Case Study: A Major Manufacturer of Fabrication Machinery
Pre-Campaign Data Creation, Cleaning and Enhancement
- Cleaned, standardized, and filtered web logs: removed requests that were unlikely to come from actual users’ IP range;
- Cleaned, standardized, and filtered e-mail logs:
- Files containing information on client’s e-mail bounce, click, open, non-open, and unsubscribe rates combined into comprehensive file;
- E-mail address domains appended and clicked URLs split into their component parts for easier processing and analysis;
- Scrubbed client’s files further using pinning, CASS, NCOA, and geocoding processes;
- Appended OSHA data and firmographic data on basis of industry code, employees, and sales to gain insights into equipment usage.
- Compiled prospect lists based upon current customer profiles.
Embedded AI-Based Analytics, Real-Time Scoring, and Translation of Findings
- Pinpointed both fulfillment and marketing channels;
- Summarized datasets to identify successful campaigns and orders;
- Identified popular item baskets in dollar amounts and customer counts;
- Isolated individual items and groups of items likely to be purchased after an initial order;
- Calculated dealer performance in terms of close rate, order size, and order count over time across different product categories;
- Created multiple maps to visualize various metrics including:
- New account indices;
- Five-year revenue per capita by state;
- Active customers and revenue by state;
- Quote conversion by state;
- Total revenue by state;
- Constructed quote close, up-sell, and cross-sell models for equipment sales; 39.7 million Calexus-built models tested before most accurate models presented to client and integrated into existing analytic infrastructure.
- Compiled reports based upon direct and wholesale channels; data-generated insights graphically presented as:
- Purchase correlation matrices;
- Quote-and-order summaries;
- Dealer statistics tables;
- End-user multi-purchase tables;
- Developed strategies for increasing wholesale volume.