Granted Patent B2

US 12,499,459 · App. 18/100,922 · Granted Dec 16, 2025

Inventors: Frank Passantino (Mountain View, CA); Paul Edwards (Mountain View, CA); Ate Douma (Mountain View, CA); Shekhar Kumar (Mountain View, CA)

Assignee: BloomReach, Inc.

G06Q30/0204G06Q30/0631

Abstract

Claims (45)

1 . A customer engagement method comprising:

receiving a first product search query from a visitor identified in a visitor session, wherein the first product search query includes at least customer segments associated with a visitor identifier of the visitor;

conducting a first product search using the first product search query to generate first product search results;

returning the first product search results based on the first product search query, wherein first product search results include a first listing of products;

displaying the first product search results in a web browser of the visitor;

updating, after the first product search query, the customer segments associated with the visitor identifier based on real time evaluation of visitor characteristics and/or actions, wherein the updated customer segments are exposed to product searching in response to tracked changes in at least one of visitor characteristics or actions;

generating product relevance scores per each customer segment exposed across a plurality of different visitors for a pre-defined period of time using a trained machine learning algorithm;

receiving a second product search query from the visitor, wherein the second product search query includes at least the updated customer segments associated with the visitor identifier;

conducting a second product search using the second product search query to generate second product search results;

returning, based at least on the second product search query and the product relevance scores generated by the trained machine learning algorithm, the second product search results, wherein second product search results include a second listing of products; and

displaying the second product search results in the web browser.

2 . The method of claim 1 , wherein the product relevance score ranks relevance of products in a customer segment and wherein the second product search results are based on the relevance score of the customer segments associated with the visitor identifier.

3 . The method of claim 2 , wherein when the updated customer segments are exposed to product search, the customer segments and the product relevance scores are used to personalize the second product search results for the identified user.

4 . The method of claim 1 , further comprising tracking at least one of visitor characteristics or visitor actions based on the visitor identifier.

5 . The method of claim 3 , wherein the at least one of visitor characteristics or visitor actions are tracked on a web site or in an application.

6 . A system for customer engagement comprising:

a processor configured to:

receive a first product search query from a visitor identified in a visitor session, wherein the first product search query includes at least customer segments associated with a visitor identifier of the visitor;

conduct a first product search using the first product search query to generate first product search results;

return the first product search results based on the first product search query, wherein first product search results include a first listing of products;

display the first product search results in a web browser of the visitor;

update, after the first product search query, the customer segments associated with the visitor identifier based on real time evaluation of visitor characteristics and/or actions, wherein the updated customer segments are exposed to product searching in response to tracked changes in at least one of visitor characteristics or actions;

generate product relevance scores per each customer segment exposed across a plurality of different visitors for a pre-defined period of time using a trained machine learning algorithm;

receive a second product search query from the visitor, wherein the second product search query includes at least the updated customer segments associated with the visitor identifier;

conduct a second product search using the second product search query to generate second product search results;

return, based on the second product search query and the product relevance scores generated by the trained machine learning algorithm, second product search results, wherein second product search results include a second listing of products; and

displaying the second product search results in the web browser.

7 . The system of claim 6 , wherein the relevance score ranks relevance of products in a customer segment and wherein the second product search results are based on the relevance score of the customer segments associated with the visitor identifier.

8 . The system of claim 6 , further comprising tracking at least one of visitor characteristics or visitor actions based on the visitor identifier.

9 . The system of claim 8 , wherein the at least one of visitor characteristics or visitor actions are tracked on a web site or in an application.

10 . A non-transitory computer readable medium, storing thereon computer readable instructions that when read by a computer cause a processor to perform a customer engagement method comprising:

receiving a first product search query from a visitor identified in a visitor session, wherein the first product search query includes at least customer segments associated with a visitor identifier of the visitor;

conducting a first product search using the first product search query to generate first product search results;

returning the first product search results based on the first product search query, wherein first product search results include a first listing of products;

displaying the first product search results in a web browser of the visitor;

updating, after the first product search query, the customer segments associated with the visitor identifier based on real time evaluation of visitor characteristics and/or actions, wherein the updated customer segments are exposed to product searching in response to tracked changes in at least one of visitor characteristics or actions;

generating product relevance scores per each customer segment exposed across a plurality of different visitors for a pre-defined period of time using a trained machine learning algorithm;

receiving a second product search query from the visitor, wherein the second product search query includes at least the updated customer segments associated with the visitor identifier;

conducting a second product search using the second product search query to generate second product search results;

returning, based at least on the second product search query and the product relevance scores generated by the trained machine learning algorithm, the second product search results wherein second product search results include a second listing of products; and

displaying the second product search results in the web browser.

11 . The method of claim 10 , wherein the relevance score ranks relevance of products in a customer segment and wherein the second product search results are based on the relevance score of the customer segments associated with the visitor identifier.

12 . The method of claim 10 , wherein when the updated customer segments are exposed to product search, the customer segments and the product relevance scores are used to personalize the second product search results for the identified user.

13 . The method of claim 10 , further comprising tracking at least one of visitor characteristics or visitor actions based on the visitor identifier.

14 . The method of claim 13 , wherein the at least one of visitor characteristics or visitor actions are tracked on a web site or in an application.

Assignments (2)

Continuity (1)

References Cited (15)