Call mapping systems and methods using variance algorithm (VA) and/or distribution compensation

• US 8,879,715 B2
• Filed: 03/15/2013
• Issued: 11/04/2014
• Est. Priority Date: 03/26/2012
• Status: Active Grant

• Current Assignee: Afiniti Ltd
• Original Assignee: Satmap International Holdings, Ltd.
• Inventors: Spottiswoode, S. James P., Chishti, Zia
• Primary Examiner(s): NGUYEN, QUYNH H
• Application Number:
  US13/843,724
• Publication Number:
  US 20140119533A1
• Time in Patent Office: 599 Days
• Field of Search: 379/265.06, 379/265.01, 379/265.05, 379/265.1, 379/265.11, 379/265.12, 379/266.08
• US Class Current: 379/265.06
• CPC Class Codes: G06F 16/23 Updating
• CPC Class Codes: G06Q 10/06398 Performance of employee with respect to a job function
• CPC Class Codes: H04M 2203/2061 Language aspects
• CPC Class Codes: H04M 2203/401 Performance feedback
• CPC Class Codes: H04M 2203/556 Statistical analysis and interpretation
• CPC Class Codes: H04M 3/5175 Call or contact centers supervision arrangements
• CPC Class Codes: H04M 3/5191 interacting with the Internet
• CPC Class Codes: H04M 3/523 with call distribution or queueing
• CPC Class Codes: H04M 3/5232 Call distribution algorithms
• CPC Class Codes: H04M 3/5233 Operator skill based call distribution
• CPC Class Codes: H04M 3/5234 Uniform load distribution
• CPC Class Codes: H04M 3/5235 Dependent on call type or called number [DNIS]
• CPC Class Codes: H04M 3/5238 with waiting time or load prediction arrangements
• CPC Class Codes: Y04S 10/50 Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Abstract

Method, system and program product, comprising obtaining agent performance data; ranking, agents based the agent performance data; dividing agents into agent performance ranges; partitioning callers based on criteria into a set of partitions; determining for each partition an outcome value for a first agent performance range and a second agent performance range; calculating for the partitions a respective outcome value difference indicator based on the outcome value for the first agent performance range and the outcome value for the second agent performance range for the partition; matching a respective agent to a respective caller in one of the partitions, based on the outcome value difference indicators for the partitions.

Claims

1. A method, comprising:

obtaining, by one or more computers, agent performance data for a set of agents;

ranking, by the one or more computers, the agents based at least in part on the agent performance data;

dividing, by the one or more computers, the agents into agent performance ranges based at least in part on the ranking step;

partitioning, by the one or more computers, callers in a set of callers based on one or more criteria into a set of partitions;

determining for each of the partitions, by the one or more computers, an outcome value comprising a statistical chance or likelihood for a particular outcome at least for a first one of the agent performance ranges and for a second one of the agent performance ranges;

calculating, by the one or more computers, for each of the partitions an outcome value difference indicator based at least in part on a difference between the outcome value for the first agent performance range and the outcome value for the second agent performance range; and

matching, by the one or more computers, a respective agent with respective performance data to a respective caller in one of the partitions, based at least in part on the outcome value difference indicators for the partitions.

2. The method as defined in claim 1, wherein the determining an outcome value step comprises determining a mean outcome value for a particular outcome for the agent performance range.

3. The method as defined in claim 1, wherein the partition is based at least in part on one or more selected from the group of demographic data, area code, zip code, NPANXX, VTN, geographic area, 800 number, and transfer number.

4. The method as defined in claim 1, wherein the outcome value comprises one or more selected from the group of sale, number of items sold per call, and revenue per call coupled with handle time.

5. The method as defined in claim 1, wherein the matching step is based at least in part on a rule to assign a higher performing agent to a caller from one of the partitions when the outcome value difference indicator for the partition of the caller is higher relative to the outcome value difference indicators of other of the partitions, and to assign a lower performing agent to a caller from one of the partitions when the outcome value difference indicator for the partition of the caller is lower relative to the outcome value difference indicators of other of the partitions, wherein the higher performing agent and the lower performing agent are determined based at least in part on the agent performance data for the respective agent relative to the agent performance data for other of the agents.

6. The method as defined in claim 1, wherein the matching step further comprises the steps:

calculating, by the one or more computers, a ranking or percentile for the partitions by outcome value difference indicator from a high outcome value difference indicator to a low outcome value difference indicator based at least in part on a first number of calls;

calculating a ranking or percentile of the agent performances, by the one or more computers, for the respective agents in the set of agents based at least in part on the respective agent performance data and a second number of calls; and

matching, by the one or more computers, the respective agent to the respective caller based at least in part on a rule to minimize a difference between the partition percentile of the outcome value difference indicator for the partition of the respective caller and the performance percentile of the respective agent.

