US 10,650,403 B2Granted

Distributing online ads by targeting online ad requests

AssigneeAdobe Inc.

InventorsDeepak Pai, Anil Kamath

Life of this patent

Estimating the term…

Loading patent…

Abstract

Techniques for distributing online ads by targeting online ad requests using test data to predict performance. The techniques can target ad requests in automated online advertising systems in which ad requests are generated by an ad exchange server and bids are placed by marketer devices in real time. The techniques aggregate bid units and compare bid unit characteristics to select bid units to target in ways that address data sparsity, variance, and volume issues. Data sparsity issues are addressed by aggregating bid units to avoid using bid units having insufficient data. Data variance issues are addressed by computing stability metrics for bid units that enable discounting the effect of outliers. Data volume and processing efficiency issues are addressed by grouping similar bid units based on similar metrics (e.g., normalized ROI) and/or similar stability scores, and then ranking the bid units and selecting the top ranked bid units to target.

Claims19

Dependent claims

A method for distributing online ads by targeting online ad requests in automated online advertising systems in which online ad requests are generated by an ad exchange server and bids are placed for marketer devices in real time as web pages are loaded for viewing, the method comprising:

tracking, by a targeting server, performance metric data for online test ads provided in the web pages in response to the online ad requests generated by the ad exchange server, the performance metric data providing performance metrics for bid units, the bid units corresponding to combinations of values for online ad request dimensions;
selecting sparse bid units based on identifying those of the bid units having sparse performance metric data;
aggregating, by the targeting server, the sparse bid units with one another to obtain an aggregated bid unit;
selecting, by the targeting server, one or more of the bid units, including the aggregated bid unit, as targets on which to spend a budget of a campaign to place online ads into the web pages as the web pages are loaded for viewing and in response to the online ad requests, the one or more bid units selected as the targets based on the performance metric for the bid units;
selecting, by the targeting server and based on the one or more bid units selected as the targets, a subset of the online ads corresponding to the online test ads that are responsive to the online ad requests which have values for the online ad request dimensions corresponding to the one or more bid units selected as targets;
distributing, by the targeting server, the subset of the online ads to first ad recipients within the web pages being loaded for viewing by the first ad recipients;
selecting, by the targeting server, bid units having performance metric data with high variance as additional targets;
distributing, by the targeting server, the subset of the online ads based on the bid units selected as the additional targets;
tracking, by the targeting server, additional performance metric data for the subset of the online ads distributed based on the bid units selected as the additional targets;
selecting, by the targeting server, bid units as updated targets based on the additional performance metric data for the bid units selected as the additional targets;
tracking, by the targeting server, subsequent performance metric data to iteratively refine the bid units selected as the updated targets based on the additional performance metric data;
selecting, by the targeting server and based on the bid units selected as the updated targets, an updated subset of the online ads which have values for the online ad request dimensions corresponding to the bid units selected as the updated targets;
and
distributing, by the targeting server, the updated subset of the online ads to second ad recipients within the web pages being loaded for viewing by the second ad recipients.

11 more elements

A method for distributing online ads by targeting online ad requests in automated online advertising systems in which ad requests are generated by an ad exchange server and bids are placed by marketer devices in real time as web pages are loaded for viewing, the method comprising:

tracking, by a targeting server, performance metric data for online test ads provided in the web pages in response to the online ad requests generated by the ad exchange server, the performance metric data providing performance metrics for bid units, the bid units corresponding to combinations of values for online ad request dimensions;
determining, by the targeting server, stability scores for the bid units based on stability of the performance metric data for the respective bid units;
selecting, by the targeting server, bid units with similar stability scores;
aggregating, by the targeting server, the selected bid units with one another to obtain an aggregated bid unit;
selecting, by the targeting server, one or more of the bid units, including the aggregated bid unit, as targets based on the performance metric and stability scores for the bid units, the bid units selected as targets to spend a budget of a campaign to place online ads into the web pages as the web pages are loaded for viewing and in response to online ad requests;
selecting, by the targeting server and based on the one or more bid units selected as the targets, a subset of the online ads corresponding to the online test ads that are responsive to the online ad requests which have values for the online ad request dimensions corresponding to the one or more bid units selected as targets;
distributing, by the targeting server, the subset of the online ads to first ad recipients within the web pages being loaded for viewing by the first ad recipients;
selecting, by the targeting server, bid units having performance metric data with high variance as additional targets;
distributing, by the targeting server, the subset of the online ads based on the bid units selected as the additional targets;
tracking, by the targeting server, additional performance metric data for the subset of the online ads distributed based on the bid units selected as the additional targets;
selecting, by the targeting server, bid units as updated targets based on the additional performance metric data for the bid units selected as the additional targets;
tracking, by the targeting server, subsequent performance metric data to iteratively refine the bid units selected as the updated targets based on the additional performance metric data;
selecting, by the targeting server and based on the bid units selected as the updated targets, an updated subset of the online ads which have values for the online ad request dimensions corresponding to the bid units selected as the updated targets;
and
distributing, by the targeting server, the updated subset of the online ads to second ad recipients within the web pages being loaded for viewing by the second ad recipients.

12 more elements

tracking, by a targeting server, performance metric data for online test ads provided in the web pages in response to the online ad requests generated by the ad exchange server, the performance metric data providing performance metrics for bid units, the bid units corresponding to combinations of values for online ad request dimensions;
aggregating, by the targeting server, bid units based on similarity of the performance metrics of the bid units to one another, to obtain an aggregated bid unit;
ranking, by the targeting server, the bid units, including the aggregated bid unit;
selecting, by the targeting server, top ranked bid units as targets, wherein a number of top ranked bid units to select as targets is based on predicted spend amounts for the bid units and a budget of a campaign to place online ads into the web pages as the web pages are loaded for viewing and in response to the online ad requests;
selecting, by the targeting server and based on the top ranked bid units selected as the targets, a subset of the online ads corresponding to the online test ads that are responsive to the online ad requests which have values for the online ad request dimensions corresponding to the top ranked bid units selected as the targets;
distributing, by the targeting server, the subset of the online ads to first ad recipients within the web pages being loaded for viewing by the first ad recipients;
selecting, by the targeting server, bid units having performance metric data with high variance as additional targets;
distributing, by the targeting server, the subset of the online ads based on the bid units selected as the additional targets;
tracking, by the targeting server, additional performance metric data for the subset of the online ads distributed based on the bid units selected as the additional targets;
selecting, by the targeting server, bid units as updated targets based on the additional performance metric data for the bid units selected as the additional targets;
tracking, by the targeting server, subsequent performance metric data to iteratively refine the bid units selected as the updated targets based on the additional performance metric data;
selecting, by the targeting server and based on the bid units selected as the updated targets, an updated subset of the online ads which have values for the online ad request dimensions corresponding to the bid units selected as the updated targets;
and
distributing, by the targeting server, the updated subset of the online ads to second ad recipients within the web pages being loaded for viewing by the second ad recipients.

11 more elements

Dependencies are inferred from the claim text.