More About Google Analytics?
Absolutely!
If 4-6 hours of YouTube videos exist teaching anyone who wants to watch them, I am sure I could find a thing or two more to write about.
"Platform Principles" is the title of the second chapter of the Google Analytics Academy.
The dry, somewhat monotonous, hand-gestury, and my professor's old highschool band mate, Justin Cutroni, continues to build my knowledge on the excellent, and genius, program that Google Analytics (GA) is.
Cutroni, in this chapter, describes the four Platform Components to GA; Collection, Processing, Configuration, and Reporting.
First, I should relay a few definitions that GA possesses. Users are individuals that visit your website. It's worth noting, that GA sees a new visitor, even if it is the same person, when they reset their cookies, change browsers, or, for a mobile application, deletes the application, then reinstalls it. Sessions are visits that an individual has on your website. Again, it's worth noting that a new session will be sent to GA servers after a default of 30 minutes of inactivity. In the configuration setting, the GA user can adjust this time. Interactions are hits on the website, such as page views, events, and transactions.
Moving along now...
In the Collection unit, he describes the process of how GA collects the data from each session a user has on a website, mobile application or and web-connected device. On a website, the JavaScript code, written in the header of each page on the website, collects and sends hits, or interactions, to the GA servers for configuration and processing. On a mobile application, a Software Developer Kit (SDK) must be used because of how applications are written. One interesting thing about mobile applications, compared to websites, is the hits are stored on the device and collected in batches, then sent to GA servers. GA does this because of lack of service and to reduce the battery use. Ironically enough, the batches are sent from Android phones are sent to GA every 30 minutes, while iOS data batches are sent every 2 minutes.
What a difference in battery life that must have on the different phone platforms...
Cutroni then lumps processing and configuration in the next few videos as they are highly related. Processing does 4 main things:
- Organizes data into users and sessions.
- Imported data is joined with GA data
- Additional configurations are applied
- Data is aggregated into tables.
Configuration, on the other hand, takes the processed data and transforms and aggregates it into meaningful data. Filters exclude, include, and change how data looks. Goals observe conversions and transforms it into conversion rates. Grouping aggregates data so the observer can analyze the collective performance of the website.
Last, and surely not least, GA reports the collected, processed, and configured data into graphs and tables for the owner of the website to process in their own mind to gain insight on the performance of the landing pages, campaigns, and advertisements. GA tables are well organized. The first column will be the dimension of the session or interaction, while the rest of the columns will contain the metrics of the session or interaction. For example, the Country that the user came from will be in the first column, while visits, unique visits, pages per visit, and average duration will be in the columns to follow. GA can also utilize sampling to report data faster. The website owner may want to sample due to new segments, custom reports, and various other custom requests. The website owner can also change the size of the sample to ensure the accuracy, or speed, of results.
The Google Analytics Demo Account, provides an ample amount of data for those learning GA to practice with looking at and gaining valuable insight from.
If you are learning GA, or want additional practice and do not already have a website that is gaining traffic, I highly suggest using the demo account (spoilers below).
The Demo Account is tracking data for the Google Merchandise Store. I am sure the data is not from the actual Google Merchandise Store, but who knows?
The first thing I did was compare December of 2015 and 2016 on a few metrics. For the most part, the two months, a year apart, were pretty similar. The pages per session was one metric where I did notice a bit of difference. Pages per session were consistently down in December 2016, as compared to December 2015 by a range of over 55% to about 3%. As seen below, the pages per session in December 2016 never want above the same metric a year earlier. I wasn't sure why this happened, so I had to dig further..
The next thing I examined was different campaigns the Merchandise store held and the effectiveness of those campaigns. One particular campaign stood out to me; "AW-Dynamic Search Ads Whole Site." These campaigns stood out to me because of its effectiveness.
The Dynamic Search Ads campaign had a pretty clear effectiveness. The amount of New Users increased over 2000 percentage points from the same month in the previous year. Even more impressive was the over all goal completion rate, which increased almost 9000 percentage points from the previous year! Being a bit more curious about specific e-commerce goals, I glanced into Goal 4-Entered Checkout, and Goal 1-Purchase Completions.
Both Goals that were set were clearly met, but they also provided more insight. Of 122 individuals that entered the Checkout, 90 completed purchases. All things considered, I would argue that is a good final conversion rate. In sum, I would argue the "AW-Dynamic Search Ads Whole Site" was an effective campaign put into place by the Google Merchandise store.
When I am conducting campaigns in the future, I will highly consider using Dynamic Search Ads in order to help bring new users to my website and hopefully create a higher goal conversion rate, especially on an e-commerce website.
For more information I suggest checking out the Google Support Page on Dynamic Search Ads.