Implement Tracking
Enable ‘Click’-events
After installing the ezrecommendation extension, the global pagelayout.tpl is overridden to track events automatically.
If you have customized your global pagelayout.tpl, you can do these changes manually, following the steps below:
The first step is to include the following code in your pagelayout.tpl. The code must be included in the body of the pagelayout.tpl but outside of any caching block. You can place it e.g. right before the </body> tag:
{* enable tracking of click events in pagelayout.tpl *}{include uri='design:content/ezrecommendation_html.tpl'
content=$module_result track=true()}After this step you will see that ´click´-events will be created for any node that is marked as recommendable through the class definition. In our case all content of the classes ´Article for recommendation´, ´Product for recommendation´ and ´Image for recommendation´.
´Click´-events are created in the moment you click on the link to a recommendable object.
You can verify whether an event has been created by checking the ´debug.log´ in the directory ´<eZ publish root directory>/var/log´. You will find entries like the following for each tracking event being sent to the eZ recommender engine depending on the event chosen:
[ Sep 11 2011 14:17:52 ] event.yoochoose.net/news/<strong><CustomerID></strong>/click/ 131574345055084112/2/110?categorypath=%2F2%2F107%2F110
If you change from track=true() to track=false() in the above include you will obviously disable tracking for any page.
In this section
- Create other common events ('owns', 'consume', 'blacklist') Exponential Documentation
- 'Consume'-event Exponential Documentation
- 'Rate'-event Exponential Documentation
- 'Purchase'-event Exponential Documentation
- 'Delete'-event Exponential Documentation
- 'Rendered'-event Exponential Documentation
- 'Transfer user'-event Exponential Documentation
- 'Click recommend'-event This event is of outmost importance if you want to measure the acceptance of your recommendations. In the statistics we use the ratio of ´Click recommended´-events” divided by the total number of recommendation calls to calculate the so-called conversion rate:
Contributors
- Andrea Melo Updated
Comments
Add a comment