Method and system for scheduling inbound inquiries

• US 6,956,941 B1
• Filed: 04/12/2000
• Issued: 10/18/2005
• Est. Priority Date: 04/12/2000
• Status: Expired due to Term

• Current Assignee: CommWorks Solutions, LLC (IP Investments Group LLC)
• Original Assignee: ALI Solutions, Inc. (Noble Systems Corporation)
• Inventors: Miller, Thomas J., Svoronos, Alexander N., Duncan, Daniel N.
• Primary Examiner(s): Bui, Bing Q.
• Application Number:
  US09/547,627
• Time in Patent Office: 2,015 Days
• Field of Search: 379/67.01, 379/207.14, 379/207.15, 379/210.01, 379/265.01, 379/265.02, 379/265.09, 379/266.01, 379/309, 379/900, 379/88.18, 379/88.17
• US Class Current: 379/265.01
• CPC Class Codes: H04L 51/04 Real-time or near real-time messaging, e.g. instant messaging [IM]
• CPC Class Codes: H04L 51/226 Delivery according to priorities
• CPC Class Codes: H04M 3/523 with call distribution or queueing

Abstract

A method and system schedules inbound inquiries, such as inbound telephone calls, for response by agents in an order that is based in part on the forecasted outcome of the inbound inquiries. A scheduling module applies inquiry information to a model to forecast the outcome of an inbound inquiry. The forecasted outcome is used to set a priority value for ordering the inquiry. The priority value may be determined by solving a constrained optimization problem that seeks to maximize an objective function, such as maximizing an agent'"'"'s productivity to produce sales or to minimize inbound call attrition. The inbound call may be placed on a virtual hold or be responded to on a real-time basis based on the inbound inquiry'"'"'s priority value. A modeling module generates models that forecast inquiry outcomes based on a history and inquiry information. Statistical analysis such as regression analysis determines the model with the outcome related to the nature of the inquiry. Forecasted outcomes are based on the goal of the inbound calls and include factors such as probability an inbound caller will hang up, probability that an inbound caller will alter a business relationship based on hold time, probability that an inbound caller will make a purchase, and the relative probable reward of responding to an inbound call.

Claims

1. A method for ordering inbound inquiries, the method comprising:

receiving plural inbound inquiries, each inbound inquiry having associated inquiry information;

applying a model to the inquiry information to determine a priority value for each inquiry, the model estimating the probability of an outcome of an inbound inquiry having a predetermined result; and

ordering the inbound inquiries with the priority values.

2. The method of claim 1 wherein the method inquiries comprise e-mail messages.

3. The method of claim 1 wherein the method inquiries comprise instant messages.

4. The method of claim 1 wherein the inbound inquiries comprise inbound telephone calls having associated caller information.

5. The method of claim 4 wherein the caller information comprises automatic number identification information.

6. The method of claim 4 wherein the caller information comprise destination number identification information.

7. The method of claim 4 further comprising:

gathering the caller information with a voice response unit.

8. The method of claim 4 further comprising:

associating demographic information with each inbound telephone call based on the caller information of the inbound call; and

applying the model to the caller information to determine the priority value for each telephone call.

9. The method of claim 4 wherein the model predicts caller behavior.

10. The method of claim 9 wherein the priority value comprises a probability that the telephone call will result in a purchase.

11. The method of claim 9 wherein the priority value comprises a probability that the caller associated with the telephone call will terminate the call after a hold time.

12. The method of claim 1 further comprising:

developing plural models from a history of inbound inquiries to forecast plural outcomes that determine the priority value.

13. The method of claim 12 wherein developing the model further comprises:

applying regression analysis to the history to calculate the priority value.

14. The method of claim 12 further comprising:

determining the outcomes of the plural inbound inquiries; and

updating the history with the outcomes of the plural inbound inquiries.

15. The method of claim 12 wherein developing the caller model further comprises:

updating the model with the updated history.

16. A method for determining inbound telephone call priority, the method comprising:

developing one or more models from a history of inbound calls, the history having caller information and outcome results from inbound telephone calls;

applying the model to caller information of a pending inbound call to predict an outcome of the pending inbound call; and

associating a priority with the pending inbound call, the priority based on the predicted outcome.

17. The method of claim 16 wherein the caller information comprises telephony information received with the pending inbound call.

18. The method of claim 17 wherein the telephony information comprises automatic number identification information.

19. The method of claim 17 wherein the telephony information comprises destination number identification information.

20. The method of claim 17 wherein the caller information further comprises account information, the method further comprising:

obtaining account information for the pending inbound call, the account information stored in a database by association with the telephony information.

21. The method of claim 17 wherein the telephony information further comprises information input by the caller through a voice response unit.

22. The method of claim 21 further comprising:

obtaining account information for the pending inbound call based on the telephony information.

23. The method of claim 16 wherein developing a model further comprises:

using the caller information as predictive variables that model outcome results.

24. The method of claim 23 wherein the model comprises a logistic regression model.

25. The method of claim 23 wherein the model comprises a linear regression model.

26. The method of claim 16 further comprising:

placing the pending inbound call in the queue of an automatic call distribution system in an order based on the priority of the pending inbound call.

27. The method of claim 26 wherein the predicted outcome comprises a purchase resulting from the pending inbound call.

28. The method of claim 26 wherein the predicted outcome comprises the hold time of the pending inbound call.

29. The method of claim 16 wherein associating a priority further comprises optimizing the order for the inbound telephone calls.

30. The method of claim 29 wherein optimizing the order comprises solving a constrained optimization problem using one or estimates from one or more models.

31. The method of claim 29 wherein optimizing further comprises maximizing agent productivity to minimize caller attrition.

32. The method of claim 29 wherein optimizing further comprises maximizing agent productivity to produce sales.

33. A system for scheduling inbound calls, the system comprising:

a receiving device operable to receive plural inbound inquiries and to provide the inbound inquiries to one or more agents; and

a scheduling module interfaced with the receiving device, the scheduling module operable to order the inbound inquiries for handling by the receiving device, the order based in part on the predicted outcome of the inbound inquiries.

34. The system of claim 33 wherein the inbound inquiries comprise inbound telephone calls.

35. The system of claim 33 wherein the receiving device comprises an automatic call distribution system.

36. The system of claim 33 wherein the receiving device comprises a server that supports voice over internet protocol.

37. The system of claim 33 wherein the receiving device comprises a voice response unit.

38. The system of claim 34 further comprising:

an inbound call history data base operable to store outcome results and caller information from plural completed inbound calls; and

a modeling module interfaced with the history database and operable to model inbound call outcomes from the stored outcome results and caller information.

39. A system for responding to inbound calls, the system comprising:

a telephone call receiving device interfaced with a network to receive plural inbound calls; and

a scheduling system associated with the receiving device and having a scheduling module that prioritizes the inbound calls in accordance with forecasted outcomes for the inbound calls;

wherein the scheduling system places one or more inbound calls on hold and then releases the inbound call from hold based on the priority of the inbound call.

40. The system of claim 39 wherein the telephone call receiving device comprises an automatic call distribution system that incorporates the scheduling system.

41. The system of claim 39 wherein the scheduling system forecasts outcomes with a model derived from a history of inbound calls.

42. The system of claim 39 wherein the scheduling system orders the inbound calls to optimize an objective function.

43. The system of claim 42 wherein the objective function comprises agent productivity to minimize inbound call attrition.