I have implemented content targeting using Engage based on the visitor's profile (collected via registration/login page).
What I want to know is, is it possible to target content based on the click stream or navigation pattern/behavior of the visitor?
Please let me know where I can find pointers/documentation related to this kind of implementation.
It does not appear there is anything out of the box to do this. Here is another blog entry on the issue: https://blogs.oracle.com/webcenter/entry/the_power_of_automated_targeting
I couldn't find anything OOTB either. But I might just be able to manufacture something using the History attributes, perhaps, capturing links, storing in history attributes and applying segmentation rules on that.
Let me see how far I can get with this. will post back if successful.
I finally managed to complete this (been juggling it with three other tasks!).
What I did:
I created History Definition and History Attributes.
Coded the templates (with vdm:recordhistory tags) so that clicking a link fetches values from content attribute and store as history attribute values.
Created segments like SportsFan, MarvelFan, etc. with segment criteria on values of the history attributes.
Created the recommendations.
The result: When site visitors click on links or browse the site, they get classified into segments (based on the content clicked of course!) and the home page calculates the segment and shows the recommendations. Also, since the classification into segments is stored in the database (using history attributes), a visitor gets to see recommendation based on his earlier visits.
Let me know if this helps anyone.