IN-APPLICATION ADVERTISEMENT PLACEMENT

1 . A computing system, comprising:

one or more processors; and

memory storing program instructions that when executed by the one or more processors cause the one or more processors to at least:

receive an indication of an application executing on a user device;

determine at least one of a user profile associated with a user executing the application, a user device profile, or a condition representative of a location of the user device;

determine a plurality of candidate advertisement placement positions within the application at which advertisements may be presented in the application to a user of the application;

determine, based at least in part on one or more of the user profile, the user device profile, or the condition:

an expected access duration; and

a quantity of advertisements to be presented during the expected access duration;

determine a predicted engagement value for each of the plurality of candidate advertisement placement positions;

select a sub-plurality of the plurality of candidate advertisement placement positions having highest predicted engagement values, the sub-plurality of candidate advertisement placement positions corresponding to the quantity; and

send a first advertisement for presentation in the application at a first advertisement placement position of the sub-plurality of advertisement placement positions.

2 . The computing system of claim 1 , wherein the program instructions further cause the one or more processors to at least:

receive an engagement response indicating an actual engagement by the user with the first advertisement; and

based at least in part on the engagement response, update at least one of a quantity of advertisements to present or a second position of the sub-plurality of positions.

3 . The computing system of claim 1 , wherein the program instructions further cause the one or more processors to at least:

determine an application content for at least a portion of the application preceding the first advertisement placement position; and

select an advertisement associated with the content.

4 . The computing system of claim 1 , wherein the predicted engagement value is determined for a first position of the plurality of candidate advertisement placement positions based at least in part on actual user engagement with advertisements presented at the first position within the application.

5 . The computing system of claim 1 , wherein the program instructions when executed further cause the one or more processors to at least:

receive an indication that first advertisement has been presented; and

send a second advertisement for presentation in the application at a second advertisement placement position of the sub-plurality of advertisement placement positions.

6 . A computer-implemented method, comprising:

determining a user profile of a user executing an application on a user device;

determining, based at least in part on the user profile, an expected access duration that the user will access the application;

determining a quantity of advertisements desired to present to the user during the expected access duration;

determining, based at least in part on the expected access duration and the quantity of advertisement, a first advertisement placement position within the application at which a first advertisement of the quantity of advertisements is to be presented in the application to the user; and

sending to the user device a first advertisement for presentation at the first advertisement placement position.

7 . The computer-implemented method of claim 6 , wherein the expected access duration is further determined based at least in part on a condition at a location of the user device.

8 . The computer-implemented method of claim 7 , wherein the condition is at least one of a time of day, a day of week, or weather at a location of the user device.

9 . The computer-implemented method of claim 7 , wherein the expected access duration is further based at least in part on actual access duration by the user during similar conditions.

10 . The computer-implemented method of claim 7 , wherein the expected access duration is further based at least in part on actual access duration by other users during similar conditions.

11 . The computer-implemented method of claim 10 , further comprising:

determining that the user has a similar user profile to the other users.

12 . The computer-implemented method of claim 6 , further comprising:

receiving an engagement response indicating an engagement by the user with the first advertisement when presented at the first advertisement placement position.

13 . The computer-implemented method of claim 12 , further comprising:

altering, based at least in part on the engagement response, a second advertisement placement position or the quantity of advertisements.

14 . The computer-implemented method of claim 6 , wherein:

the first advertisement placement position is determined from a plurality of candidate advertisement placement positions available within the application at which advertisements may be presented in the application; and

wherein the plurality of candidate advertisement placement positions are indicated by a publisher of the application.

15 . A non-transitory computer-readable storage medium storing instruction that, when executed by at least one processor of a computing system, cause the computing system to at least:

receive, from a publisher, an indication of a plurality of candidate advertisement placement positions within an application at which advertisements may be presented in the application;

determine for a user an expected access duration;

determine a quantity of advertisements to present to the user during the expected access duration;

determine a first advertisement placement position from the plurality of candidate advertisement placement positions at which to present a first advertisement of the quantity of advertisements;

determine a second advertisement placement position from the plurality of candidate advertisement placement positions at which to present a second advertisement of the quantity of advertisements; and

send to a user device a first advertisement for presentation at the first advertisement placement position.

16 . The non-transitory computer-readable storage medium of claim 15 , wherein the quantity of advertisements is determined based at least in part on an advertisement engagement history of the user.

17 . The non-transitory computer-readable storage medium of claim 15 , wherein the expected access duration is determined based at least in part on at least one of an actual application use history of the user, or actual application use history of a plurality of users determined to have a similarity with the user.

18 . The non-transitory computer-readable storage medium of claim 15 , wherein the instructions further cause the computing system to at least:

develop, for the first advertisement placement position of the plurality of candidate advertisement placement positions, a model indicating a likelihood of engagement with an advertisement when presented in the application at the respective candidate advertisement placement position.

19 . The non-transitory computer-readable storage medium of claim 18 , wherein the instructions further cause the computing system to at least:

receive an engagement response indicating an engagement by the user with the first advertisement when presented at the first advertisement placement position; and

update the model based at least in part on the engagement response.

20 . The non-transitory computer-readable storage medium of claim 15 , wherein the instructions further cause the computing system to at least:

determine an amount of the application expected to be exposed to the user during the access duration; and

wherein the first advertisement placement position and the second advertisement placement position are within the amount of the application expected to be exposed to the user.