Predictive Dialing Mechanism for Agent Productivity Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional predictive dialers rely on historical call data to determine the number of calls to make, leading to inefficiencies such as nuisance calls and reduced agent productivity, as they do not account for the probability of contact for each customer at specific times and the number of available agents.
Innovation Solution
A predictive dialing mechanism that determines the probability of contact for each customer over a predetermined period and adjusts the number of calls based on the number of available agents, optimizing the selection of customers to call by considering their individual probabilities of contact.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If conventional predictive dialers use historical call data to determine the number of calls to make, then they can automate calling operations, but they generate nuisance calls and reduce agent productivity
Solution Approach 1:
The patent changes the parameter for determining call volume from historical aggregate data to real-time probability-based predictions. Each customer is assigned a probability of contact, and the dialer adjusts call volume dynamically based on these probabilities and current agent availability, transforming the automation from rigid historical patterns to adaptive real-time decision-making
Solution Approach 2:
The system incorporates continuous feedback loops where call outcomes, customer responses, and agent availability are monitored in real-time. This feedback adjusts the probability calculations and call scheduling dynamically, allowing the automated dialer to learn from actual performance and optimize future calling decisions to maximize agent productivity
2Ease of operation
If conventional predictive dialers dial based on historical call history, then they can maintain simple operations, but they fail to account for individual customer contact probabilities and agent availability
Solution Approach 1:
The patent segments the customer base into individual records with unique probability assessments. Instead of treating all customers uniformly based on aggregate history, each customer is individually evaluated for contact probability, allowing precise targeting while maintaining automated operation through systematic processing of individual records
Solution Approach 2:
The system transitions from static historical averages to dynamic real-time probability calculations. The dialer continuously updates contact probabilities based on current conditions including time of day, customer behavior patterns, and agent availability, making the operation adaptable rather than rigid while maintaining automation
3Quantity of substance
If the dialer makes more calls to increase coverage, then more customers can be contacted, but the number of nuisance calls increases
Solution Approach 1:
The patent changes the parameter for call volume determination from fixed historical ratios to variable probability-based calculations. The dialer adjusts the number of calls made based on the summed probabilities of individual customers and current agent capacity, optimizing the balance between coverage and nuisance call reduction
Solution Approach 2:
The system applies partial action by making calls only to customers with sufficient probability thresholds rather than attempting to contact everyone. This selective approach reduces nuisance calls while maintaining adequate coverage by focusing resources on customers most likely to answer
Data Source
AI summary
A method, apparatus and computer program product for performing predictive dialing is presented. A probability of contact for each customer of a plurality of customers for a predetermined period of time is determined. A number of agents available during the predetermined period of time are also determined. A selection is then made regarding which customers to call based on the number of agents available during the time period and the probability of contact for each of the plurality of customers for a predetermined period of time. The number of calls are then placed to the selected customers.