SYSTEMS AND METHODS FOR SELECTING AN AD CAMPAIGN AMONG ADVERTISING CAMPAIGNS HAVING MULTIPLE BID STRATEGIES

• US 20160292722A1
• Filed: 04/02/2015
• Published: 10/06/2016
• Est. Priority Date: 04/02/2015
• Status: Active Grant

• Current Assignee: Vungle, Inc.
• Original Assignee: Vungle, Inc.
• Inventors: Guerin, Hammond, Myers, Raymond, Kang, Daniel
• Granted Patent: US 9,996,853 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 30/0242 Determining effectiveness of advertisements
• CPC Class Codes: G06Q 30/0243 Comparative campaigns
• CPC Class Codes: G06Q 30/0244 Optimization
• CPC Class Codes: G06Q 30/0247 Calculate past, present or future revenues
• CPC Class Codes: G06Q 30/08 Auctions
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34
• CPC Class Codes: H04L 67/55 Push-based network services

Abstract

Methods and systems are described for selecting an engaging ad campaign among advertising campaigns having different types of bid strategies. In one embodiment, an advertising system designed for selecting relevant and engaging ad campaigns for delivering to a device of a user includes an adaptive decision unit having filter logic for filtering eligible ad campaigns, a storage medium to store instructions of the system, and processing logic coupled to the storage medium. The processing logic is configured to execute the instructions of the system to receive and process an ad request from the device upon initiation of a software application on the device, filter eligible ad campaigns, convert each bid strategy of the filtered ad campaigns into an effective cost-per-mille (CPM) strategy, compare effective CPM strategies for the filtered ad campaigns, and select an ad campaign based on the comparison of the effective CPM strategies.

Claims

1. A system for selecting a relevant ad campaign for delivering to a wireless device of a user, comprising:

an ad server that includes an adaptive decision unit having filter logic for filtering eligible ad campaigns, a storage medium to store instructions, and a processing system coupled to the storage medium and the adaptive decision unit, the processing system is configured to execute the instructions to receive and process an ad request from a wireless device upon initiation of a non-web browser software application on the wireless device, filter eligible ad campaigns with the filter logic, compare effective cost-per-mille CPM strategies with predicted conversion rates for the filtered ad campaigns, and select an ad campaign based on the comparison of the effective CPM strategies and the predicted conversion rates; and

at least one prediction server coupled to the ad server, the at least one prediction server to receive the filtered eligible ad campaigns from the ad server, to assign weighted values for different parameters of a plurality of bid strategies of the filtered ad campaigns including a first weighted value if at least one other software application is installed on the wireless device, and a second weighted value if the user has viewed at least one video ad on the wireless device, to determine a predicted conversion rate for each filtered ad campaigns based on the weighted values of the different parameters, and to convert each bid strategy including a cost per install (CPI) into an effective CPM strategy.

2. The system of claim 1, wherein the processing system is further configured to execute the instructions of the system to:

generate an ad response that includes the selected ad campaign; and

send the ad response to the wireless device in real time just prior to an ad play event for a video ad for display on the wireless device in-application (in-app) of the non-web browser software application during the ad play event.

3. The system of claim 1, wherein the filter logic of the adaptive decision unit includes geo-location data and cost budgets of publishers for different bid strategies including CPM, cost-per-click (CPC), cost-per-install (CPI), cost-per-action (CPA), and cost-per-completed-view (CPCV).

4. The system of claim 1, further comprising:

prediction servers for storing a plurality of machine learning models including a machine learning model that assigns a weighted value for different parameters of a plurality of bid strategies of the filtered ad campaigns, determines a predicted conversion rate for a plurality of filtered ad campaigns based on the different parameters, and converts each bid strategy into an effective CPM strategy.

5. The system of claim 4, wherein the different parameters include user characteristics for the user including user engagement rate during ad play events and behavior data including previous app installs and a time of last app install for the user.

6. The system of claim 5, wherein the different parameters further include device characteristics of the wireless device including make and model of the user'"'"'s wireless device, network connection for the user'"'"'s wireless device, volume, screen size, and display orientation of the user'"'"'s wireless device.

17. The system of claim 1, wherein the ad server comprises a special purpose dedicated computer to perform advertising and publisher functions.

