Patent Issued for System And Method For Managing Routing Of Customer Calls To Agents (USPTO 10,860,937)

2020 DEC 21 (NewsRx) -- By a News Reporter-Staff News Editor at Insurance Daily News -- Massachusetts Mutual Life Insurance Company (Springfield, Massachusetts, United States) has been issued patent number 10,860,937, according to news reporting originating out of Alexandria, Virginia, by NewsRx editors.

The patent’s inventor is Merritt, Sears (Groton, MA).

This patent was filed on February 14, 2020 and was published online on December 21, 2020.

From the background information supplied by the inventors, news correspondents obtained the following quote: “Customer contact centers provide an important interface for customers/partners of an organization to contact the organization. The contact can be for a request for a product or service, for trouble reporting, service request, etc. The contact mechanism in a conventional call center is via a telephone, but it could be via a number of other electronic channels, including e-mail, online chat, etc.

“The contact center consists of a number of human agents, each assigned to a telecommunication device, such as a phone or a computer for conducting email or Internet chat sessions, that is connected to a central switch. Using these devices, the agents generally provide sales, customer service, or technical support to the customers or prospective customers of a contact center, or of a contact center’s clients. Conventionally, a contact center operation includes a switch system that connects callers to agents. In an inbound contact center, these switches route inbound callers to a particular agent in a contact center, or, if multiple contact centers are deployed, to a particular contact center for further routing. When a call is received at a contact center (which can be physically distributed, e.g., the agents may or may not be in a single physical location), if a call is not answered immediately, the switch will typically place the caller on hold and then route the caller to the next agent that becomes available. This is sometimes referred to as placing the caller is in a call queue. In conventional methods of routing inbound callers to agents, high business value calls can be subjected to a long wait while the low business value calls are often answered more promptly, possibly causing dissatisfaction on the part of the high business value caller.

“There is a need for a system and method for identifying high business value inbound callers at a call center during a time period in which inbound callers are awaiting connection to an agent. Additionally, there is a need to improve traditional methods of routing callers, such as ‘round-robin’ caller routing, to improve allocation of limited call center resources to high business value inbound callers.”

Supplementing the background information on this patent, NewsRx reporters also obtained the inventor’s summary information for this patent: “Embodiments described herein can automatically route an inbound call from a customer to one of a plurality of queues (e.g., two queues) based on predicted value of the inbound telephone call. Upon identifying the customer, the process retrieves customer demographic data associated with a customer identifier for the identified customer. A predictive model determines a value prediction signal for the identified customer. Based on the value prediction signal determined, the predictive model classifies the identified customer into one of a first value group and a second value group. In the event the predictive model classifies the identified customer into the first value group, the call management system routes the identified customer to a first call queue for connection to one of a first pool of call center agents who are authorized to present the offer to purchase the product. In the event the predictive model classifies the identified customer into the second value group, the call management system routes the identified customer to a second call queue for connection to one of a second pool of call center agents who are not authorized to present the offer to purchase the product.

“The predictive model can include a logistic regression model and a tree based model. In an embodiment, the predictive model determines the value prediction signal in real time by applying a logistic regression model in conjunction with a tree based model to the retrieved customer demographic data. In an embodiment, the logistic regression model employs l.sub.1 regularization. In an embodiment, the logistic regression model employs l.sub.2 regularization. In an embodiment, the tree based model is a random forests ensemble learning method for classification.

“The value prediction signal can include one or more of a first signal representative of a likelihood that the identified customer will accept an offer to purchase a product, a second signal representative of a likelihood that the identified customer will lapse in payments for a purchased product, and a third signal representative of a likelihood that the identified customer will accept an offer to purchase the product and will not lapse in payments for the purchased product. In various embodiments, the a value prediction signal is a buy-only signal, a lapse-only signal, a buy-don’t-lapse signal, or combination of these signals.

“The customer identifier can include two or more of name of the identified customer, address of the identified customer, and zip code of the identified customer. In an embodiment, the customer management system obtains a customer identifier for the received customer call from caller information associated with the inbound calls. In an embodiment, the customer management system obtains a customer identifier for the received customer call via an automated telegreeter of a Voice Response Unit (‘VRU’) system. In an embodiment, the customer management system obtains a customer identifier for the received customer call from a third party directory service.

