Determining lookback windows

• US 9,665,890 B1
• Filed: 11/21/2012
• Issued: 05/30/2017
• Est. Priority Date: 11/21/2012
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Kent, James
• Primary Examiner(s): Sittner, Matthew T
• Application Number:
  US13/683,092
• Time in Patent Office: 1,651 Days
• CPC Class Codes: G06Q 30/0272 Period of advertisement exposure

Abstract

Methods, systems, and apparatus, including computer programs encoded on a computer-readable storage medium, for determining a suggested time for linking interactions with events. A method includes: identifying data for a content provider that includes events and data related to presentation of and interactions of users with content; identifying a model for linking interactions with events that includes criteria for use in evaluating an interaction to determine when to link an interaction to a given event; for each event, determining a number of interactions that occurred in each of multiple time periods prior to a given event; performing a mathematical operation on the number of interactions in each time period producing a result; and determining a suggested time for linking interactions and events based on the result for each time period, the suggested time representing an amount of time before an event in which interactions are appropriately linked to the event.

Claims

1. A method comprising

logging, based on cookie information, interactions associated with a content provider, wherein the interactions represent interactions of users with content presented by the content provider;

logging a plurality of conversion events associated with products or services that are represented in the presentations of content;

evaluating, by one or more processors, logged conversion and interaction data including;

for each given conversion event in the plurality of logged conversion events;

identifying one or more interactions that are linked to the given conversion event;

determining, based on a comparison of a first timestamp for each identified interaction and a second timestamp of the conversion event, a time window before the given conversion event that is associated with each identified interaction; and

for each identified interaction, incrementing an interaction count associated with the time window associated with the interaction; and

determining one or more suggested time windows for the content provider for linking interactions and conversion events based on the interaction counts for each time window, wherein a suggested time window represents an amount of time before a conversion in which interactions are appropriately linked to the conversion;

determining that a given suggested time window of the one or more time windows exceeds an average cookie age of cookies from which the cookie information was received;

reducing the given suggested time window based on the determination that the given suggested time window exceeds the average cookie age;

providing, through a user interface presented to the content provider, a presentation of the given suggested time window and others of the suggested one or more time windows;

receiving, through the user interface, an acceptance of a suggested time window to be used; and

using the accepted time window to determine attribution credit to be applied to specific interactions that are correlated to specific conversion events.

2. The method of claim 1 wherein determining a suggested time includes selecting a suggested time window as an aggregate of multiple possible time windows.

3. The method of claim 1 wherein determining a suggested time includes selecting a time window from among multiple possible time windows that is slowest to converge.

4. The method of claim 1 further comprising identifying a sensitivity to using the suggested time, and applying a correction factor to the suggested time based at least in part on the sensitivity.

5. The method of claim 1 further comprising adjusting the suggested time based at least in part on cookie deletion data, wherein the cookie deletion data indicates an amount of time a cookie is likely to be valid, and wherein cookies are used to assist in accumulating the events and interactions.

6. The method of claim 1 wherein the criteria include criteria that define different types of sponsored content or different mediums for presentation of the sponsored content and where the suggested time is attributable to the given criteria.

7. The method of claim 1 further comprising:

identifying a current lookback window associated with the content provider,

comparing the current lookback window to the suggested time including

providing a comparison of a number of interactions or dollar amount that is discounted based on the comparison; and

providing a summary of the comparison to the content provider.

8. The method of claim 1 wherein linking includes attributing the interaction to the event.

9. The method of claim 1 wherein the time windows are non-overlapping.

10. The method of claim 1 wherein the suggested time defines a time window for linking interactions with events.

11. The method of claim 1 wherein linking includes, for a given event, crediting an entity associated with an interaction that occurred in a suggested time prior to an occurrence of the given event.

12. A non-transitory computer program product tangibly embodied in a computer-readable storage device and comprising instructions that, when executed by a processor, cause the processor to:

log, based on cookie information, interactions associated with a content provider, wherein the interactions represent interactions of users with content presented by the content provider;

log a plurality of conversion events associated with products or services that are represented in the presentations of content;

evaluate, by one or more processors, logged conversion and interaction data including;

for each given conversion event in the plurality of logged conversion events;

identify one or more interactions that are linked to the given conversion event;

determine, based on a comparison of a first timestamp for each identified interaction and a second timestamp of the conversion event, a time window before the given conversion event that is associated with each identified interaction; and

for each identified interaction, increment an interaction count associated with the time window associated with the interaction; and

determine one or more suggested time windows for the content provider for linking interactions and conversion events based on the interaction counts for each time window, wherein a suggested time window represents an amount of time before a conversion in which interactions are appropriately linked to the conversion;

determine that a given suggested time window of the one or more time windows exceeds an average cookie age of cookies from which the cookie information was received;

reduce the given suggested time window based on the determination that the given suggested time window exceeds the average cookie age;

provide, through a user interface presented to the content provider, a presentation of the given suggested time window and others of the suggested one or more time windows;

receive, through the user interface, an acceptance of a suggested time window to be used; and

use the accepted time window to determine attribution credit to be applied to specific interactions that are correlated to specific conversion events.

13. A system comprising:

a content management system;

an interaction data store;

wherein the content management system is configured to;

log, in the interaction data store and based on cookie information, interactions associated with a content provider, wherein the interactions represent interactions of users with content presented by the content provider;

log, in the interaction data store, a plurality of conversion events associated with products or services that are represented in the presentations of content;

evaluate, by one or more processors, logged conversion and interaction data including;

for each given conversion event in the plurality of logged conversion events;

identify one or more interactions that are linked to the given conversion event;

determine, based on a comparison of a first timestamp for each identified interaction and a second timestamp of the conversion event, a time window before the given conversion event that is associated with each identified interaction; and

for each identified interaction, increment an interaction count associated with the time window associated with the interaction; and

determine one or more suggested time windows for the content provider for linking interactions and conversion events based on the interaction counts for each time window, wherein a suggested time window represents an amount of time before a conversion in which interactions are appropriately linked to the conversion;

determine that a given suggested time window of the one or more time windows exceeds an average cookie age of cookies from which the cookie information was received;

reduce the given suggested time window based on the determination that the given suggested time window exceeds the average cookie age;

provide, through a user interface presented to the content provider, a presentation of the given suggested time window and others of the suggested one or more time windows;

receive, through the user interface, an acceptance of a suggested time window to be used; and

use the accepted time window to determine attribution credit to be applied to specific interactions that are correlated to specific conversion events.