Granted Patent B1

US 10,417,658 · App. 14/084,568 · Granted Sep 17, 2019

Inventors: Kiril Tsemekhman (San Francisco, CA); Vadim Tsemekhman (Seattle, WA); Arun Ahuja (Stamford, CT)

Assignee: Integral Ad Science, Inc.

G06Q30/0246G06Q30/0242

Abstract

Claims (87)

1. A method for placing online advertisements, the method comprising:

receiving, using a hardware processor, an advertising budget for a particular portion of an online advertising campaign;

causing, using the hardware processor, monitoring code to be loaded in association with advertisements presented as part of the online advertising campaign, wherein the monitoring code is configured to determine advertisement viewability information indicative of a probability that an advertisement was viewable to a consumer based at least in part on a comparison of an amount of time that the advertisement of a particular size on a particular web page was within a viewport of a web browser with past measurements of in-view time of advertisements of the particular size on the particular webpage;

(a) receiving, using the hardware processor, conversion information corresponding to a plurality of consumers each associated with one of a plurality of consumer devices that each loaded an advertisement as part of a web page loaded by the web browser executed by that consumer device based on instructions to load the advertisement received using a first online advertising channel, wherein the advertisement was placed using the first online advertising channel as part of the online advertising campaign according to an initial allocation amongst a plurality of online advertising channels including the first online advertising channel;

(b) receiving advertisement viewability information indicative of a probability that each of the plurality of consumers viewed the advertisement, wherein the advertisement viewability information was sent by the monitoring code loaded in association with the advertisement loaded as part of the web page by the web browser executed by the consumer device corresponding to that consumer;

(c) determining that a subset of the plurality of consumers did not view the advertisement based on the probability that the advertisement was viewed;

(d) placing the plurality of consumers into a control group and a test group based on the advertisement viewability information corresponding to each of the plurality of consumers, wherein the subset of the plurality of consumers are included in the control group and all consumers associated with a consumer device that loaded the advertisement based on instructions received using the first online advertising channel are included in the test group;

(e) calculating a causal conversion metric for the first online advertising channel based on a comparison of the conversion information corresponding to consumers of the control group and conversion information corresponding to consumers of the test group;

(f) repeating (a) through (e) for each of the plurality of online advertising channels other than the first online advertising channel;

allocating the budget among the plurality of online advertising channels based on the causal conversion metric corresponding to the first online advertising channel and the causal conversion metrics corresponding to the plurality of online advertising channels other than the first online advertising channel; and

transmitting instructions to at least one remote computing device indicating in what proportion advertisements for the campaign are to be placed using each of the plurality of online advertising channels based at least in part on the causal conversion metric and the budget, wherein the proportion is based on the allocation of the budget.

2. The method of claim 1 , wherein calculating a causal conversion metric further comprises:

calculating a first conversion rate for the control group based on the conversion information corresponding to consumers included in the control group;

calculating a second conversion rate for the test group based on the conversion information corresponding to consumers included in the test group;

comparing the first conversion rate to the second conversion rate; and

calculating a causal conversion rate based on the comparison.

3. The method of claim 1 , wherein calculating a causal conversion metric further comprises:

receiving advertisement cost information corresponding to the cost of presenting each of the plurality of users with the advertisement;

calculating a first return on investment for the control group based on the conversion information corresponding to consumers included in the control group and advertising cost information corresponding to consumers included in the control group;

calculating a second return on investment for the test group based on the conversion information corresponding to consumers included in the test group and advertising cost information corresponding to consumers included in the test group;

comparing the first return on investment to the second return on investment; and

calculating a causal return on investment based on the comparison.

4. The method of claim 1 , wherein allocating the budget further comprises determining whether to place more advertisements on the first online advertising channel or one of the plurality of online advertising channels other than the first online advertising channel based on a comparison of the causal conversion metric corresponding to the first online advertising channel and the causal conversion metrics corresponding to the plurality of advertising channels other than the first online advertising channel.

5. The method of claim 1 , further comprising:

categorizing the plurality of consumers into a second subset of consumers based on a contextual category of a web site with which the advertisement was presented;

calculating a third causal conversion metric for the second subset of consumers;

comparing the third causal conversion metric to the causal conversion metric; and

determining whether to place one or more advertisements on web sites in the contextual category using the advertising channel based on the comparison.

