GENERATING AND HANDLING OPTIMIZED CONSUMER SEGMENTS

1 . A computer-implemented method, comprising:

receiving, in a server, a raw data from multiple consumer devices;

refining the raw data to capture a data pattern;

predicting a consumer behavior based on the data pattern, the consumer behavior defining one or more attributes;

identifying a consumer segment based on the consumer behavior and a sharing of a one or more attributes among multiple consumers in the consumer segment;

selecting at least one of an advertising message or a promotional offer to one or more consumers in the consumer segment to include in a payload content;

identifying a media channel to deliver the payload content to one or more consumer devices; and

providing the consumer segment to a display in a client device, upon request.

2 . The computer-implemented method of claim 1 , wherein identifying the media channel comprises selecting one of an in-store printer, a mobile video, a desktop display, or a third party advertisement, based on a type of the one or more consumer devices and a current location of the consumers.

3 . The computer-implemented method of claim 1 , further comprises receiving, in the server, from a client device, a pre-selected universe of consumers and an impact goal for the payload content, wherein the pre-selected universe of consumers includes the consumer segment and is based on a product or brand identified in the payload content, and the impact goal comprises a desired metric associating the consumer segment with the product or brand identified in the payload content.

4 . The computer-implemented method of claim 1 , further comprising determining a time duration of the promotional offer, promotion or recommendation in the payload content based on the one or more attributes of the consumers in the consumer segment.

5 . The computer-implemented method of claim 1 , further comprising selecting a list of products or brands to be included in the payload content based on the one or more attributes of the consumers in the consumer segment.

6 . The computer-implemented method of claim 1 , further comprising:

selecting a metric for the payload content, the metric associating a product or brand in the payload content to a consumer behavior;

selecting a group of consumers to form a control group based on the one or more attributes, wherein the control group does not receive the payload content;

determining an impact of the payload content on the consumer segment based on a comparison of a value of the metric for the control group with a value of the metric for the consumer segment;

and ranking the consumer segment based on the impact of the payload content on the consumer segment.

7 . The computer-implemented method of claim 1 , further comprising generating a segment profile with a list of attributes and consumer behavior associated with a percentage of consumers in the consumer segment, and providing a graphical view of the segment profile to the display in the client device, the graphical view including an indicator of the percentage of consumers in a consumer universe associated with the list of attributes and consumer behavior.

8 . The computer-implemented method of claim 1 , further comprising determining an audience extension beyond the consumer segment for the payload content when a budget and a goal of a campaign for the payload content is not reachable within the consumer segment.

9 . The computer-implemented method of claim 1 , further comprising:

predicting a campaign performance for the consumer segment based on a number of reachable users and a contact frequency of the payload content; and

accounting for a deterioration of the campaign performance based on an audience extension.

10 . The computer-implemented method of claim 1 , further comprising receiving, in the server, a request from a use to split the consumer segment into a maximum number of sub-segments to increase an impact of the payload content, wherein a sub-segment includes one or more consumers from the consumer segment.

11 . A system, comprising:

a data acquisition layer configured to test, standardize, partition and format a raw data received by a server;

a data enrichment layer, configured to refine the raw data by transformation, feature computation, and training of an auxiliary model to capture a data pattern in the raw data;

a targeting imputation module configured to impute one or more consumer attributes to define a target audience and a consumer segment;

a consumer preference module storing multiple consumer preferences for multiple products or brands and multiple consumer sensitivities for marketing impulses;

a behavior prediction module configured to predict a consumer behavior based on the consumer preferences for products and the marketing impulses; and

an application layer configured to provide a payload content to a consumer device, the payload content including a personalized advertisement or coupon for a selected product or brand based on the consumer behavior.

12 . The system of claim 11 , wherein the targeting imputation module is further configured to select a group of consumers to form a control group based on the one or more consumer attributes, wherein the control group does not receive the payload content.

13 . The system of claim 11 , wherein the behavior prediction module is configured to evaluate a metric for the payload content associating the selected product or brand in the payload content to a measured consumer behavior.

14 . The system of claim 11 , wherein the behavior prediction module is configured to evaluate an impact of the payload content on the consumer segment based on a comparison of a metric value for a control group with a metric value for the consumer segment, and to rank the consumer segment based on the impact of the payload content on the consumer segment.

15 . The system of claim 11 , wherein the behavior prediction module is configured to generate a segment profile with a list of attributes and consumer behavior associated with a percentage of consumers in the consumer segment, and to provide a graphical view of the segment profile to a display in a client device, wherein the graphical view includes an indicator of the percentage of consumers in a consumer universe associated with the list of attributes and consumer behavior.

16 . A non-transitory, computer readable medium storing instructions which, when executed by a processor, cause a computer to execute a method, the method comprising:

receiving, in a server, a raw data from multiple consumer devices;

refining the raw data to capture a data pattern;

predicting a consumer behavior based on the data pattern, the consumer behavior defining one or more attributes;

identifying a consumer segment based on the consumer behavior and a sharing of a one or more attributes among multiple consumers in the consumer segment;

selecting at least one of an advertising message or a promotional offer to one or more consumers in the consumer segment to include in a payload content;

identifying a media channel to deliver the payload content to one or more consumer devices; and

providing the consumer segment to a display in a client device, upon request.

17 . The non-transitory, computer readable medium of claim 16 wherein, in the method, identifying the media channel comprises selecting one of an in-store printer, a mobile video, a desktop display, or a third party advertisement, based on a type of the one or more consumer devices and a current location of the consumers.

18 . The non-transitory, computer readable medium of claim 16 , wherein the method further comprises receiving, in the server, from a client device, a pre-selected universe of consumers and an impact goal for the payload content, wherein the pre-selected universe of consumers includes the consumer segment and is based on a product or brand identified in the payload content, and the impact goal comprises a desired metric associating the consumer segment with the product or brand identified in the payload content.

19 . The non-transitory, computer readable medium of claim 16 , wherein the method further comprises determining a time duration of the promotional offer, promotion or recommendation in the payload content based on the one or more attributes of the consumers in the consumer segment.

20 . The non-transitory, computer readable medium of claim 16 , wherein the method further comprises selecting a list of products or brands to be included in the payload content based on the one or more attributes of the consumers in the consumer segment.