Using visitor context and web page features to select web pages for display
In one embodiment, a method includes accessing a current-visitor context of a current visitor to a web page in a current web-browsing session. The current-visitor context includes one or more data associated with or concerning the current visitor. The method includes selecting based on the current-visitor context a particular one of multiple possible instances of the web page for presentation to the current visitor. The particular one of the multiple possible instances of the web page is substantially most likely to generate a highest expected outcome from interaction with the web page by the current visitor as indicated by the current-visitor context.
1. A method comprising:
receiving, by one or more computer systems, a request for a landing page to be presented to a current visitor, the landing page having one or more landing-page components each of which results from combining different feature variations;
predicting, by the one or more computer systems, a plurality of responses of a current visitor to possible instances of the landing page based on a current visitor context and possible feature variations of the possible instances of the landing page,
selecting, by the one or more computer systems, one of the possible: instances of the landing page based on the predicted responses;
constructing, by the one or more computer systems, an instance of the landing page based on the selected one of the possible instances of the landing page; and
transmitting, by the one or more computer systems, the constructed instance of the landing page to the current visitor.
2. The method of claim 1 , wherein the current visitor context comprises one or more of a group of current visitor contexts comprising: demographic data of the current visitor;
conditions of a current web-browsing session;
network-access data of the current web-browsing session; and
network-traffic data of the current web-browsing session.
3. The method of claim 2 :
further comprising accessing one or more previous-visitor contexts of one or more previous visitors to the constructed landing page in previous web-browsing sessions; and
wherein:
the one of the possible instances of the landing page is selected based on the previous-visitor contexts in addition to the current visitor context; and
the one of the possible instances of the landing page is substantially most likely to generate a highest expected outcome from interaction with the landing page by the current visitor as indicated by the previous visitor context as well as the current visitor context.
4. The method of claim 3 , wherein previous web-browsing sessions are analogous to current web-browsing sessions.
5. The method of claim 4 , wherein the previous visitor contexts comprise one or more of a group of previous visitor contexts comprising:
demographic data of the previous visitor;
background data of the previous visitor;
conditions of the previous web-browsing sessions;
network-access data of the previous web-browsing sessions; and
network-traffic data of the previous web-browsing sessions.
6. The method of claim 4 , wherein the previous visitor contexts comprise one or more of a group of previous visitor contexts comprising:
an action taken by the previous visitor;
a response of the previous visitor to the landing page; and
a traffic source from which the previous visitor accessed the landing page.
7. The method of claim 2 , wherein:
the constructed landing page has one or more web-page components that each have one or more possible feature variations; and
the plurality of possible instances of the landing page result from different combinations of at least some of the possible feature variations of the web-page components.
8. The method of claim 7 , wherein the features are selected from a group comprising a background color, an audio element, a video element, a text element, a size, a dimension, and a web-page layout.
9. The method of claim 2 , wherein the constructed landing page comprises one or more advertisements.
10. The method of claim 9 , wherein one or more of the advertisements include banner advertisements.
11. The method of claim 10 , wherein one or more of the advertisements include text-based content.
12. The method of claim 1 , wherein the current visitor context comprises one or more of a group of current visitor contexts comprising:
an action taken by the current visitor;
a response of the current visitor to the constructed landing page; and
a traffic source from which the current visitor is accessing the constructed landing page.
13. The method of claim 1 , further comprising receiving a response from the current visitor to the constructed instance of the landing page; and
updating a performance estimate of a feature variation based on the response from the current visitor.
14. A system comprising:
one or more processors and a memory storing executable instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
receiving, by one or more computer systems, a request for a landing page to be presented to a current visitor, the landing page having one or more landing-page components each of which results from combining different feature variations;
predicting, by the one or more computer systems, a plurality of responses of a current visitor to possible instances of the landing page based on a current visitor context and possible feature variations of the possible instances of the landing page,
selecting, by the one or more computer systems, one of the possible instances of the landing page based on the predicted responses;
constructing, by the one or more computer systems, an instance of the landing page based on the selected one of the possible instances of the landing page; and
transmitting, by the one or more computer systems, the constructed instance of the landing page to the current visitor.
15. The system of claim 14 , wherein the one or more processors are further configured to receive a response from the current visitor to the constructed instance of the landing page; and
update a performance estimate of a feature variation based on the response from the current visitor.
16. The system of claim 14 , wherein the current visitor context comprises one or more of a group of current visitor contexts comprising:
demographic data of the current visitor;
conditions of a current web-browsing session;
network-access data of the current web-browsing session; and
network-traffic data of the current web-browsing session.
17. The system of claim 14 , wherein the current visitor context comprises one or more of a group of current visitor contexts comprising:
an action taken by the current visitor;
a response of the current visitor to the constructed landing page; and
a traffic source from which the current visitor is accessing the constructed landing page.
18. The system of claim 14 :
wherein the one or more processors are further configured to access one or more previous-visitor contexts of one or more previous visitors to the constructed landing page in previous web-browsing sessions; and
wherein:
the one of the possible instances of the landing page is selected based on the previous-visitor contexts in addition to the current visitor context; and
the one of the possible instances of the landing page is substantially most likely to generate a highest expected outcome from interaction with the landing page by the current visitor as indicated by the previous visitor context as well as the current visitor context.
19. A non-transitory machine-readable medium including instructions which, when read by a machine, cause the machine to perform operations comprising, at least:
receiving, by one or more computer systems, a request for a landing page to be presented to a current visitor, the landing page having one or more landing-page components each of which results from combining different feature variations;
predicting, by the one or more computer systems, a plurality of responses of a current visitor to possible instances of the landing page based on a current visitor context and possible feature variations of the possible instances of the landing page,
selecting, by the one or more computer systems, one of the possible instances of the landing page based on the predicted responses;
constructing, by the one or more computer systems, an instance of the landing page based on the selected one of the possible instances of the landing page; and
transmitting, by the one or more computer systems, the constructed instance of the landing page to the current visitor.
20. The non-transitory machine-readable medium of claim 19 , wherein the operations further comprise receiving a response from the current visitor to the constructed instance of the landing page; and
updating a performance estimate of a feature variation based on the response from the current visitor.