REAL-TIME UPDATES TO DIGITAL MARKETING FORECAST MODELS

• US 20140379460A1
• Filed: 06/24/2013
• Published: 12/25/2014
• Est. Priority Date: 06/24/2013
• Status: Active Grant

• Current Assignee: Adobe Inc.
• Original Assignee: Adobe Systems Incorporated (Adobe Inc.)
• Inventors: Schein, Andrew I.
• Granted Patent: US 10,181,130 B2
• Time in Patent Office: Days
• US Class Current: 705/14.43
• CPC Class Codes: G06Q 30/0244 Optimization

Abstract

Techniques are disclosed for automatically creating or updating predictive models, including digital marketing forecast models. A predictive model is updated in real-time or near real-time using a stochastic gradient descent optimization method based on one or more predictive values associated with an advertising impression that is won in an online advertising auction. Each predictive value, which is obtained from the predictive model, is encoded as an argument in a uniform resource locator (URL) corresponding to the ad impression being auctioned. If and when the ad impression is won, the predictive value(s) and other information can be tracked and immediately available for updating the model using information encoded in the URL.

Claims

1. A computer-implemented method of digital marketing forecasting comprising:

receiving, subsequent to winning a bid to buy an advertising impression via an online auction, a uniform resource locator (URL) including a forecast value obtained from a predictive model associated with the advertising impression;

computing a prediction error representing a difference between an actual value associated with the advertising impression and the forecast value; and

updating the predictive model based at least in part on the prediction error using a stochastic gradient decent optimization method.

2. The method of claim 1, further comprising:

receiving a request to bid on the advertising impression via the online auction;

computing the forecast value based at least in part on the predictive model; and

encoding the forecast value into the URL.

3. The method of claim 1, wherein the predictive model includes a plurality of predictive dimensions, and wherein the method further comprises updating the predictive model for each of the plurality of predictive dimensions independently.

4. The method of claim 1, wherein the actual value includes actual cost per advertising impression, and wherein the forecast value includes predicted cost per advertising impression.

5. The method of claim 1, wherein the actual value includes actual revenue per advertising impression, and wherein the forecast value includes predicted revenue per advertising impression.

6. The method of claim 1, further comprising computing the forecast value based at least in part on the updated predictive model.

7. A computer-implemented method of digital marketing forecasting comprising:

receiving a request to bid on an advertising impression via an online auction;

computing a forecast value associated with the advertising impression based at least in part on a predictive model; and

encoding the forecast value into a uniform resource locator (URL).

8. The method of claim 7, further comprising:

receiving, subsequent to winning a bid to buy the advertising impression via the online auction, the URL including the forecast value associated with the advertising impression;

computing a prediction error representing a difference between an actual value associated with the advertising impression and the forecast value; and

updating the predictive model based at least in part on the prediction error using a stochastic gradient decent optimization method.

9. The method of claim 8, wherein the actual value includes actual cost per advertising impression, and wherein the forecast value includes predicted cost per advertising impression.

10. The method of claim 8, wherein the actual value includes actual revenue per advertising impression, and wherein the forecast value includes predicted revenue per advertising impression.

11. The method of claim 8, further comprising computing the forecast value based at least in part on the updated predictive model.

12. A digital marketing forecasting system, comprising:

a storage; and

a processor operatively coupled to the storage and configured to;

receive, subsequent to winning a bid to buy an advertising impression via an online auction, a uniform resource locator (URL) including a forecast value obtained from a predictive model stored in the storage and associated with the advertising impression;

compute a prediction error representing a difference between an actual value associated with the advertising impression and the forecast value; and

update the predictive model based at least in part on the prediction error using a stochastic gradient decent optimization method.

13. The system of claim 12, wherein the processor is further configured to:

receive a request to bid on the advertising impression via the online auction;

compute the forecast value based at least in part on the predictive model; and

encode the forecast value into the URL.

14. The system of claim 12, wherein the predictive model includes a plurality of predictive dimensions, and wherein the processor is further configured to update the predictive model for each of the plurality of predictive dimensions independently.

15. The system of claim 12, wherein the actual value includes actual cost per advertising impression, and wherein the forecast value includes predicted cost per advertising impression.

16. The system of claim 12, wherein the actual value includes actual revenue per advertising impression, and wherein the forecast value includes predicted revenue per advertising impression.

17. The system of claim 12, wherein the processor is further configured to compute the forecast value based at least in part on the updated predictive model.

18. A non-transient computer-readable medium having instructions encoded thereon that when executed by a processor cause the processor to:

receive, subsequent to winning a bid to buy an advertising impression via an online auction, a uniform resource locator (URL) including a forecast value obtained from a predictive model stored in the storage and associated with the advertising impression;

compute a prediction error representing a difference between an actual value associated with the advertising impression and the forecast value; and

update the predictive model based at least in part on the prediction error using a stochastic gradient decent optimization method.

19. The computer-readable medium of claim 18, further comprising instructions that when executed by the processor cause the processor to:

receive a request to bid on the advertising impression via the online auction;

compute the forecast value based at least in part on the predictive model; and

encode the forecast value into the URL.

20. The computer-readable medium of claim 18, wherein the predictive model includes a plurality of predictive dimensions, and wherein the computer-readable medium further comprises instructions that when executed by the processor cause the processor to update the predictive model for each of the plurality of predictive dimensions independently.