System and method for digital advertising campaign optimization
A technique for dynamically adjusting a digital advertising campaign during an active campaign flight is discussed. Using feedback from a digital survey over an exposed audience of user populations, brand lift may be calculated on a per ad creative and/or per site basis. User characteristics derived from content consumption patterns may be used to optimize ongoing campaigns and formulate target audiences and target creative formats for new campaigns.
1. A computing device-implemented method for deriving user characteristics via content consumption patterns, the computing device including at least one processor, the method comprising:
initiating a digital advertising campaign for an advertiser, the digital advertising campaign displaying ad creatives on one or more digital properties during an advertising campaign flight;
identifying a training set of data of content consumption patterns of digital data by known users;
providing the training set of data to a machine learning algorithm to train the machine learning algorithm;
acquiring a set of data of a content consumption pattern of digital data by an unknown user using a site-specific Ad Tag, the site-specific Ad Tag including custom configurations for ad placements on a specific digital property and further configured to;
request data from one or more external sources, and
dynamically form a request for an ad creative for an ad placement using data received from the one or more external sources;
providing the set of data of the content consumption pattern of digital data by the unknown user to the trained machine learning algorithm;
receiving user characteristics derived from the set of data of the content consumption pattern of digital data by the unknown user from the trained machine learning algorithm;
providing a digital survey during the campaign flight of the digital advertising campaign to a plurality of individuals that have viewed at least two ad creatives being displayed in the digital advertising campaign, answers of the individuals to the digital survey forming a plurality of survey results;
providing the plurality of survey results to the trained machine learning algorithm; and
adjusting the advertising campaign flight based at least in part on the derived user characteristics and the plurality of survey results.
2. The method of claim 1 wherein the advertising campaign flight is adjusted during the flight based on the derived characteristics.
3. The method of claim 1 wherein the advertising campaign flight is adjusted after the campaign flight based on the derived characteristics.
4. The method of claim 1 wherein contextualization information and website visit metrics are dynamically retrieved by the site-specific Ad Tag.
5. The method of claim 1 wherein data retrieved by the site-specific Ad Tag includes one or more of data regarding an amount of user dwell time on a web page, data regarding an amount of time a user spent viewing an ad creative, data regarding user interaction metrics with an ad creative on the web page, data regarding user location, phone or carrier type, data regarding scroll depth and/or scroll speed, viewability data, likelihood to engage data, data regarding user-defined ad preferences or data regarding user interaction with previously seen ad creatives.
6. The method of claim 1 , wherein the training set of data of content consumption patterns includes registration data acquired from ad publishers.
7. The method of claim 1 , further comprising:
adjusting the advertising campaign flight by increasing an allocation of ad creatives in the ongoing campaign flight that exhibit a pre-determined level of result.
8. The method of claim 1 , further comprising:
adjusting the advertising campaign flight by decreasing an allocation of ad creatives in the ongoing campaign flight that exhibit a pre-determined level of result.
9. A non-transitory medium holding processor-executable instructions for deriving user characteristics via content consumption patterns, the instructions when executed causing at least one computing device equipped with a processor to:
initiate a digital advertising campaign for an advertiser, the digital advertising campaign displaying ad creatives on one or more digital properties during an advertising campaign flight;
identify a training set of data of content consumption patterns of digital data by known users;
provide the training set of data to a machine learning algorithm to train the machine learning algorithm;
acquire a set of data of a content consumption pattern of digital data by an unknown user using a site-specific Ad Tag, the site-specific Ad Tag including custom configurations for ad placements on a specific digital property and further configured to:
request data from one or more external sources, and
dynamically form a request for an ad creative for an ad placement using data received from the one or more external sources;
provide the set of data of the content consumption pattern of digital data by the unknown user to the trained machine learning algorithm;
receive user characteristics derived from the set of data of the content consumption pattern of digital data by the unknown user from the trained machine learning algorithm;
provide a digital survey during the campaign flight of the digital advertising campaign to a plurality of individuals that have viewed at least two ad creatives being displayed in the digital advertising campaign, answers of the individuals to the digital survey forming a plurality of survey results;
provide the plurality of survey results to the trained machine learning algorithm; and
adjust the advertising campaign flight based at least in part on the derived user characteristics and the plurality of survey results.
10. The medium of claim 9 wherein the advertising campaign flight is adjusted during the flight based on the derived characteristics.
11. The medium of claim 9 wherein the advertising campaign flight is adjusted after the campaign flight based on the derived characteristics.
12. The medium of claim 9 wherein contextualization information and website visit metrics are dynamically retrieved by the site-specific Ad Tag.
13. The medium of claim 9 wherein data retrieved by the site-specific Ad Tag includes one or more of data regarding an amount of user dwell time on a web page, data regarding an amount of time a user spent viewing an ad creative, data regarding user interaction metrics with an ad creative on the web page, data regarding user location, phone or carrier type, data regarding scroll depth and/or scroll speed, viewability data, likelihood to engage data, data regarding user-defined ad preferences or data regarding user interaction with previously seen ad creatives.
14. The medium of claim 9 , wherein the training set of data of content consumption patterns includes registration data acquired from ad publishers.
15. The medium of claim 9 , wherein the instructions when executed further cause the computing device to:
adjust the advertising campaign flight by increasing an allocation of ad creatives in the ongoing campaign flight that exhibit a pre-determined level of result.
16. The medium of claim 9 , wherein the instructions when executed further cause the computing device to:
adjust the advertising campaign flight by decreasing an allocation of ad creatives in the ongoing campaign flight that exhibit a pre-determined level of result.