Methods, systems, and devices for counterfactual-based incrementality measurement in digital ad-bidding platform
First Claim
Patent Images
1. A computer-implemented method for adjusting future bid requests based on a causal ad impact of previous bid requests, the method comprising:
- receiving, by a computer system of a demand side platform, a first bid request from an ad exchange, wherein the first bid request is indicative of a first bidding opportunity at the ad exchange;
matching, by the computer system of the demand side platform, ad campaigns of one or more advertisers to the first bid request to identify a list of one or more eligible advertisers to submit a bid to the first bidding opportunity to place one or more advertisements from the ad campaigns of the one or more eligible advertisers on user computing systems of a plurality of users;
executing, by the computer system of the demand side platform, a pre-bid randomization scheme prior to submitting bids, whereby each of a plurality of user identifiers associated with the plurality of users is categorized in a control group or a test group;
logging, by the computer system of the demand side platform, a phantom control impression, the phantom control impression comprising the control group of the plurality of user identifiers, wherein the ad exchange does not transmit the one or more advertisements from the ad campaigns of the one or more eligible advertisers to user computing systems of the control group;
submitting, by the computer system of the demand side platform, one or more bids associated with the ad campaigns of the one or more eligible advertisers for placing the one or more advertisements from the ad campaigns of the one or more eligible advertisers on user computing systems of the test group;
receiving, by the computer system of the demand side platform, results of the one or more submitted bids, wherein the results indicate whether each of the one or more submitted bids was won or lost;
logging, by the computer system of the demand side platform, a test-win impression, the test-win impression comprising a first sub-group of one or more user identifiers of the test group, wherein the ad exchange transmits at least one of the one or more advertisements from the ad campaigns of the one or more eligible advertisers to user computing systems of the first sub-group of one or more user identifiers of the test group;
logging, by the computer system of the demand side platform, a test-lost impression, the test-lost impression comprising a second sub-group of one or more user identifiers of the test group, wherein the ad exchange does not transmit the one or more advertisements from the ad campaigns of the one or more eligible advertisers to user computing systems of the second sub-group of one or more user identifiers of the test group;
accessing, by the computer system of the demand side platform, cookie data associated with the control group of the plurality of user identifiers of the phantom control impression, the first sub-group of the test group of the plurality of user identifiers of the test-win impression, and the second sub-group of the test group of the plurality of user identifiers of the test-lost impression, wherein the cookie data comprises a cookie ID that can be utilized to identify browser actions of associated users;
identifying, by the computer system of the demand side platform, browser actions associated with the control group of the plurality of user identifiers of the phantom control impression, the first sub-group of the test group of the plurality of user identifiers of the test-win impression, and the second sub-group of the test group of the plurality of user identifiers of the test-lost impression by utilizing the accessed cookie data;
determining, by the computer system of the demand side platform, the causal ad impact based at least in part on the identified browser actions associated with the control group of the plurality of user identifiers of the phantom control impression, the first sub-group of the test group of the plurality of user identifiers of the test-win impression, and the second sub-group of the test of the plurality of user identifiers of the test-lost impression; and
generating, by the computer system, a graphical user interface that allows the one or more advertisers to utilize the determined causal ad impact to adjust one or more bids in a second bidding opportunity,wherein the computer system comprises a computer processor and an electronic storage medium.
5 Assignments
0 Petitions
Accused Products
Abstract
A digital ad-buying platform uses counterfactual-based incrementality measurement by implementing randomization and/or a correction for auction win bias to avoid the need to identify counterfactual winner types in the control group. This approach can estimate impact at the individual consumer level. Confidence levels can be determined using Gibbs sampling in the context of causal analysis in the presence of non-compliance.