6. The method of claim 1 , wherein determining that the subset of the plurality of consumers did not view the advertisement based on the probability that the advertisement was viewed comprises determining that the probability that the advertisement was viewed is less than a threshold probability.

7. A system for placing advertisements, the system comprising:

a hardware processor that is configured to:

receive an advertising budget for a particular portion of an online advertising campaign;

cause monitoring code to be loaded in association with advertisements presented as part of the online advertising campaign, wherein the monitoring code is configured to determine advertisement viewability information indicative of a probability that an advertisement was viewable to a consumer based at least in part on a comparison of an amount of time that the advertisement of a particular size on a particular web page was within a viewport of a web browser with past measurements of in-view time of advertisements of the particular size on the particular webpage;

(a) receive conversion information corresponding to a plurality of consumers each associated with one of a plurality of consumer devices that each loaded an advertisement as part of a web page loaded by the web browser executed by that consumer device based on instructions to load the advertisement received using an advertising channel, wherein the advertisement was placed using the first online advertising channel as part of the online advertising campaign according to an initial allocation amongst a plurality of online advertising channels including the first online advertising channel;

(b) receive advertisement viewability information indicative of a probability that each of the plurality of consumers viewed the advertisement, wherein the advertisement viewability information was sent by the monitoring code loaded in association with the advertisement loaded as part of the web page by the web browser executed by the consumer device corresponding to that consumer;

(c) determine that a subset of the plurality of consumers did not view the advertisement based on the probability that the advertisement has been viewed;

(d) place the plurality of consumers into a control group and a test group based on the advertisement viewability information corresponding to each of the plurality of consumers, wherein the subset of the plurality of consumers are included in the control group and all consumers associated with a consumer device that loaded the advertisement based on instructions received using the first online advertising channel are included in the test group;

(e) calculate a causal conversion metric for the first online advertising channel based on a comparison of the conversion information corresponding to consumers of the control group and conversion information corresponding to consumers of the test group;

(f) repeat (a) through (e) for each of the plurality of online advertising channels other than the first online advertising channel;

allocate the budget among the plurality of online advertising channels based on the causal conversion metric corresponding to the first online advertising channel and the causal conversion metrics corresponding to the plurality of online advertising channels other than the first online advertising channel; and

transmit instructions to at least one remote computing device indicating in what proportion advertisements for the campaign are to be placed using each of the plurality of online advertising channels based at least in part on the causal conversion metric and the budget, wherein the proportion is based on the allocation of the budget.

8. The system of claim 7 , wherein the hardware processor is further configured to:

calculate a first conversion rate for the control group based on the conversion information corresponding to consumers included in the control group;

calculate a second conversion rate for the test group based on the conversion information corresponding to consumers included in the test group;

compare the first conversion rate to the second conversion rate; and

calculate a causal conversion rate based on the comparison.

9. The system of claim 7 , wherein the hardware processor is further configured to:

receive advertisement cost information corresponding to the cost of presenting each of the plurality of users with the advertisement;

calculate a first return on investment for the control group based on the conversion information corresponding to consumers included in the control group and advertising cost information corresponding to consumers included in the control group;

calculate a second return on investment for the test group based on the conversion information corresponding to consumers included in the test group and advertising cost information corresponding to consumers included in the test group;

compare the first return on investment to the second return on investment; and

calculate a causal return on investment based on the comparison.

10. The system of claim 7 , wherein the hardware processor is further configured to determine whether to place more advertisements on the first online advertising channel or one of the plurality of online advertising channels other than the first online advertising channel based on a comparison of the causal conversion metric corresponding to the first online advertising channel and the causal conversion metrics corresponding to the plurality of advertising channels other than the first online advertising channel.

11. The system of claim 7 , wherein the hardware processor is further configured to:

categorize the plurality of consumers into a second subset of consumers based on a contextual category of a web site with which the advertisement was presented;

calculate a third causal conversion metric for the second subset of consumers;

compare the third causal conversion metric to the causal conversion metric; and

determine whether to place one or more advertisements on web sites in the contextual category using the advertising channel based on the comparison.

12. The system of claim 7 , wherein the hardware processor is further configured to determine that the probability that the advertisement was viewed is less than a threshold probability.

