Call Routing Using Purchase-Likelihood Scoring and Priority
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Call centers typically provide uniform call treatment, failing to account for individual caller situations, particularly in large companies with diverse product offerings, making it difficult to identify and prioritize callers interested in new products or services.
Innovation Solution
A predictive model determines a likelihood that a caller will accept an offer to discuss new products or services, routing calls to either Enhanced Deepening (ED) or Expedited Servicing (ES) representatives based on a generated score, using customer data such as recent purchases, life events, and digital behavior.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a single uniform call treatment is provided to all callers, then call center operations are simple and easy to manage, but individual caller situations are not accounted for and sales opportunities are lost
Solution Approach 1:
The patent segments callers into different categories (ED and ES) based on their likelihood to purchase, using a predictive model that analyzes customer data. This segmentation enables differentiated call treatment while maintaining manageable system complexity through automated classification.
Solution Approach 2:
The system changes the parameter of call treatment based on the predicted purchase likelihood score. Callers are routed to different representative types (ED for enhanced deepening, ES for expedited servicing) based on dynamic parameter assessment, allowing customization without proportionally increasing system complexity.
2Productivity
If callers are routed to specialized representatives based on purchase likelihood, then sales opportunities are enhanced, but call routing system complexity increases
Solution Approach 1:
The predictive model performs preliminary analysis of customer data before the call is connected, determining the caller's purchase likelihood in advance. This preliminary action enables efficient routing to appropriate representatives without adding complexity during the actual call handling process.
Solution Approach 2:
The predictive modeling system acts as an intermediary between the incoming call and the representative assignment. It processes customer data and generates routing recommendations, separating the complexity of analysis from the simplicity of execution, thereby enhancing productivity without proportionally increasing overall system complexity.
3Productivity
If all calls are handled with standard customer service treatment, then resource allocation is simple, but sales opportunities for new products and services are missed
Solution Approach 1:
The system performs preliminary analysis of customer data and generates a purchase likelihood score before the call is answered. This advance preparation eliminates time loss during the call itself, allowing representatives to immediately focus on sales opportunities without delay for analysis.
Solution Approach 2:
The predictive model automatically analyzes customer data and generates routing recommendations without requiring manual assessment during the call. This self-service approach to call classification minimizes time loss while improving sales conversion by identifying opportunities beforehand.
Data Source
AI summary
A system can route a call based on a minimum expected delay, classification, and priority of the call. In some embodiments, when a call is received, the call is classified and prioritized within the classification. A minimum expected delay time is determined. Based on the minimum expected delay time, the classification, and the prioritization, the call is routed to a representative with instructions on how the call should be treated.