-
Citations
20 Claims
-
1. A computer-implemented method for adjusting future bid requests based on a causal ad impact of previous bid requests, the method comprising:
-
receiving, by a computer system of a demand side platform, a first bid request from an ad exchange, wherein the first bid request is indicative of a first bidding opportunity at the ad exchange; matching, by the computer system of the demand side platform, ad campaigns of one or more advertisers to the first bid request to identify a list of one or more eligible advertisers to submit a bid to the first bidding opportunity to place one or more advertisements from the ad campaigns of the one or more eligible advertisers on user computing systems of a plurality of users; executing, by the computer system of the demand side platform, a pre-bid randomization scheme prior to submitting bids, whereby each of a plurality of user identifiers associated with the plurality of users is categorized in a control group or a test group; logging, by the computer system of the demand side platform, a phantom control impression, the phantom control impression comprising the control group of the plurality of user identifiers, wherein the ad exchange does not transmit the one or more advertisements from the ad campaigns of the one or more eligible advertisers to user computing systems of the control group; submitting, by the computer system of the demand side platform, one or more bids associated with the ad campaigns of the one or more eligible advertisers for placing the one or more advertisements from the ad campaigns of the one or more eligible advertisers on user computing systems of the test group; receiving, by the computer system of the demand side platform, results of the one or more submitted bids, wherein the results indicate whether each of the one or more submitted bids was won or lost; logging, by the computer system of the demand side platform, a test-win impression, the test-win impression comprising a first sub-group of one or more user identifiers of the test group, wherein the ad exchange transmits at least one of the one or more advertisements from the ad campaigns of the one or more eligible advertisers to user computing systems of the first sub-group of one or more user identifiers of the test group; logging, by the computer system of the demand side platform, a test-lost impression, the test-lost impression comprising a second sub-group of one or more user identifiers of the test group, wherein the ad exchange does not transmit the one or more advertisements from the ad campaigns of the one or more eligible advertisers to user computing systems of the second sub-group of one or more user identifiers of the test group; accessing, by the computer system of the demand side platform, cookie data associated with the control group of the plurality of user identifiers of the phantom control impression, the first sub-group of the test group of the plurality of user identifiers of the test-win impression, and the second sub-group of the test group of the plurality of user identifiers of the test-lost impression, wherein the cookie data comprises a cookie ID that can be utilized to identify browser actions of associated users; identifying, by the computer system of the demand side platform, browser actions associated with the control group of the plurality of user identifiers of the phantom control impression, the first sub-group of the test group of the plurality of user identifiers of the test-win impression, and the second sub-group of the test group of the plurality of user identifiers of the test-lost impression by utilizing the accessed cookie data; determining, by the computer system of the demand side platform, the causal ad impact based at least in part on the identified browser actions associated with the control group of the plurality of user identifiers of the phantom control impression, the first sub-group of the test group of the plurality of user identifiers of the test-win impression, and the second sub-group of the test of the plurality of user identifiers of the test-lost impression; and generating, by the computer system, a graphical user interface that allows the one or more advertisers to utilize the determined causal ad impact to adjust one or more bids in a second bidding opportunity, wherein the computer system comprises a computer processor and an electronic storage medium. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
9. A system for adjusting future bid requests based on a causal ad impact of previous bid requests, wherein the system comprises:
-
one or more computer readable storage devices configured to store a plurality of computer executable instructions; and one or more hardware computer processors in communication with the one or more computer readable storage devices and configured to execute the plurality of computer executable instructions in order to cause the system to; receive a first bid request from an ad exchange, wherein the first bid request is indicative of a first bidding opportunity at the ad exchange; match ad campaigns of one or more advertisers to the first bid request to identify a list of one or more eligible advertisers to submit a bid to the first bidding opportunity to place one or more advertisements from the ad campaigns of the one or more eligible advertisers on user computing systems of a plurality of users; execute a randomization scheme prior to submitting bids, whereby each of a plurality of user identifiers associated with the plurality of users is categorized in a control group or a test group based on a fraction probability; log a phantom