Routing callers from a set of callers based on caller data
View Patent ↗Methods and systems are provided for routing callers to agents in a call-center routing environment. An exemplary method includes pooling incoming callers, and causing a caller from the pool of callers to be routed. The caller may be routed from the pool of callers to an agent, placed in another pool of callers, or placed in a queue of callers. The caller data may include demographic or psychographic data. The caller may be routed from the pool of callers based on comparing the caller data with agent data associated with an agent via a pattern matching algorithm and/or computer model for predicting a caller-agent pair outcome. Additionally, if a caller is held beyond a hold threshold (e.g., a time, “cost” function, or the like) the caller may be routed to the next available agent.
1. A computer-implemented method for routing callers to agents in a call-center routing environment, the method comprising the acts of:
identifying, by one or more computers, caller data for each caller of a set of callers, wherein the caller data comprises demographic and/or psychographic data;
calculating for each caller of the set of callers, by the one or more computers, for each of at least two desired variables, a desired variable score using a multi-element pattern matching algorithm based at least in part on the demographic and/or psychographic data of the respective caller;
calculating, by the one or more computers, a Z-score for each of the at least two desired variable scores of the multi-element pattern matching algorithm, for each of the callers in the set of callers;
calculating, by the one or more computers, for each of the callers, a linear combination of the Z-scores to obtain a caller linear combination value;
identifying, by one or more computers, agent data for each agent of a set of agents, wherein the agent data comprises psychographic and/or historical data;
calculating for each agent of the set of agents, by the one or more computers, for each of the at least two desired variables, a desired variable score using a multi-element pattern matching algorithm based at least in part on the psychographic and/or historical data of the respective agent;
calculating, by the one or more computers, a Z-score for each of the at least two desired variable scores of the multi-element pattern matching algorithm, for each of the agents in the set of agents;
calculating, by the one or more computers, for each of the agents, a linear combination of the Z-scores to obtain an agent linear combination value;
selecting, by the one or more computers, one of the agents of the set of agents and one of the callers of the set of callers based at least in part on the agent linear combination value of the one agent and the caller linear combination value of the one caller; and
causing, by the one or more computers, the one caller of the set of callers to be routed to the one agent of the set of agents based on the selecting step.
2. The computer-implemented method of claim 1 , wherein the set of callers are in a queue and the caller is routed out of order within the queue.
3. The computer-implemented method of claim 1 , wherein the set of callers are in a pool of callers.
4. The computer-implemented method of claim 1 , further comprising routing the caller to the next available agent if a hold threshold for the caller is exceeded.
5. The computer-implemented method of claim 4 , wherein the hold threshold comprises one or more of a time period, number of times the caller is skipped by other callers, or a multiple of a predicted hold time.
6. The computer-implemented method of claim 1 , wherein the selecting step selects the one agent based at least in part on availability data for the one agent.
7. The computer-implemented method of claim 1 , further comprising:
weighting, by the one or more computers, the Z-score for one of the at least two desired variable scores of the pattern matching algorithm.
8. The computer-implemented method of claim 1 , wherein the selecting step is based at least in part on a criterion to maximize a combination of the agent linear combination value of the one agent and the caller linear combination value of the one caller.
9. The computer-implemented method of claim 1 , wherein the selecting step is based at least in part on a criterion that a combination of the agent linear combination value of the one agent and the caller linear combination value of the one caller exceeds a preset threshold.
10. A system for routing callers to agents in a call-center environment, the apparatus comprising:
one or more computers configured with the following components:
a collector to identify, by the one or more computers, caller data for each caller of a set of callers, wherein the caller data comprises demographic and/or psychographic data;
a routing engine configured to calculate for each caller of the set of callers, by the one or more computers, for each of at least two desired variables, a desired variable score using a multi-element pattern matching algorithm based at least in part on the demographic and/or psychographic data of the respective caller;
the routing engine configured to calculate, by the one or more computers, a Z-score for each of the at least two desired variable scores of a multi-element pattern matching algorithm, for each of the callers in the set of callers;
the routing engine configured to calculate, by the one or more computers, for each of the callers, a linear combination of the Z-scores to obtain a caller linear combination value;
the collector configured to identify, by one or more computers, agent data for each agent of a set of agents, wherein the agent data comprises psychographic and/or historical data;
the routing engine configured to calculate for each agent of the set of agents, by the one or more computers, for each of the at least two desired variables, a desired variable score using a multi-element pattern matching algorithm based at least in part on the psychographic and/or historical data of the respective agent;
the routing engine configured to calculate, by the one or more computers, a Z-score for each of the at least two desired variable scores of the multi-element pattern matching algorithm, for each of the agents in the set of agents;
the routing engine configured to calculate, by the one or more computers, for each of the agents, a linear combination of the Z-scores to obtain an agent linear combination value;
the routing engine configured to select, by the one or more computers, one of the agents of the set of agents and one of the callers of the set of callers based at least in part on the agent linear combination value of the one agent and the caller linear combination value of the one caller; and
the routing engine configured to cause, by the one or more computers, the one caller of the set of callers to be routed to the one agent of the set of agents based on the selecting step.
