Google Analytics Academy – How Google Analytics Works
In this rather brief technical overview, the inner-workings of Google Analytics are dissected into four key areas:
Processing and Configuration
The course makes the point to highlight the duality of this step. After collection, Google Analytics processes the data according to configuration settings. It is also noted that once the data is subsequently stored in the database, it is saved that way permanently. In other words, get your configuration right!
This is the pretty Google Analytics interface that most are familiar with. It is noted that you can access data via the Google Analytics interface as well as an API.
Dimensions: Characteristics of users, and their sessions and actions
Examples of Dimensions
Metrics: Quantitative measures of users, and their sessions and actions; Helps you understand behavior
Examples of Metrics
Sessions and Session Timeouts
Understanding the concept of sessions, and session timeouts, is important for planning a Google Analytics strategy.
Session: A period of consecutive activity by the same user
Default Session Timeout: After 30 minutes of no new activity, Google Analytics assumes the session is over. This can be customized to the needs of the website. The example given was a video streaming website where a user might watch a video for 45 minutes.
Page Views vs. Events
Page Views: Every time a page is shown on the website
Events: Additional interaction that requires customization (ex. clicking on a video). Keep the visitors session “active”
Both keep the session “active”
Bounce Rate: Percentage of sessions with only one interaction
The historical definition of “bounce rate” reflected the percentage of website visitors that immediately left a site after landing. Google Analytics, however, defines bounce rate as the percentage of visitors with only one interaction. In theory, a visitor can read one single page for 20 minutes and then close the site. This would still count as a “bounce.”
Possible Reason for High Bounce Rate: Mismatched Content / Expectations
Possible Reason for High Bounce Rate: Poor User Experience
Possible Reason for High Bounce Rate: Lots of individual articles (i.e. blog or news site)
Alternative use: Testing landing page effectiveness
Final Note for Sites with AJAX and Flash
nce Google Analytics uses visitor interaction, sites with a lot of AJAX and Flash may need to add Event Tracking methods to properly capture the needed data.