Probability multiplier process for call center routing

• US 8,781,100 B2
• Filed: 06/24/2009
• Issued: 07/15/2014
• Est. Priority Date: 01/28/2008
• Status: Active Grant

• Current Assignee: Afiniti Ltd
• Original Assignee: Satmap International Holdings, Ltd.
• Inventors: Spottiswoode, S. James P., Chishti, Zia
• Primary Examiner(s): DEANE JR, WILLIAM J
• Application Number:
  US12/490,949
• Publication Number:
  US 20090323921A1
• Time in Patent Office: 1,847 Days
• Field of Search: 37926501-26514, 37926601-2661, 379/309
• US Class Current: 379/265.06
• CPC Class Codes: H04M 2201/18 Comparators
• CPC Class Codes: H04M 2201/36 Memories
• CPC Class Codes: H04M 2203/551 Call history
• CPC Class Codes: H04M 3/5232 Call distribution algorithms
• CPC Class Codes: H04M 3/5233 Operator skill based call distribution
• CPC Class Codes: H04M 3/5235 Dependent on call type or called number [DNIS]

Abstract

Systems and processes are disclosed for routing callers to agents in a contact center based on similar probabilities for an outcome variable. An exemplary probability multiplier process includes determining agent performance of a set of agents for an outcome variable (e.g., sales) and determining caller propensity of a set of callers for the outcome variable (e.g., the propensity or statistical chance of purchasing). Callers and agents are matched based on corresponding agent performance and propensity for the outcome variable of the caller, e.g., matching callers and agents having similar relative performance for the outcome variable, such as matching the highest ranked caller to the highest ranked agent, the worst ranked caller to the worst ranked agent, and so on. The performance and propensity of the callers and agents may be converted to percentile rankings, and callers and agents can be matched based on a closest match of percentile rankings.

Claims

1. A computer-implemented method for routing callers to agents in a call-center routing environment, the method comprising:

identifying, by one or more computers, caller data for each of a set of waiting callers including multiple data items from one or more of the group of demographic data and psychographic data of the caller;

identifying, by the one or more computers, agent data for a set of agents including multiple data items from one or more of the group of demographic data and a psychographic data;

determining, by one or more computers, agent performance of the set of agents for an outcome variable, wherein the agent performance comprises a respective relative ranking for each of two or more of the set of agents for at least the variable outcome of revenue generation;

determining, by the one or more computers, caller propensity of the set of waiting callers for the outcome variable, wherein the caller propensity comprises a respective relative ranking for each of two or more of the set of waiting callers for the outcome variable of revenue generation;

using a multi-data element pattern matching algorithm, by the one or more computers, to create a model that matches each agent of the set of agents to each caller of the set of waiting callers, in a pair-wise fashion to determine a potential for the outcome variable for each agent-caller match; and

matching and routing, by the one or more computers, one of the waiting callers to one of the agents based at least in part on the relative ranking of the one agent in relation to the relative ranking of the one waiting caller and the potential for the outcome variable for this agent-caller match from the multi-data element pattern matching algorithm.

2. The method of claim 1, wherein the ranking for each agent of the two of more agents of the set of agents comprises percentile ranking, and wherein the ranking for each of the two or more waiting caller of the set of waiting callers comprises a percentile ranking, and

wherein the matching and routing step comprises comparing the percentile ranking of the one agent to percentile rankings of the waiting callers in the set of waiting callers.

3. The method of claim 2, further comprising matching an agent to the closest matching percentile ranking caller.

4. The method of claim 2, further comprising matching a caller to the closest matching percentile ranking agent.

5. The method of claim 1, wherein the outcome variable is associated with a sales rate.

6. The method of claim 1, wherein the caller propensity comprises a predicted value for the outcome variable based on at least one caller data.

7. The method of claim 1, wherein the agents are grouped into at least two subsets based on performance and the callers are grouped into at least two subsets based on propensity, and wherein callers from a first subset of callers are matched to agents from a first subset of agents and callers from a second subset of callers are matched to agents from a second subset of agents.

8. The method of claim 1, further comprising comparing demographic data of the agents to demographic data of the callers via the multi-data element pattern matching algorithm.

