Predictive intelligence for service and support

• US 10,311,364 B2
• Filed: 11/19/2014
• Issued: 06/04/2019
• Est. Priority Date: 11/19/2013
• Status: Active Grant

• Current Assignee: Salesforce.com, Inc.
• Original Assignee: Salesforce.com, Inc.
• Inventors: Beechuk, Scott D., Ramsey, Michael, Krishnan, Arvind
• Primary Examiner(s): Nilsson, Eric
• Application Number:
  US14/548,188
• Publication Number:
  US 20150142720A1
• Time in Patent Office: 1,658 Days
• CPC Class Codes: G06N 5/048 Fuzzy inferencing

Abstract

Embodiments regard predictive intelligence for service and support. An embodiment of a method includes receiving at an interface for a database system data entered in one or more data fields regarding service of a first case for a first client, the data being received from the first client or from an agent entering data regarding the first case; automatically comparing by the database system the data for the first case with data for other cases stored in a data store of the database system; identifying by the database system one or more similar cases based on the comparison of data; automatically generating by a predictive intelligence algorithm applied by the database system one or more predictive actions for the first case based on the identified one or more similar cases; and transmitting data regarding the one or more predictive actions to the first client or agent.

Claims

1. A method performed by a database system having at least a processor and a memory therein, wherein the method comprises:

displaying to an agent'"'"'s display screen interface, data entered into a plurality of data fields initiating a service request of a first case for a first client, wherein the data is received from the agent regarding the first case and entered into the database system via the agent'"'"'s display screen interface;

wherein the data received for the plurality of data fields includes at least a subset of the plurality of data fields received as blank entries;

returning results of a predictive query against the database system to the agent'"'"'s display screen interface to complete one or more of the plurality of data fields received as blank entries with predicted values to form a case report more complete than the data entered into the plurality of data fields regarding the first case from the first client or the agent;

comparing, via the database system, the first case with other cases stored within the database system;

identifying, via the database system, one or more similar cases to the first case based on the comparison and displaying the one or more similar cases identified to a related cases interface at the agent'"'"'s display screen interface;

generating, via a predictive intelligence algorithm applied by the database system, one or more predictive actions for the first case based on the identified one or more similar cases, wherein the one or more predictive actions comprises at least making a determination of severity for a problem defined by the case report;

calculating a confidence value for the one or more predictive actions; and

executing a routing rules module via the processor and memory of the database system causing the first case to be routed past one or more lower tiers of service via a pre-escalation procedure based on the determination of severity for the problem and the confidence value for the determination of severity for the problem exceeding a threshold.

2. The method of claim 1, wherein comparing the case report with other cases comprises comparing the content of the case report with earlier cases;

and

wherein the comparing includes at least analyzing case object fields and case child object fields of the case report and the earlier cases and extracting keywords from one or more of subject, description, and other text fields of the case report and the earlier cases.

3. The method of claim 1, wherein generating the one or more predictive action includes applying a confidence factor representing a level of confidence in a prediction.

4. The method of claim 1:

wherein service request data is received in connection with the service request from the first client and wherein the service request data is displayed to the agent via the agent'"'"'s display screen interface; and

wherein the one or more similar cases include one or more resolved cases, one or more on-going cases, or both.

5. The method of claim 1, wherein performing a predictive query against the database system to complete one or more of the plurality of data fields received as blank entries with predicted values comprises predicting category and application area fields based on a free text description of the first case from an email channel associated with the first case.

6. The method of claim 1, wherein the one or more predictive actions include replacement of data for one or more data fields that are determined to be erroneous.

7. The method of claim 1, wherein the one or more predictive actions include one or more of:

the determination of the severity of the problem for the first case, and pre-escalation of service for the first case based on the severity of the problem for the first case and based further on a SLA (Service Level Agreement) violation risk; and

wherein the pre-escalation completely bypasses the one or more lower tiers of service when the predictive intelligence algorithm calculates the confidence value for the one or more predictive actions in excess of the threshold indicating that the one or more lower tiers of service will be wasted.

8. The method of claim 1, wherein the predictive action includes one or more of:

recommendation of questions to be directed to the first client;

routing of the case;

recommendation of a solution based on a solution for one or more similar cases; and

identification of one or more experts for the first case.

