Stop Guessing: Analytics Can Tell You Who Your Customers Are • Start with 30-40 modeled variables – “Feature Vectors” • Each feature vector is like a gene, which describes a facet, or set of customer behavior traits • 8-15 Feature Vectors are used to define Action Clusters, all 40 can be used to create sub-groups 5-dimensional typical segmentation ≈ 16,000 views 12-dimensional Action Clusters ≈ 3,138,000,000,000 views 30 Feature Vectors ≈ 17,400,000,000,000,000,000,000,000,000,000 views Most segmentation approaches only focus here Preferred Channel Preferred Product Categories Time until Repurchase in Key Categories Econometric: Real-estate & Unemployment Annual Spend Level Annual Transactions Age + Income + Geography Length of Time as Customer Recency + Frequency + Value Response to Media Breadth of Categories Shopped Return / Exchange Behavior Use of Service Programs Use of In-House Credit Card Participation in Loyalty Program Case Study on Decision Optimization: Foresight Customer Strategies Articulates and Optimally Funds Top Initiatives Media Types Computes Optimal Media Effectiveness Optimize Across Identified Strategies Cluster 1 Cluster 2 Cluster 3 Cluster 4 Cluster 5 Optimally Funds Across Action Clusters Optimize Across Customer Action Clusters Cluster 1 Cluster 2 Cluster 3 Cluster 4 Cluster 5 Customer Cluster 1 Cluster 2 Cluster 3 Cluster 4 Cluster 5 Foresight optimizes resource investment against the Action Cluster portfolio to maximize business objectives. It includes dimensions of customer strategies (or initiatives), media types and geography, and is constrained by budgets and policies. Optimize Across Available Media Allocates Against Policy and Budget Constraints Bounded By Budget Constraints Cluster 1 Cluster 2 Cluster 3 Cluster 4 Cluster 5 Annual Budget Geography Allocates to Stores / Regions / DMAs 80,000+ Total Decisions Optimize Across Geographies