Predicting User Interactions With Objects Associated With Advertisements On An Online System

Abstract

Based on prior interactions associated with a user, an online system predicts an amount of interaction by the user with an object associated with an advertisement. Using the predicted amount of user interaction, the online system determines an expected value of presenting the advertisement to the user. The advertisement is ranked among other advertisements based on the expected values associated with the advertisements, and one or more advertisements are selected for presentation to the user based on the ranking. An advertisement may also specify a threshold amount of interaction with an associated object as targeting criteria, so the predicted amount of interaction with the object associated with the advertisement may determine if a user is eligible to be presented with the advertisement.

What is claimed is: 
     
         1 . A method comprising:
 receiving, at an online system, information describing an advertisement from an advertiser;   identifying an opportunity to present the advertisement to a user of the online system via a client device;   retrieving data stored by the online system describing interactions previously performed by the user;   determining a likelihood of one or more user interactions with an object associated with the advertisement based at least in part on the retrieved data;   determining a predicted amount of interaction by the user with the object associated with the advertisement based at least in part on the determined likelihood;   determining an expected revenue from the predicted amount of interaction by the user with the object associated with the advertisement;   ranking a plurality of advertisements including the advertisement based at least in part on the expected revenue from the predicted amount of interaction by the user with the object associated with the advertisement;   selecting an advertisement for presentation to the user based at least in part on the ranking; and   sending the selected advertisement for presentation to the client device for presentation to the user.   
     
     
         2 . The method of  claim 1 , wherein the one or more user interactions with the object associated with the advertisement are selected from a group consisting of: spending an amount of money interacting with the object associated with the advertisement, spending an amount of time interacting with the object associated with the advertisement, accessing the object associated with the advertisement a number of times, subscribing to a number of services from the object associated with the advertisement, purchasing a number of products from the object associated with the advertisement, attending a number of events associated with the object associated with the advertisement, achieving a status determined by the object associated with the advertisement, and any combination thereof. 
     
     
         3 . The method of  claim 1 , wherein the object associated with the advertisement is selected from a group consisting of: an application, a retailer, a third-party system, and any combination thereof. 
     
     
         4 . The method of  claim 1 , wherein the retrieved data is selected from a group consisting of: information from a user profile associated with the user and maintained by the online system, actions performed by the user on the online system, actions performed by the user on one or more third-party systems, actions performed by one or more users of the online system connected to the user, information from one or more user profiles associated with one or more users connected to the user, a geographic location associated with the user, and any combination thereof 
     
     
         5 . The method of  claim 1 , wherein determining the expected revenue from the predicted amount of interaction by the user with the object associated with the advertisement comprises:
 determining a bid amount for the advertisement based at least in part on the predicted amount of interaction by the user with the object associated with the advertisement and a return on investment associated with the advertisement and included in the information describing the advertisement.   
     
     
         6 . The method of  claim 5 , wherein ranking the plurality of advertisements including the advertisement comprises:
 ranking the plurality of advertisements based at least in part on bid amounts associated with each of the plurality of advertisements.   
     
     
         7 . The method of  claim 5 , wherein determining the expected revenue from the predicted amount of interaction by the user with the object associated with the advertisement comprises:
 determining a confidence interval providing a measure of a reliability of the predicted amount of interaction by the user with the object associated with the advertisement; and   modifying the bid amount for the advertisement based at least in part on the confidence interval.   
     
     
         8 . The method of  claim 1 , wherein determining the expected revenue from the predicted amount of interaction by the user with the object associated with the advertisement comprises:
 retrieving a return on investment associated with the advertisement from the information describing the advertisement;   retrieving a budget for presenting the advertisement from the information describing the advertisement; and   determining a bid amount for presenting the advertisement based at least in part on the return on investment, the budget, and the predicted amount of interaction by the user with the object associated with the advertisement.   
     
     
         9 . The method of  claim 1 , wherein the expected revenue from the predicted amount of interaction by the user with the object associated with the advertisement is further based on one or more targeting criteria associated with the advertisement and identifying one or more characteristics of users of the online system eligible to be presented with the advertisement. 
     