7. A machine-accessible non-transitory medium containing executable computer program instructions which when executed by an ad system cause the ad system to perform a method for delivering an ad campaign to a wireless device of a user, the method comprising:

receiving, with a processing system of the ad system, an ad request from a wireless device upon initiation of a non-web browser software application on the wireless device;

processing the ad request;

filtering, with filter logic of the ad system, eligible ad campaigns and associated bid strategies to generate filtered ad campaigns;

receiving with at least one prediction server filtered eligible ad campaigns from the ad server, and assigning weighted values for different parameters of a plurality of bid strategies of the filtered ad campaigns including a first weighted value if at least one other software application is installed on the wireless device, and a second weighted value if the user has viewed at least one video ad on the wireless device;

determining a predicted conversion rate for each filtered ad campaign based on the weighted values of the different parameters;

converting each bid strategy of the filtered ad campaigns into an effective cost-per-mille (CPM) strategy;

comparing effective CPM strategies with the predicted conversion rates for the filtered ad campaigns; and

selecting an ad campaign from the filtered ad campaigns based on the comparison of the effective CPM strategies.

8. The machine-accessible non-transitory medium of claim 7, the method further comprising:

generating an ad response that includes the selected ad campaign; and

sending the ad response in real time just prior to an ad play event for a video ad to the wireless device for display on the wireless device during the ad play event.

9. The machine-accessible non-transitory medium of claim 7, wherein filtering eligible ad campaigns and associated bid strategies to generate filtered ad campaigns includes determining geo-location data and cost budgets of publishers for different bid strategies of the eligible ad campaigns including CPM, cost-per-click (CPC), cost-per-install (CPI), cost-per-action (CPA), and cost-per-completed-view (CPCV).

10. The machine-accessible non-transitory medium of claim 7, the method further comprising:

storing with at least one prediction server a plurality of machine learning models;

assigning with a machine learning model a weighted value for different parameters of a plurality of bid strategies of the filtered ad campaigns;

determining a predicted conversion rate for the plurality of filtered ad campaigns based on the different parameters; and

converting each bid strategy into an effective CPM strategy.

11. The machine-accessible non-transitory medium of claim 10, wherein the different parameters include user characteristics for the user including user engagement rate during ad play events and behavior data including previous app installs and a time of last app install for the user.

12. The machine-accessible non-transitory medium of claim 11, wherein the different parameters further include device characteristics of the wireless device including make and model of the user'"'"'s wireless device, network connection for the user'"'"'s wireless device, volume, screen size, and display orientation of the user'"'"'s wireless device.

13. A device for displaying an ad campaign, comprising:

a touch display device for receiving user input of a user and displaying output;

a storage medium to store instructions of advertising services software; and

a processing unit coupled to the storage medium and the touch display device, the processing unit is configured to execute the instructions of the advertising services software to generate an ad request in response to initiation of a non-web browser software application on the device, send the ad request to an ad system, and receive an ad response in real time prior to an ad play event for a video ad with the ad response including a selected ad campaign to be displayed on the touch display device in-application (in-app) during the ad play event of the non-web browser software application, wherein the ad request includes different types of information including publisher settings for a publisher of the initiated non-web browser software application, an application id to identify the selected non-web browser software application, and user characteristics including user engagement rate during ad play events, wherein the device comprises a mobile device or a tablet device, wherein the selected ad campaign is selected based on device characteristics including a network connection for the device, operating system version, volume, screen size and orientation, whether at least one other software application is installed on the device, and whether the user has viewed at least one video ad on the device.

14. The device of claim 13, wherein at least one of the device or an ad system filter eligible ad campaigns, convert each bid strategy of the filtered ad campaigns into an effective cost-per-mille (CPM) strategy, determine predicted conversion rates for the filtered ad campaigns based on whether at least one other software application is installed on the device, and whether the user has viewed at least one video ad on the device, and compare effective CPM strategies with the predicted conversion rates for the filtered ad campaigns.

15. The device of claim 14, wherein at least one of the device or the ad system select an ad campaign based on the comparison of the effective CPM strategies and the predicted conversion rates.

16. The device of claim 13, wherein the ad campaign is selected based on the different types of information of the ad request including publisher settings for a publisher of the initiated software application, an application id to identify the selected software application, user characteristics including user engagement rate during ad play events, and device characteristics including a device id and geographical data.