“In one embodiment, a processor-based method comprises executing, by a processor, a predictive machine-learning model configured to determine, for each lead profile of a plurality of lead records stored in an internal database of a contact center, a value prediction signal by inputting customer demographic data, payment data, marketing costs data, and lapse data into a logistic regression model operating in conjunction with a tree based model, the predictive machine-learning model outputting a first subset of the plurality of lead records into a first value group and a second subset of the plurality of lead records into a second value group, wherein the value prediction signal comprises one or more of a first signal representative of a likelihood that the identified customer will accept an offer to purchase a product, a second signal representative of a likelihood that the identified customer will lapse in payments for a purchased product, and a third signal representative of a likelihood that the identified customer will accept an offer to purchase the product and will not lapse in payments for the purchased product, and wherein the predictive machine-learning model is continually trained using customer demographic data, updated payment data, updated marketing costs data, and updated lapse data; and running, by the processor, the predictive machine-learning model upon receiving a customer call from an identified customer at an inbound call receiving device of the contact center to: retrieve the customer demographic data for the identified customer; classify the identified customer into one of the first value group and the second value group; and direct the inbound call receiving device: in the event the processor classifies the identified customer into the first value group, to route the identified customer to a first call queue for connection to one of a first pool of call center agents; in the event the processor classifies the identified customer into the second value group, to route the identified customer to a second call queue for connection to one of a second pool of call center agents.

“In another embodiment, a processor based method for managing customer calls within a call center comprises retrieving, by a processor, customer demographic data associated with a customer identifier for an identified customer in a customer call; determining, by a predictive model executing on the processor, a value prediction signal comprising one or more of a first signal representative of a likelihood that the identified customer will accept an offer to purchase a product, a second signal representative of a likelihood that the identified customer will lapse in payments for a purchased product, and a third signal representative of a likelihood that the identified customer will accept an offer to purchase the product and will not lapse in payments for the purchased product; wherein the predictive model comprises a logistic regression model operating in conjunction with a tree based model; classifying, by the predictive model executing on the processor based on the value prediction signal determined by the predictive model, the identified customer into one of a first value group and a second value group, wherein the first value group comprises customers having a first set of modeled lifetime values, and the second value group comprises customers having a second set of modeled lifetime values, wherein modeled lifetime values in the first set of modeled lifetime values are higher than modeled lifetime values in the second set of modeled lifetime values; and in the event the classifying step classifies the identified customer into the first value group, routing, by the processor, the identified customer to a first call queue for connection to one of a first pool of call center agents; in the event the classifying step classifies the identified customer into the second value group, routing, by the processor, the identified customer to a second call queue for connection to one of a second pool of call center agents.

“In a further embodiment, a system for managing customer calls within a call center, comprises an inbound telephone call receiving device for receiving a customer call to the call center; non-transitory machine-readable memory that stores historical information for leads of the call center comprising payment data, marketing costs data, and lapse data; a predictive modeling module that stores a predictive model of customer value, wherein the predictive model comprises a logistic regression model operating in conjunction with a tree based model; and a processor, configured to execute an inbound queue management module, wherein the processor in communication with the non-transitory machine-readable memory and the predictive models module executes a set of instructions instructing the processor to: retrieve customer demographic data associated with a customer identifier for an identified customer in the customer call received by the inbound telephone call receiving device, wherein the customer identifier comprises two or more of name of the identified customer, address of the identified customer, and zip code of the identified customer; retrieve from the non-transitory machine readable memory the historical information for leads of the call center comprising payment data, marketing costs data, and lapse data; determine a value prediction signal for the identified customer via analysis by the predictive model of the customer demographic data associated with the customer identifier for the identified customer and via analysis of historical payment data and lapse data of the call center; wherein the value prediction signal comprises one or more of a first signal representative of a likelihood that the identified customer will accept an offer to purchase a product, a second signal representative of a likelihood that the identified customer will lapse in payments for a purchased product, and a third signal representative of a likelihood that the identified customer will accept an offer to purchase the product and will not lapse in payments for the purchased product; classify the identified customer into one of a first value group and a second value group based on the value prediction signal, wherein the first value group comprises customers having a first set of modeled lifetime values, and the second value group comprises customers having a second set of modeled lifetime values, wherein modeled lifetime values in the first set of modeled lifetime values are higher than modeled lifetime values in the second set of modeled lifetime values; and direct the inbound telephone call receiving device: in the event the inbound queue management module classifies the identified customer into the first value group, to route the identified customer to a first call queue for connection to one of a first pool of call center agents; in the event the inbound queue management module classifies the identified customer into the second value group, to route the identified customer to a second call queue for connection to one of a second pool of call center agents.