13. A non-transitory computer-readable medium containing computer-executable instructions that, when executed by a processor, cause the processor to perform a method for placing advertisements, the method comprising:

receiving an advertising budget for a particular portion of an online advertising campaign;

causing monitoring code to be loaded in association with advertisements presented as part of the online advertising campaign, wherein the monitoring code is configured to determine advertisement viewability information indicative of a probability that an advertisement was viewable to a consumer based at least in part on a comparison of an amount of time that the advertisement of a particular size on a particular web page was within a viewport of a web browser with past measurements of in-view time of advertisements of the particular size on the particular webpage;

(a) receiving conversion information corresponding to a plurality of consumers each associated with one of a plurality of consumer devices that each loaded an advertisement as part of a web page loaded by the web browser executed by that consumer device based on instructions to load the advertisement received using a first online advertising channel, wherein the advertisement was placed using the first online advertising channel as part of the online advertising campaign according to an initial allocation amongst a plurality of online advertising channels including the first online advertising channel;

(b) receiving advertisement viewability information indicative of a probability that each of the plurality of consumers viewed the advertisement, wherein the advertisement viewability information was sent by the monitoring code loaded in association with the advertisement loaded as part of the web page by the web browser executed by the consumer device corresponding to that consumer;

(c) determining that a subset of the plurality of consumers did not view the advertisement based on the probability that the advertisement was viewed;

(d) placing the plurality of consumers into a control group and a test group based on the advertisement viewability information corresponding to each of the plurality of consumers, wherein the subset of the plurality of consumers are included in the control group and all consumers associated with a consumer device that loaded the advertisement based on instructions received using the first online advertising channel are included in the test group;

(e) calculating a causal conversion metric for the first online advertising channel based on a comparison of the conversion information corresponding to consumers of the control group and conversion information corresponding to consumers of the test group;

(f) repeating (a) through (e) for each of the plurality of online advertising channels other than the first online advertising channel;

allocating the budget among the plurality of online advertising channels based on the causal conversion metric corresponding to the first online advertising channel and the causal conversion metrics corresponding to the plurality of online advertising channels other than the first online advertising channel; and

transmitting instructions to at least one remote computing device indicating in what proportion advertisements for the campaign are to be placed using each of the plurality of online advertising channels based at least in part on the causal conversion metric and the budget, wherein the proportion is based on the allocation of the budget.

14. The non-transitory computer-readable medium of claim 13 , wherein calculating a causal conversion metric further comprises:

calculating a first conversion rate for the control group based on the conversion information corresponding to consumers included in the control group;

calculating a second conversion rate for the test group based on the conversion information corresponding to consumers included in the test group;

comparing the first conversion rate to the second conversion rate; and

calculating a causal conversion rate based on the comparison.

15. The non-transitory computer-readable medium of claim 13 , wherein calculating a causal conversion metric further comprises:

receiving advertisement cost information corresponding to the cost of presenting each of the plurality of users with the advertisement;

calculating a first return on investment for the control group based on the conversion information corresponding to consumers included in the control group and advertising cost information corresponding to consumers included in the control group;

calculating a second return on investment for the test group based on the conversion information corresponding to consumers included in the test group and advertising cost information corresponding to consumers included in the test group;

comparing the first return on investment to the second return on investment; and

calculating a causal return on investment based on the comparison.

16. The non-transitory computer-readable medium of claim 13 , wherein allocating the budget further comprises determining whether to place more advertisements on the first online advertising channel or one of the plurality of online advertising channels other than the first online advertising channel based on a comparison of the causal conversion metric corresponding to the first online advertising channel and the causal conversion metrics corresponding to the plurality of advertising channels other than the first online advertising channel.

17. The non-transitory computer-readable medium of claim 13 , wherein the method further comprises:

categorizing the plurality of consumers into a second subset of consumers based on a contextual category of a web site with which the advertisement was presented;

calculating a third causal conversion metric for the second subset of consumers;

comparing the third causal conversion metric to the causal conversion metric; and determining whether to place one or more advertisements on web sites in the contextual category using the advertising channel based on the comparison.

18. The non-transitory computer-readable medium of claim 13 , wherein determining that the subset of the plurality of consumers did not view the advertisement based on the probability that the advertisement was viewed comprises determining that the probability that the advertisement was viewed is less than a threshold probability.

Assignments (9)

Continuity (2)

Cited By (3)

US 12,439,132 US 12,695,960 US 12,711,525