9. The method of claim 1, wherein the matching and routing step is based at least in part on a closest match of the relative ranking of the one waiting caller for revenue generation to the relative ranking of the one agent for revenue generation.

10. The method of claim 1, further comprising:

accessing, by the one or more computers, result data for the routing; and

modifying, by the one or more computers, the multi-data element pattern matching algorithm based on the result data.

11. A method for routing callers to agents in a call-center routing environment, the method comprising:

identifying, by one or more computers, caller data for each of a set of waiting callers including multiple data items from one or more of the group of demographic data and psychographic data of the caller;

identifying, by the one or more computers, agent data for a set of agents including multiple data items from one or more of the group of demographic data and psychographic data;

determining, by one or more computers, agent performance of the set of agents for a first outcome variable, wherein the agent performance comprises a respective relative ranking for each of two or more of the set of agents for at least the first variable outcome;

determining, by the one or more computers, caller propensity of the set of callers for a second outcome variable, wherein the caller propensity comprises a respective relative ranking for each of two or more of the set of waiting callers for the second outcome variable;

using a multi-data element pattern matching algorithm, by the one or more computers, to create a model that matches each agent of the set of agents to each caller of the set of waiting callers, in a pair-wise fashion to determine a potential for the first outcome variable and the second outcome variable for each agent-caller match; and

matching and routing, by the one or more computers, one of the waiting callers to one of the agents based at least in part on the relative ranking of the one agent in relation to the relative ranking of the one waiting caller and the potential for the outcome variable for this agent-caller match from the multi-data element pattern matching algorithm.

12. The method of claim 11, wherein the ranking for each agent of the two of more agents of the set of agents comprises percentile ranking, and wherein the ranking for each of the two of more waiting callers of the set of waiting callers comprises a percentile ranking, and

wherein the matching and routing step comprises comparing the percentile ranking for the first variable outcome of the one agent to percentile rankings for the second variable outcome of the waiting callers in the set of waiting callers.

13. The method of claim 12, further comprising matching an agent to the closest matching percentile ranking caller.

14. The method of claim 12, further comprising matching a caller to the closest matching percentile ranking agent.

15. The method of claim 11, wherein the first outcome variable and the second outcome variable are associated with a sales rate.

16. The method of claim 11, wherein one or both of the first outcome variable and the second outcome variable is associated with a sales rate.

17. The method of claim 11, wherein the first outcome variable comprises sales data and the second outcome variable comprises propensity to purchase data.

18. The method of claim 11, wherein the caller propensity comprises a predicted value for the second outcome variable based on at least one caller data.

19. The method of claim 11, wherein the agents are grouped into at least two subsets based on performance and the callers are grouped into at least two subsets based on propensity, and wherein callers from a first subset of callers are matched to agents from a first subset of agents and callers from a second subset of callers are matched to agents from a second subset of agents.

20. The method of claim 11, further comprising comparing demographic data of the agents to demographic data of the callers via the multi-data element pattern matching algorithm to determine the chance of either or both of the first and second outcome variables.

21. The method of claim 11, wherein the matching and routing step is based at least in part on a closest match of the relative ranking of the one waiting caller for revenue generation to the relative ranking of the one agent for revenue generation.

22. The method of claim 11, further comprising:

accessing, by the one or more computers, result data for the routing; and

modifying, by the one or more computers, the multi-data element pattern matching algorithm based on the result data.

23. A non-transitory computer-readable storage medium comprising computer-readable instructions for matching, when executed, to agents based on predicted outcome performance, the computer readable instructions comprising:

identifying, by one or more computers, caller data for each of a set of waiting callers including multiple data items from one or more of the group of demographic data and psychographic data of the caller;

identifying, by the one or more computers, agent data for a set of agents including multiple data items from one or more of the group of demographic data and psychographic data;

determining, by one or more computers, agent performance of the set of agents for an outcome variable, wherein the agent performance comprises a respective relative ranking for each of two or more of the set of agents for at least the variable outcome of revenue generation;

determining, by the one or more computers, caller propensity of the set of waiting callers for the outcome variable, wherein the caller propensity comprises a respective relative ranking for each of two or more of the set of waiting callers for the outcome variable of revenue generation;

using a multi-data element pattern matching algorithm, by the one or more computers, to create a model that matches each agent of the set of agents to each caller of the set of waiting callers, in a pair-wise fashion to determine a potential for the outcome variable for each agent-caller match; and

matching and routing, by the one or more computers, one of the waiting callers to one of the agents based at least in part on the relative ranking of the one agent in relation to the relative ranking of the one waiting caller and the potential for the outcome variable for this agent-caller match from the multi-data element pattern matching algorithm.