“In another embodiment, a processor-based method comprises executing, by a processor, a predictive machine-learning model configured to determine, for each lead profile of a plurality of lead records stored in an internal database of a contact center, a value prediction signal by inputting customer demographic data, payment data and lapse data into a logistic regression model operating in conjunction with a tree based model, the predictive machine-learning model outputting a first subset of the plurality of lead records into a first value group and a second subset of the plurality of lead records into a second value group, wherein the value prediction signal comprises one or more of a first signal representative of a likelihood that the identified customer will accept an offer to purchase a product, a second signal representative of a likelihood that the identified customer will lapse in payments for a purchased product, and a third signal representative of a likelihood that the identified customer will accept an offer to purchase the product and will not lapse in payments for the purchased product, and wherein the predictive machine-learning model is continually trained using customer demographic data, updated payment data, and updated lapse data; and running, by the processor, the predictive machine-learning model upon receiving a customer call from an identified customer at an inbound call receiving device of the contact center: to retrieve the customer demographic data for the identified customer; to classify the identified customer into one of the first value group and the second value group; and to direct the inbound call receiving device: in the event the processor classifies the identified customer into the first value group, to route the identified customer to a first call queue for connection to one of a first pool of call center agents who are authorized to present the offer to purchase the product; in the event the processor classifies the identified customer into the second value group, to route the identified customer to a second call queue for connection to one of a second pool of call center agents who are not authorized to present the offer to purchase the product.

“In another embodiment, a processor based method for managing customer calls within a call center comprises retrieving, by a processor, customer demographic data associated with a customer identifier for an identified customer in a customer call; determining, by a predictive model executing on the processor, a value prediction signal comprising one or more of a first signal representative of a likelihood that the identified customer will accept an offer to purchase a product, a second signal representative of a likelihood that the identified customer will lapse in payments for a purchased product, and a third signal representative of a likelihood that the identified customer will accept an offer to purchase the product and will not lapse in payments for the purchased product; wherein the predictive model comprises a logistic regression model operating in conjunction with a tree based model; classifying, by the predictive model executing on the processor based on the value prediction signal determined by the predictive model, the identified customer into one of a first value group and a second value group; and in the event the classifying step classifies the identified customer into the first value group, routing, by the processor, the identified customer to a first call queue for connection to one of a first pool of call center agents who are authorized to present the offer to purchase the product; in the event the classifying step classifies the identified customer into the second value group, routing, by the processor, the identified customer to a second call queue for connection to one of a second pool of call center agents who are not authorized to present the offer to purchase the product.

“In yet another embodiment, a system for managing customer calls within a call center, comprises an inbound telephone call receiving device for receiving a customer call to the call center; non-transitory machine-readable memory that stores historical information about leads, customers, and marketing costs of the call center; a predictive modeling module that stores a predictive model of customer value, wherein the predictive model comprises a logistic regression model operating in conjunction with a tree based model; and a processor, configured to execute an inbound queue management module, wherein the processor in communication with the non-transitory machine-readable memory and the predictive models module executes a set of instructions instructing the processor to: retrieve external third-party customer demographic data associated with a customer identifier for an identified customer in the customer call received by the inbound telephone call receiving device, wherein the customer identifier comprises two or more of name of the identified customer, address of the identified customer, and zip code of the identified customer; retrieve from the non-transitory machine readable memory the historical information about leads, customers, and marketing costs of the call center; determine a value prediction signal for the identified customer via analysis by the predictive model of the third-party customer demographic data associated with the customer identifier for the identified customer and via analysis of the historical information about leads, customers, and marketing costs of the call center; wherein the value prediction signal comprises one or more of a first signal representative of a likelihood that the identified customer will accept an offer to purchase a product, a second signal representative of a likelihood that the identified customer will lapse in payments for a purchased product, and a third signal representative of a likelihood that the identified customer will accept an offer to purchase the product and will not lapse in payments for the purchased product; classify the identified customer into one of a first value group and a second value group based on the value prediction signal; and direct the inbound telephone call receiving device: in the event the inbound queue management module classifies the identified customer into the first value group, to route the identified customer to a first call queue for connection to one of a first pool of call center agents who are authorized to present the offer to purchase the product; in the event the inbound queue management module classifies the identified customer into the second value group, to route the identified customer to a second call queue for connection to one of a second pool of call center agents who are authorized to present the offer to purchase the product.

“Other objects, features, and advantages of the present disclosure will become apparent with reference to the drawings and detailed description of the illustrative embodiments that follow.”

