Multi-touch attribution model for valuing impressions and other online activities
View Patent ↗To implement a multi-touch attribution model, a conversion event resulting from user activity is programmatically detected. A set of events that precede the conversion event is identified. Respective events of the set occur on respective websites. The conversion event is attributed to multiple websites of the respective websites.
1. A method of valuing impressions over a network, the method being implemented by one or more processors and comprising:
monitoring activity of a plurality of computer systems operated by a corresponding plurality of individuals;
programmatically detecting a common conversion event from the monitored activity of two or more computer systems of the plurality of computer systems, resulting from user activity;
programmatically analyzing activities performed by corresponding users on the two or more computer systems to identify a set of events that occur in the activities performed by each of the two or more computer systems which precede the common conversion event; and
programmatically associating the common conversion event with a set of online resources that are relevant to the set of events.
2. The method of claim 1 , wherein:
the set of events comprises an event associated with a visit by a first user to a first website and a visit by the first user to a second website; and
associating the common conversion event comprises assigning partial credit for the common conversion event to the first website and partial credit for the common conversion event to the second website.
3. The method of claim 1 , wherein the activities are limited to a specified time window.
4. The method of claim 1 , wherein:
the set of events comprises impressions displayed on the respective websites; and
associating the common conversion event comprises assigning partial credit for the common conversion event to respective impressions on the multiple websites.
5. The method of claim 1 , wherein associating the common conversion event comprises:
selecting, from the set of events, a group of events relating to the common conversion event; and
weighting the events of the group using a time decay function.
6. The method of claim 5 , wherein the time decay function is an exponential function having an associated half-life.
7. The method of claim 1 , wherein associating the common conversion event comprises:
selecting, from the set of events, a group of events relating to the common conversion event; and
weighting the events of the group based on the times of occurrence of individual events in the set of events, using weights that increase in response to user interactions and subsequently decrease.
8. The method of claim 7 , wherein associating the common conversion event further comprises weighting the events of the group using a time decay function.
9. The method of claim 7 , wherein:
the common conversion event comprises an on-line purchase; and
the group of events relating to the common conversion event comprises on-line displays of advertisements relating to the purchase.
10. The method of claim 1 , wherein a respective event of the set of events corresponds to a cluster of impressions.
11. The method of claim 1 , wherein the set of events is limited to a specified number of events.
12. A non-transitory computer-readable storage medium storing one or more programs configured to be executed by a computer system, the one or more programs comprising instructions to:
monitor activity of a plurality of computer systems operated by a corresponding plurality of individuals;
detect a common conversion event from the monitored activity of two or more computer systems of the plurality of computer systems, resulting from user activity;
analyze activities performed by corresponding users on the two or more computer systems to identify a set of events that occur in the activities performed by each of the two or more computer systems which precede the common conversion event; and
associate the common conversion event with a set of online resources that are relevant to the set of events.
13. The computer-readable storage medium of claim 12 , wherein:
the set of events comprises an event associated with a visit by a first user to a first website and a visit by the first user to a second website; and
the instructions to associate comprise instructions to assign partial credit for the common conversion event to the first website and partial credit for the common conversion event to the second website.
14. The computer-readable storage medium of claim 12 , wherein:
the set of events comprises impressions displayed on the respective websites; and
the instructions to associate comprise instructions to assign partial credit for the common conversion event to respective impressions on the multiple websites.
15. The computer-readable storage medium of claim 12 , wherein the instructions to associate comprise instructions to:
select, from the set of events, a group of events relating to the common conversion event; and
weight the events of the group using a time decay function.
16. The computer-readable storage medium of claim 12 , wherein the instructions to associate comprise instructions to:
select, from the set of events, a group of events relating to the common conversion event; and
weight the events of the group based on the times of occurrence of the individual events in the set of events, using weights that increase in response to user interactions and subsequently decrease.
17. The computer-readable storage medium of claim 16 , wherein the instructions to associate further comprise instructions to weight the events of the group using a time decay function.
18. The computer-readable storage medium of claim 16 , wherein:
the common conversion event comprises an on-line purchase; and
the group of events relating to the common conversion event comprises on-line displays of advertisements relating to the purchase.