9. A database system comprising:

a processor and a memory to execute instructions;

an agent'"'"'s display screen interface to execute via the processor and the memory of the database system, wherein the agent'"'"'s display screen interface is to display data entered into a plurality of data fields initiating a service request of a first case for a first client, wherein the data is received from the agent regarding the first case and entered into the database system via the agent'"'"'s display screen interface;

wherein the data received for the plurality of data fields includes at least a subset of the plurality of data fields received as blank entries;

a query interface to return results of a predictive query against the database system to the agent'"'"'s display screen interface to complete one or more of the plurality of data fields received as blank entries with predicted values to form a case report more complete than the data entered into the plurality of data fields regarding the first case from the first client or the agent;

an analysis engine to compare the first case with other cases stored within the database system;

wherein the analysis engine is to further identify one or more similar cases to the first case based on the comparison and displaying the one or more similar cases identified to a related cases interface at the agent'"'"'s display screen interface;

a predictive intelligence engine to generate via a predictive intelligence algorithm applied by the database system, one or more predictive actions for the first case based on the identified one or more similar cases, wherein the one or more predictive actions comprises at least making a determination of severity for a problem defined by the case report;

wherein the predictive intelligence engine is to further calculate a confidence value for the one or more predictive actions; and

a routing rules module to execute via the processor and memory of the database system, wherein execution of the routing rules module causes the first case to be routed past one or more lower tiers of service via a pre-escalation procedure based on the determination of severity for the problem and the confidence value for the determination of severity for the problem exceeding a threshold.

10. The database system of claim 9, wherein comparing the case report with other cases comprises comparing the content of the case report with earlier cases;

and

wherein the comparing includes at least analyzing case object fields and case child object fields of the case report and the earlier cases and extracting keywords from one or more of subject, description, and other text fields of the case report and the earlier cases.

11. The database system of claim 9, wherein the interface is to receive the data regarding the first case in connection with the service request from the first client received by the agent.

12. The database system of claim 9:

wherein the database system is a multi-tenant database system; and

wherein the received data is entered in a local computing apparatus, the local computer apparatus being linked to the multi-tenant database system by a network connection.

13. The database system of claim 9, wherein generating the one or more predictive actions includes applying a confidence factor representing a level of confidence in a prediction.

14. Non-transitory computer-readable storage media having instructions stored thereupon that, when executed by a processor of a database system, the instructions cause the database system to perform operations including:

displaying to an agent'"'"'s display screen interface, data entered into a plurality of data fields initiating a service request of a first case for a first client, wherein the data is received from the agent regarding the first case and entered into the database system via the agent'"'"'s display screen interface;

wherein the data received for the plurality of data fields includes at least a subset of the plurality of data fields received as blank entries;

returning results of a predictive query against the database system to the agent'"'"'s display screen interface to complete one or more of the plurality of data fields received as blank entries with predicted values to form a case report more complete than the data entered into the plurality of data fields regarding the first case from the first client or the agent;

comparing, via the database system, the first case with other cases stored within the database system;

identifying, via the database system, one or more similar cases to the first case based on the comparison and displaying the one or more similar cases identified to a related cases interface at the agent'"'"'s display screen interface;

generating, via a predictive intelligence algorithm applied by the database system, one or more predictive actions for the first case based on the identified one or more similar cases, wherein the one or more predictive actions comprises at least making a determination of severity for a problem defined by the case report;

calculating a confidence value for the one or more predictive actions; and

executing a routing rules module via the processor and memory of the database system causing the first case to be routed past one or more lower tiers of service via a pre-escalation procedure based on the determination of severity for the problem and the confidence value for the determination of severity for the problem exceeding a threshold.

15. The non-transitory computer-readable storage media of claim 14, wherein comparing the case report with other cases comprises comparing the content of the case report with earlier cases;

and

wherein the comparing includes at least analyzing case object fields and case child object fields of the case report and the earlier cases and extracting keywords from one or more of subject, description, and other text fields of the case report and the earlier cases.

16. The non-transitory computer-readable storage media of claim 14, wherein generating the one or more predictive action includes applying a confidence factor representing a level of confidence in a prediction.

17. The non-transitory computer-readable storage media of claim 14:

wherein service request data is received in connection with the service request from the first client and wherein the service request data is displayed to the agent via the agent'"'"'s display screen interface; and

wherein the one or more similar cases include one or more resolved cases, one or more on-going cases, or both.

18. The non-transitory computer-readable storage media of claim 14, wherein the one or more predictive actions includes one or more of pre-population of one or more data fields for the first case, and completion of one or more data fields for the first case that are left blank.

19. The non-transitory computer-readable storage media of claim 14, wherein the one or more predictive actions include replacement of data for one or more data fields that are determined to be erroneous.

20. The non-transitory computer-readable storage media of claim 14, wherein the one or more predictive actions include one or more of:

the determination of the severity of the problem for the first case, and pre-escalation of service for the first case based on the severity of the problem for the first case and based further on a SLA (Service Level Agreement) violation risk; and

wherein the pre-escalation completely bypasses the one or more lower tiers of service when the predictive intelligence algorithm calculates the confidence value for the one or more predictive actions in excess of the threshold indicating that the one or more lower tiers of service will be wasted.

21. The non-transitory computer-readable storage media of claim 14, wherein the predictive action includes one or more of:

recommendation of questions to be directed to the first client;

routing of the case;

recommendation of a solution based on a solution for one or more similar cases; and

identification of one or more experts for the first case.