7. The method as defined in claim 1, wherein the matching step is performed based at least in part using a multi-data element pattern matching algorithm in a pair-wise fashion to obtain a score or other indicator for each of multiple caller-agent pairs from a set of agents and a set of callers.

8. The method as defined in claim 1, further comprising correcting, by the one or more computers, for time effects by subtracting from agent performance data an epoch average from a desired outcome data sample, wherein the epoch average comprises an average of one selected from the group of hour-of-day desired outcome data, day-of-week desired outcome data, and hour-of-week desired outcome data from an epoch in which a respective call occurs, to obtain a new target outcome variable.

9. The method as defined in claim 1, further comprising correcting, by the one or more computers, for time effects by dividing agent performance data by an epoch average from a desired outcome data sample, wherein the epoch average comprises an average of one selected from the group of hour-of-day desired outcome data, day-of-week desired outcome data, and hour-of-week desired outcome data from an epoch in which a respective call occurs, to obtain a new target outcome variable.

10. The method as defined in claim 1, further comprising correcting, by the one or more computers, for time effects by forming an outer product by combining hour-of-day desired outcome data, day-of-week desired outcome data, and hour-of-week desired outcome data from an epoch in which a respective call occurs to obtain a time effects factor;

and

calculating, by the one or more computers, the agent performance using the time effects factor in a Bayesian Mean Regression calculation of the agent performance.

11. The method as defined in claim 1, further comprising:

switching, by the one or more computers, between the outcome value difference indicator matching algorithm and a second matching algorithm that is different from the outcome value difference indicator matching algorithm, based on one or more criteria; and

when there has been switching to the second matching algorithm, then performing, by the one or more computers, performing the second matching algorithm to match a respective one of the agents with a respective one of the callers.

12. The method as defined in claim 11, further comprising:

obtaining results data, by the one or more computers, using each of the outcome value difference indicator matching algorithm and the second matching algorithm in the matching step;

determining a switch-over point in a distribution for agent performance and/or a distribution for caller propensity for the particular outcome where better results are obtained from one of the algorithms relative to the other of the algorithms; and

using, by the one or more computers, this switch-over point to switch between the algorithms.

13. The method as defined in claim 1, further comprising:

switching, by the one or more computers, between the outcome value difference indicator matching algorithm and a second matching algorithm configured to match a respective agent with a respective ranking or percentile to a caller in one of the partitions with a closest ranking or percentile, based on one or more criteria; and

when there has been switching to the second matching algorithm, then performing, by the one or more computers, the second matching algorithm to match a respective one of the agents with a respective one of the callers.

28. The method as defined in claim 6, wherein the partition is based at least in part on one or more selected from the group of demographic data, area code, zip code, NPANXX, VTN, geographic area, 800 number, and transfer number.

14. A system, comprising:

one or more computers, configured with program code, that when executed, perform the steps;

obtaining, by the one or more computers, agent performance data for a set of agents;

ranking, by the one or more computers, the agents based at least in part on the agent performance data;

dividing, by the one or more computers, the agents into agent performance ranges based at least in part on the ranking step;

partitioning, by the one or more computers, callers in a set of callers based on one or more criteria into a set of partitions;

determining for each of the partitions, by the one or more computers, an outcome value comprising a statistical chance or likelihood for a particular outcome at least for a first one of the agent performance ranges and for a second one of the agent performance ranges;

calculating, by the one or more computers, for each of the partitions an outcome value difference indicator based at least in part on a difference between the outcome value for the first agent performance range and the outcome value for the second agent performance range; and

matching, by the one or more computers, a respective agent with respective performance data to a respective caller in one of the partitions, based at least in part on the outcome value difference indicators for the partitions.

15. The system as defined in claim 14, wherein the determining an outcome value step comprises determining a mean outcome value for a particular outcome for the agent performance range.

16. The system as defined in claim 14, wherein the partition is based at least in part on one or more selected from the group of demographic data, area code, zip code, NPANXX, VTN, geographic area, 800 number, and transfer number.

17. The system as defined in claim 14, wherein the outcome value comprises one or more selected from the group of sale, number of items sold per call, and revenue per call coupled with handle time.

18. The system as defined in claim 14, wherein the matching step is based at least in part on a rule to assign a higher performing agent to a caller from one of the partitions when the outcome value difference indicator for the partition of the caller is higher relative to the outcome value difference indicators of other of the partitions, and to assign a lower performing agent to a caller from one of the partitions when the outcome value difference indicator for the partition of the caller is lower relative to the outcome value difference indicators of other of the partitions, wherein the higher performing agent and the lower performing agent are determined based at least in part on the agent performance data for the respective agent relative to the agent performance data for other of the agents.

