In my last post, I addressed the question: Why Google Analytics? Now that I have begun to dive deeper into what makes this software program tick, I must ask the question: How Google Analytics? or how does Google Analytics work?
Why the “How”?
But why is it important to know the “how”? If I can interact with the software and interpret the information that it generates, then why do I need to know what is behind the curtain? Well, sometimes knowing what is behind the curtain can make all of the difference; Just ask Dorothy! No reputable researcher would ever accept or publish any findings without first knowing the source of information and how that information was generated. So what can the “how” tell the researcher? It can tell them whether or not the data being collected is trustworthy, whether or not their experiments are empirically sound, and whether or not the data is being structured in meaningful ways. Knowing these things is important so that the researcher can make accurate, reliable, and useful conclusions based off of the information that has been generated. For these same reasons, it is important to know how Google Analytics works. Knowing the “how” can mean the difference between making interpretations and making meaningful and reliable interpretations.
With all of that in mind, let’s explore some of the very basics on how Google Analytics generates information. Data goes through four main stages in Google Analytics:
Data is collected using a tracking code that is embedded into the programming language on each page of a website. When certain actions are detected on a web page, the tracking code sends out an “image request” to Google Analytics servers called a “hit”. Each time a certain action is detected, a hit is sent out to the server and that is how the data is collected. The data that Google Analytics collects falls into three main categories: Users, Sessions, and Interactions. All three of these categories are closely related but differ in scope. Google Analytics tracks information on each user (visitor to the site), information about each session (visit to the site) that a user has, and the amount and type of interactions (hits) that the user has with the website based on what they click on. A user can have multiple sessions (visits) to a website and a session can consist of multiple interactions (hits).
After being collected, the data is configured or transformed based on settings that filter, group, and select relevant metrics based on specified goals. The data is then processed by organizing the data into user and session models, importing external data from other predetermined sources, applying configurations, and aggregating the data so that it can be placed into a table. Finally, the data is placed in tables that display relevant quantitative metrics with specified qualitative dimensions in the reporting stage. These tables and their corresponding charts can then be used by analysts to make interpretations about the information that has been generated.
The Need for Understanding How and Why
Now that I have briefly explained how Google Analytics works and why it is important to know how it works, I would like to add more value to this discourse by quickly looking at how and why any of this information is significant at all. More specifically, I will look at it from the angle of industry need. Justin Cutroni posted an article on the Google Analytics blog that explains the importance of data fluency and identifies the increasing industry need for proficient data analysts. He states that:
This would suggest that not only is data analysis important for decision making, it is important because of decision making. And as I mentioned in my last post, decisions are what drive a company forward or into the ground. Corporations need skilled data analysts that can generate and interpret useful information, allowing for the initiation of excellent decision making that will drive organizations forward. But you shouldn’t take my word for it. Take Justin’s.