11. The system of claim 10 , wherein the set of callers are in a queue and the caller is routed out of order within the queue.
12. The system of claim 10 , wherein the set of callers are in a pool of callers.
13. The system of claim 10 , wherein the routing engine is configured to route the caller to the next available agent if a hold threshold for the caller is exceeded.
14. The system of claim 13 , wherein the hold threshold comprises one or more of a time period, number of times the caller is skipped by other callers, or a multiple of a predicted hold time.
15. The system of claim 10 , wherein the routing engine is configured to select the one agent based at least in part on availability data for the one agent.
16. The system of claim 10 , wherein the routing engine is configured to weight, by the one or more computers, the Z-score for one of the at least two desired variable scores of the pattern matching algorithm.
17. The system of claim 10 , wherein the routing engine is configured to select based at least in part on a criterion to maximize a combination of the agent linear combination value of the one agent and the caller linear combination value of the one caller.
18. The system of claim 10 , wherein the routing engine is configured to select based at least in part on a criterion that a combination of the agent linear combination value of the one agent and the caller linear combination value of the one caller exceeds a preset threshold.
19. Non-transitory computer readable storage medium comprising computer readable instructions for carrying out, when executed on one or more computers, a method comprising:
identifying, by the one or more computers, caller data for each caller of a set of callers, wherein the caller data comprises demographic and/or psychographic data;
calculating for each caller of the set of callers, by the one or more computers, for each of at least two desired variables, a desired variable score using a multi-element pattern matching algorithm based at least in part on the demographic and/or psychographic data of the respective caller;
calculating, by the one or more computers, a Z-score for each of the at least two desired variable scores of the multi-element pattern matching algorithm, for each of the callers in the set of callers;
calculating, by the one or more computers, for each of the callers, a linear combination of the Z-scores to obtain a caller linear combination value;
identifying, by one or more computers, agent data for each agent of a set of agents, wherein the agent data comprises psychographic and/or historical data;
calculating for each agent of the set of agents, by the one or more computers, for each of the at least two desired variables, a desired variable score using a multi-element pattern matching algorithm based at least in part on the psychographic and/or historical data of the respective agent;
calculating, by the one or more computers, a Z-score for each of the at least two desired variable scores of the multi-element pattern matching algorithm, for each of the agents in the set of agents;
calculating, by the one or more computers, for each of the agents, a linear combination of the Z-scores to obtain an agent linear combination value;
selecting, by the one or more computers, one of the agents of the set of agents and one of the callers of the set of callers based at least in part on the agent linear combination value of the one agent and the caller linear combination value of the one caller; and
causing, by the one or more computers, the one caller of the set of callers to be routed to the one agent of the set of agents based on the selecting step.
20. The computer readable medium of claim 19 , wherein the set of callers are in a queue and the caller is routed out of order within the queue.
21. The computer readable medium of claim 19 , wherein the set of callers are in a pool of callers.
22. The computer readable medium of claim 19 , further comprising instructions for routing the caller to the next available agent if a hold threshold for the caller is exceeded.
23. The computer readable medium of claim 22 , wherein the hold threshold comprises one or more of a time period, number of times the caller is skipped by other callers, or a multiple of a predicted hold time.
24. The computer readable medium of claim 19 , further comprising computer readable instructions to perform the selecting step based at least in part on availability data for the one agent.
25. The computer readable medium of claim 19 , further comprising:
weighting, by the one or more computers, the Z-score for one of the at least two desired variable scores of the pattern matching algorithm.
26. The computer readable medium of claim 19 , wherein the selecting step is based at least in part on a criterion to maximize a combination of the agent linear combination value of the one agent and the caller linear combination value of the one caller.
27. The computer readable medium of claim 19 , wherein the selecting step is based at least in part on a criterion that a combination of the agent linear combination value of the one agent and the caller linear combination value of the one caller exceeds a preset threshold.