control impression, the phantom control impression comprising the control group of the plurality of user identifiers, wherein the ad exchange does not transmit the one or more advertisements from the ad campaigns of the one or more eligible advertisers to user computing systems of the control group; submit one or more bids associated with the ad campaigns of the one or more eligible advertisers for placing the one or more advertisements from the ad campaigns associated with the one or more eligible advertisers on user computing systems of the test group; receive results of the one or more submitted bids, wherein the results indicate whether each of the one or more submitted bids was won or lost; log a test-win impression, the test-win impression comprising a sub-group of one or more user identifiers of the test group, wherein the ad exchange transmits at least one of the one or more advertisements from the ad campaigns of one or more eligible advertisers to user computing systems of the sub-group of one or more user identifiers of the test group; access cookie data associated with the control group of the plurality of user identifiers of the phantom control impression and the sub-group of the test group of the plurality of user identifiers of the test-win impression, wherein the cookie data comprises a cookie ID that can be utilized to identify browser actions of associated users; identify browser actions associated with the control group of the plurality of user identifiers of the phantom control impression and the sub-group of the test group of the plurality of user identifiers of the test-win impression by utilizing the accessed cookie data; determine the causal ad impact based at least in part on the identified browser actions associated with the control group of the plurality of user identifiers of the phantom control impression and the sub-group of the test group of the plurality of user identifiers of the test-win impression; and generate a graphical user interface that allows the one or more advertisers to utilize the determined causal ad impact to adjust one or more bids in a second bidding opportunity. - View Dependent Claims (10, 11, 12, 13)
-
-
14. A computer-implemented method for adjusting future bid requests based on a causal ad impact of previous bid requests, the method comprising:
-
receiving, by a computer system of the demand side platform, a first bid request from an ad exchange, wherein the first bid request is indicative of a first bidding opportunity at the ad exchange; matching, by the computer system of the demand side platform, ad campaigns of one or more advertisers to the first bid request to identify a list of one or more eligible advertisers to submit a bid to the first bidding opportunity to place one or more advertisements from the ad campaigns of the one or more eligible advertisers on user computing systems of a plurality of users; submitting, by the computer system of the demand side platform, one or more bids associated with the ad campaigns of the one or more eligible advertisers; receiving, by the computer system of the demand side platform, results of the one or more submitted bids, wherein the results indicate whether each of the one or more bids was won or lost; executing, by the computer system of the demand side platform, a randomization scheme, whereby each of a plurality of user identifiers associated with the plurality of users is categorized in a control group or a test group; logging, by the computer system of the demand side platform, a test group impression, the test group impression comprising the test group of the plurality of user identifiers, wherein the ad exchange transmits at least one or more advertisements from the ad campaigns of the one or more eligible advertisers to user computing systems of the test group of the plurality of user identifiers; logging, by the computer system of the demand side platform, a control group impression, the control group impression comprising the control group of the plurality of user identifiers, wherein the ad exchange does not transmit the one or more advertisements from the ad campaigns of the one or more eligible advertisers to user computing systems of the control group of the plurality of user identifiers; accessing, by the computer system of the demand side platform, cookie data associated with the test group of the plurality of user identifiers of the test group impression and the control group of the plurality of user identifiers of the control group impression, wherein the cookie data comprises a cookie ID that can be utilized to identify browser actions of associated users; identifying, by the computer system of the demand side platform, browser actions associated with the test group of the plurality of user identifiers of the test group impression and the control group of the plurality of user identifiers of the control group impression by utilizing the accessed cookie data; determining, by the computer system of the demand side platform, the causal ad impact based at least in part on the identified browser actions associated with the test group of the plurality of the user identifiers of the test group impression and the control group of the plurality of user identifiers of the control group impression; and generating, by the computer system, a graphical user interface that allows the one or more advertisers to utilize the determined causal ad impact to adjust one or more bids in a second bidding opportunity, wherein the computer system comprises a computer processor and an electronic storage medium. - View Dependent Claims (15, 16, 17, 18, 19, 20)
-
Specification