     
         10 . The method of  claim 9 , wherein the targeting criteria is based on one or more selected from the group consisting of: the expected value of presenting the advertisement to the user, descriptive information associated with the user and included in a user profile maintained by the online system, actions performed by the user on the online system, actions performed by the user on one or more systems external to the online system, information associated with one or more users of the online system connected to the user, a geographic location associated with the user, device parameters for the client device, and any combination thereof. 
     
     
         11 . The method of  claim 1 , wherein the expected revenue from the predicted amount of interaction by the user with the object associated with the advertisement is described relative to a threshold value. 
     
     
         12 . The method of  claim 1 , further comprising loading a tracking mechanism associated with the advertisement if the user performs an interaction from the one or more interactions with the object associated with the advertisement after presentation of the advertisement. 
     
     
         13 . A method comprising:
 receiving, at an online system, information describing an advertisement associated with an object;   receiving a request to present one or more advertisements to a user of the online system via a client device;   determining a predicted amount of interaction with the object associated with the advertisement by the user based at least in part on prior interactions between the user and one or more objects identified by the online system;   determining an expected value of presenting the advertisement to the user based at least in part on the predicted amount of interaction with the object associated with the advertisement;   ranking the advertisement among a plurality of advertisements based at least in part on the expected value of presenting the advertisement to the user;   selecting the advertisement from the plurality of advertisements based on the ranking; and   sending the advertisement to the client device for presentation to the user.   
     
     
         14 . The method of  claim 13 , wherein the one or more user interactions with the object associated with the advertisement are selected from a group consisting of: spending an amount of money interacting with the object associated with the advertisement, spending an amount of time interacting with the object associated with the advertisement, accessing the object associated with the advertisement a number of times, subscribing to a number of services from the object associated with the advertisement, purchasing a number of products from the object associated with the advertisement, attending a number of events associated with the object associated with the advertisement, achieving a status determined by the object associated with the advertisement, and any combination thereof. 
     
     
         15 . The method of  claim 13 , wherein the object associated with the advertisement is selected from a group consisting of: an application, a retailer, a third-party system, and any combination thereof. 
     
     
         16 . The method of  claim 13 , wherein the prior interactions between the user and one or more objects identified by the online system are selected from a group consisting of:
 information from a user profile associated with the user and maintained by the online system, actions performed by the user on the online system, actions performed by the user on one or more third-party systems, actions performed by one or more users of the online system connected to the user, information from one or more user profiles associated with one or more users connected to the user, a geographic location associated with the user, and any combination thereof.   
     
     
         17 . The method of  claim 13 , wherein determining the expected value of presenting the advertisement to the user comprises:
 determining a bid amount for the advertisement based at least in part on the predicted amount of interaction with the object associated with the advertisement and a return on investment associated with the advertisement and included in the information describing the advertisement.   
     
     
         18 . The method of  claim 17 , wherein ranking the plurality of advertisements including the advertisement comprises:
 ranking the plurality of advertisements based at least in part on bid amounts associated with each of the plurality of advertisements.   
     
     
         19 . The method of  claim 17 , wherein determining the expected value of presenting the advertisement to the user further comprises:
 determining a confidence interval providing a measure of a reliability of the predicted amount of interaction with the object associated with the advertisement; and   modifying the bid amount for the advertisement based at least in part on the confidence interval.   
     
     
         20 . The method of  claim 13 , wherein the expected value of presenting the advertisement to the user is further based on one or more targeting criteria associated with the advertisement and identifying one or more characteristics of users of the online system eligible to be presented with the advertisement. 
     
     
         21 . The method of  claim 20 , wherein the targeting criteria is based on one or more selected from the group consisting of: the expected value of presenting the advertisement to the user, descriptive information associated with the user and included in a user profile maintained by the online system, actions performed by the user on the online system, actions performed by the user on one or more systems external to the online system, information associated with one or more users of the online system connected to the user, a geographic location associated with the user, device parameters for the client device, and any combination thereof. 
     
     
         22 . The method of  claim 13 , wherein the expected value of presenting the advertisement to the user comprises an expected revenue based on the predicted amount of interaction with the object associated with the advertisement.