The claims supplied by the inventors are:

“What is claimed is:

“1. A processor-based method for managing customer calls within a call center, comprising: upon establishing a call session with an identified customer at an inbound call receiving device: retrieving, by a processor, customer demographic data associated with the identified customer; executing, by the processor, a predictive machine-learning model to determine a value prediction signal by inputting the customer demographic data, wherein the value prediction signal is representative of a modeled lifetime value of the identified customer, the predictive machine-learning model classifying the identified customer into a first value group or a second value group, wherein a modeled lifetime value of the first value group is higher than a modeled lifetime value of the second value group; directing, by the processor, the inbound call receiving device to route the customer call: to a first queue assignment in the event the processor classifies the identified customer into the first value group; and to a second queue assignment in the event the processor classifies the identified customer into in the second value group.

“2. The processor based method according to claim 1, wherein the first value group comprises customers having a first set of modeled lifetime values, and the second value group comprises customers having a second set of modeled lifetime values, wherein modeled lifetime values in the first set of modeled lifetime values are higher than modeled lifetime values in the second set of modeled lifetime values.

“3. The processor based method according to claim 1, wherein the predictive machine-learning model comprises a first model operating in conjunction with a second model.

“4. The processor based method according to claim 3, wherein the first model determines a likelihood of sale of a product to the identified customer and the second model determines the value prediction signal, wherein the value prediction signal is representative of modeled lifetime value of the sale of the product to the identified customer.

“5. The processor based method according to claim 1, wherein the predictive machine-learning model is further configured to determine the value prediction signal by inputting payment data and marketing costs data.

“6. The processor based method according to claim 5, wherein the predictive machine-learning model is continually trained using updated customer demographic data, updated payment data, and updated marketing costs data.

“7. The processor based method according to claim 5, wherein the predictive machine-learning model is further configured to determine the value prediction signal by inputting lapse data.

“8. The processor based method according to claim 1, wherein the predictive machine-learning model comprises a logistic regression model.

“9. The processor based method according to claim 8, wherein the predictive machine-learning model comprises one of a logistic regression model with l.sub.1 regularization and a logistic regression model with l.sub.2 regularization.

“10. The processor based method according to claim 1, wherein the predictive machine-learning model comprises a tree based model.

“11. The processor based method according to claim 10, wherein the predictive machine-learning model comprises a random forests ensemble learning method for classification.

“12. The processor based method according to claim 1, wherein the first queue assignment comprises a call queue for connection to a first group of agents, and the second queue assignment comprises a call queue for connection to a second group of agents.

“13. The processor based method according to claim 12, wherein the first group of agents comprises agents who are authorized to present an offer to purchase a product, and the second group of agents comprises agents who are not authorized to present the offer to purchase the product.

“14. A processor based method for managing customer calls within a call center, comprising: retrieving, by a processor, customer demographic data associated with a customer identifier for an identified customer in a customer call, via a lookup tool executing on the processor to perform real time matching of customer demographic data to the customer identifier for the identified customer in the customer call; determining, by a predictive machine-learning model executing on the processor, a value prediction signal representative of a modeled lifetime value of the identified customer; classifying, by the predictive machine-learning model executing on the processor based on the value prediction signal determined by the predictive machine-learning model, the identified customer into one of a first value group and a second value group, wherein a modeled lifetime value of the first value group is higher than a modeled lifetime the second value group; and in the event the classifying step classifies the identified customer into the first value group, routing, by the processor, the identified customer to a first call queue assignment; in the event the classifying step classifies the identified customer into the second value group, routing, by the processor, the identified customer to a second call queue assignment.

“15. The processor based method according to claim 14, wherein the first value group comprises customers having a first set of modeled lifetime values, and the second value group comprises customers having a second set of modeled lifetime values, wherein modeled lifetime values in the first set of modeled lifetime values are higher than modeled lifetime values in the second set of modeled lifetime values.

“16. The processor based method according to claim 14, wherein the predictive machine-learning model comprises a first model operating in conjunction with a second model.

“17. The processor based method according to claim 16, wherein the first model determines a likelihood of sale of a product to the identified customer and the second model determines the value prediction signal, wherein the value prediction signal is representative of modeled lifetime value of the sale of the product to the identified customer.

“18. The processor based method according to claim 14, wherein the predictive machine-learning model comprises a logistic regression model.

“19. The processor based method according to claim 14, wherein the predictive machine-learning model comprises a tree based model.

“20. The processor based method according to claim 14, wherein the first queue assignment comprises a call queue for connection to a first group of agents, and the second queue assignment comprises a call queue for connection to a second group of agents.”

For the URL and additional information on this patent, see: Merritt, Sears. System And Method For Managing Routing Of Customer Calls To Agents. U.S. Patent Number 10,860,937, filed February 14, 2020, and published online on December 21, 2020. Patent URL: http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=10,860,937.PN.&OS=PN/10,860,937RS=PN/10,860,937

(Our reports deliver fact-based news of research and discoveries from around the world.)