19. The system as defined in claim 14, wherein the matching step further comprises the steps:

calculating, by the one or more computers, a ranking or percentile for the partitions by outcome value difference indicator from a high outcome value difference indicator to a low outcome value difference indicator based at least in part on a first number of calls;

calculating a ranking or percentile of the agent performances, by the one or more computers, for the respective agents in the set of agents based at least in part on the respective agent performance data and a second number of calls; and

matching, by the one or more computers, the respective agent to the respective caller based at least in part on a rule to minimize a difference between the partition percentile of the outcome value difference indicator for the partition of the respective caller and the performance percentile of the respective agent.

20. The system as defined in claim 14, wherein the matching step is performed based at least in part using a multi-data element pattern matching algorithm in a pair-wise fashion to obtain a score or other indicator for each of multiple caller-agent pairs from a set of agents and a set of callers.

21. The system as defined in claim 14, further comprising correcting, by the one or more computers, for time effects by subtracting from agent performance data an epoch average from a desired outcome data sample, wherein the epoch average comprises an average of one selected from the group of hour-of-day desired outcome data, day-of-week desired outcome data, and hour-of-week desired outcome data from an epoch in which a respective call occurs, to obtain a new target outcome variable.

22. The system as defined in claim 14, further comprising correcting, by the one or more computers, for time effects by dividing agent performance data by an epoch average from a desired outcome data sample, wherein the epoch average comprises an average of one selected from the group of hour-of-day desired outcome data, day-of-week desired outcome data, and hour-of-week desired outcome data from an epoch in which a respective call occurs, to obtain a new target outcome variable.

23. The system as defined in claim 14, further comprising correcting, by the one or more computers, for time effects by forming an outer product by combining hour-of-day desired outcome data, day-of-week desired outcome data, and hour-of-week desired outcome data from an epoch in which a respective call occurs to obtain a time effects factor;

and

calculating, by the one or more computers, the agent performance using the time effects factor in a Bayesian Mean Regression calculation of the agent performance.

24. The system as defined in claim 14, wherein the one or more computers are configured with program code to perform, when executed, the steps of:

switching, by the one or more computers, between the outcome value difference indicator matching algorithm and a second matching algorithm that is different from the outcome value difference indicator matching algorithm, based on one or more criteria; and

when there has been switching to the second matching algorithm, then performing, by the one or more computers, performing the second matching algorithm to match a respective one of the agents with a respective one of the callers.

25. The system as defined in claim 24, wherein the one or more computers are configured with program code to perform, when executed, the steps of:

obtaining results data, by the one or more computers, using each of the outcome value difference indicator matching algorithm and the second matching algorithm in the matching step;

obtaining or receiving, by the one or more computers, a switch-over point in a distribution for agent performance and/or a distribution for caller propensity for the particular outcome where better results are obtained from one of the algorithms relative to the other of the algorithms; and

using, by the one or more computers, this switch-over point to switch between the algorithms.

26. The system as defined in claim 14, wherein the one or more computers are configured with program code to perform, when executed, the steps of:

switching, by the one or more computers, between the outcome value difference indicator matching algorithm and a second matching algorithm configured to match a respective agent with a respective ranking or percentile to a caller in one of the partitions with a closest ranking or percentile, based on one or more criteria; and

when there has been switching to the second matching algorithm, then performing, by the one or more computers, the second matching algorithm to match a respective one of the agents with a respective one of the callers.

29. The system as defined in claim 19, wherein the partition is based at least in part on one or more selected from the group of demographic data, area code, zip code, NPANXX, VTN, geographic area, 800 number, and transfer number.

27. A a non-transitory computer-readable medium, comprising:

program code, that when executed by one or more computers, causes the performance of the steps;

obtaining, by the one or more computers, agent performance data for a set of agents;

ranking, by the one or more computers, the agents based at least in part on the agent performance data;

dividing, by the one or more computers, the agents into agent performance ranges based at least in part on the ranking step;

partitioning, by the one or more computers, callers in a set of callers based on one or more criteria into a set of partitions;

determining for each of the partitions, by the one or more computers, an outcome value comprising a statistical chance or likelihood for a particular outcome at least for a first one of the agent performance ranges and for a second one of the agent performance ranges;

calculating, by the one or more computers, for each of the partitions an outcome value difference indicator based at least in part on a difference between the outcome value for the first agent performance range and the outcome value for the second agent performance range; and

matching, by the one or more computers, a respective agent with respective performance data to a respective caller in one of the partitions, based at least in part on the outcome value difference indicators for the partitions.