Caller-Agent Pattern Matching Routing System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional contact center routing systems typically rely on random or round-robin methods to connect callers with available agents, failing to consider the relative capabilities and individual characteristics of both callers and agents, which can lead to suboptimal interactions and increased wait times.
Innovation Solution
Implementing a pattern matching algorithm and computer model that compares caller data with agent data to predict the highest probability of a desired outcome, allowing for non-linear routing based on suitability scores and hold thresholds, while also considering demographic, psychographic, and business-relevant data to optimize interactions such as sales generation and customer satisfaction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If round-robin routing is used to connect callers to agents, then ease of operation is improved, but interaction quality and productivity deteriorate
Solution Approach 1:
The system changes the routing parameter from simple round-robin selection to pattern-based matching that considers multiple characteristics including caller needs, agent capabilities, historical performance data, and real-time availability. This transforms the routing decision from a mechanical sequence to an optimized match based on multiple variables.
Solution Approach 2:
The patent replaces the mechanical round-robin routing system with an intelligent pattern matching system that uses data analysis and algorithms to determine optimal caller-agent connections. This substitution enables the system to evaluate multiple factors simultaneously and make decisions based on predicted interaction outcomes rather than simple sequential assignment.
2Productivity
If pattern matching algorithms are implemented to optimize caller-agent matching, then interaction quality and revenue generation improve, but device complexity increases
Solution Approach 1:
The routing system is segmented into distinct functional modules: data collection components that gather caller and agent information, pattern matching engines that analyze the data against predefined criteria, scoring systems that evaluate match quality, and routing execution components that implement the decisions. This modular architecture manages complexity by separating concerns and enabling independent optimization of each component.
Solution Approach 2:
The system introduces an intermediary routing layer between the incoming caller and the agent pool. This intermediary contains the pattern matching algorithms and decision logic, acting as a mediator that processes routing decisions based on multiple factors including caller needs, agent capabilities, and historical performance data, thereby managing complexity centrally rather than distributed across all system components.
3Loss of time
If callers are routed based on hold time in queue, then loss of time for individual callers is reduced, but overall productivity and interaction quality deteriorate
Solution Approach 1:
The routing system dynamically adjusts routing decisions based on real-time conditions rather than static queue position. It continuously evaluates current agent availability, caller characteristics, and predicted interaction outcomes to make adaptive routing decisions. This dynamic approach allows the system to sometimes route shorter-waiting callers first when pattern matching indicates a better overall outcome, rather than strictly following hold time.
Data Source
AI summary
Methods and systems are provided for routing callers to agents in a call-center routing environment. An exemplary method includes identifying caller data for at least one of a set of callers on hold and causing a caller of the set of callers to be routed to an agent based on a comparison of the caller data and the agent data. The caller data and agent data may be compared via a pattern matching algorithm and/or computer model for predicting a caller-agent pair having the highest probability of a desired outcome. As such, callers may be pooled and routed to agents based on comparisons of available caller and agent data, rather than a conventional queue order fashion. If a caller is held beyond a hold threshold the caller may be routed to the next available agent. The hold threshold may include a predetermined time, “cost” function, number of times the caller may be skipped by other callers, and so on.