24. The computer readable storage medium of claim 23, wherein the ranking for each agent of the two of more agents of the set of agents comprises a percentile ranking, and wherein the ranking for each of the two of more waiting callers of the set of waiting callers comprises a percentile ranking, and

wherein the matching and routing step comprises comparing the percentile ranking of the one agent to the percentile rankings of the waiting callers in the set of waiting callers.

25. The computer readable storage medium of claim 24, further comprising instructions for matching an agent to the closest matching percentile ranking caller.

26. The computer readable storage medium of claim 24, further comprising instructions for matching a caller to the closest matching percentile ranking agent.

27. The computer readable storage medium of claim 23, wherein the outcome variable is associated with a sales rate.

28. The computer readable storage medium of claim 23, wherein the caller propensity comprises a predicted value for the outcome variable based on at least one caller data.

29. The computer readable storage medium of claim 23, further comprising instructions for comparing demographic data of the agents to demographic data of the callers via the multi-data element pattern matching algorithm to determine the chance of the outcome variable.

30. The computer readable storage medium of claim 23, wherein the matching and routing step is based at least in part on a closest match of the relative ranking of the one waiting caller for revenue generation to the relative ranking of the one agent for revenue generation.

31. The computer readable storage medium of claim 23, further comprising computer-readable instructions for performing the steps:

accessing, by the one or more computers, result data for the routing; and

modifying, by the one or more computers, the multi-data element pattern matching algorithm based on the result data.

32. A system for routing callers to agents in a call center routing environment based on predicted outcome performance, the system comprising one or more computers configured with computer-readable instructions to perform, when executed, the steps:

identifying, by one or more computers, caller data for each of a set of waiting callers including multiple data items from one or more of the group of demographic data and psychographic data of the caller;

identifying, by the one or more computers, agent data for a set of agents including multiple data items from one or more of the group of demographic data and psychographic data;

determining, by the one or more computers, agent performance of the set of agents for an outcome variable, wherein the agent performance comprises a respective relative ranking for each of two or more of the set of agents for at least the variable outcome of revenue generation;

determine, by the one or more computers, caller propensity of the set of waiting callers for the outcome variable, wherein the caller propensity comprises a respective relative ranking for each of two or more of the set of waiting callers for the outcome variable of revenue generation;

using a multi-data element pattern matching algorithm, by the one or more computers, to create a model that matches each agent of the set of agents to each caller of the set of waiting callers, in a pair-wise fashion to determine a potential for the outcome variable for each agent-caller match; and

matching and routing, by the one or more computers, one of the waiting callers to one of the agents based at least in part on the relative ranking of the one agent in relation to the relative ranking of the one waiting caller and the potential for the outcome variable for this agent-caller match from the multi-data element pattern matching algorithm.

33. The system of claim 32, wherein the ranking for each agent of the two of more agents of the set of agents comprises a percentile ranking, and wherein the ranking for each of the two of more callers of the set of waiting callers comprises a percentile ranking, and

wherein the computer-readable instructions for the matching and routing step compares the percentile ranking of the one agent to percentile rankings of the waiting callers in the set of waiting callers.

34. The system of claim 33, wherein the matching and routing step further comprises computer-readable instructions for matching an agent to the closest matching percentile ranking caller.

35. The system of claim 33, wherein the matching and routing step further comprises computer-readable instructions for matching a caller to the closest matching percentile ranking agent.

36. The system of claim 32, wherein the outcome variable is associated with a sales rate.

37. The system of claim 32, wherein the matching and routing step is based at least in part on a closest match of the relative ranking of the one waiting caller for revenue generation to the relative ranking of the one agent for revenue generation.

38. The system of claim 32, wherein the one or more computers are further configured with computer-readable instructions for performing the steps:

accessing, by the one or more computers, result data for the routing; and

modifying, by the one or more computers, the multi-data element pattern matching algorithm based on the result data.