|
Web Analytics Tutorial |
Appendix B – Technical Details of Metrics Accuracy | ||||
Visit Time Issues | ||||
|
In the introduction to this lesson, we discussed some of the problems with Visit Detection. Because HTTP requests are distinct, there is no absolute way of determining which request comprise a single visit. In addition, Proxies add another factor by obscuring the source of the requests and making it hard to distinguish different users who may be visiting at the same time. View TimeSummary tracks the time a particular page was viewed and includes this column
in some reports. The Finally, the last page of a visit is where the user either clicks a link to a new site, types in a new URL or closes the browser. In most cases, you have no record of when she left the site. This means that Summary cannot determining how long the last page view was. In the case of pages that contain extended information that a user was searching for, this could be a significant amount of time that is never recorded. Visit DurationThis last point becomes especially important when trying to calculate the
length of the entire visit. Because there is no record of when the visitor left
the site, visit duration, as summarized in the |
|
|||
|
Fortunately, there is a partial solution to this. In Lesson 7 - Determining Visitor Behavior Patterns we talked about setting up a script or configuring your web server for link following. When you do this, each link that leaves your site goes through a redirector first. This means that you get one last request in your log files as the visitor is leaving. If you install this script then you can track the view time of exit pages and your visit duration will include the entire length of the visit for visitors following links to another site. Some visitors will close their browser or type in a new URL rather than following a link. The only way to track this is to install some JavaScript code on every page of your site that makes a final request to your site before the page changes. |
|
|||
| ||||
|
Table of Contents |
1: What is Web Analytics? |
2: Where are My Visitors Coming From? |
3: Search Engines |
4: Advertising |
5: Revenue Modeling |
6: Design Considerations |
7: Determining Visitor Behavior Patterns |
8: Examining Subsets of Traffic |
9: Incorporating Business Goals |
10: Bandwidth Management |
11: Site and Server Diagnostics |
12: Investigating Troublemakers |
Appendix A: Making Reports More Usable |
Appendix B: Technical Details of Metric Accuracy Copyright 2002 by Summary.Net - Updated 16.Apr.2002 |