If you’ve worked with Google Analytics data before, then you know how quickly reports can become a hot mess when customizing data without following specific business rules or best practices. Additionally, collecting and understanding your Google Analytics data when connecting to Power BI becomes much more complex and time-consuming than it should be.
In this blog post, I will share my experience connecting Power BI+ Google Analytics data and attempt to keep the conversation going on some best practices for implementing an effective measurement plan and naming conventions for your business’ website.
I believe it is essential to begin with a solid understanding of your business objectives and translate that logic into metrics and dimensions. You can then measure by leveraging the strength of Power BI to gain real-time business insights pulled from your company’s website.
One-size does NOT fit all.
Coming from someone who has spent the better part of the last three months researching and combing through blogs, articles, forums and watching YouTube videos, the two most mentioned statements I found were, “Start with a [measurement] plan,” and, “There’s no one-size-fits-all solution.” After being involved in my first project connecting to our client’s Google Analytics data in Power BI Desktop to create interactive real-time reports, I could not agree more. Starting with a plan of how your business wants to track your website’s data is critical. Not every company is the same, which means there is no cookie-cutter way to plan how you should measure the success of your website. However, not developing a plan and/or identifying key performance indicators (KPIs) to measure at all is a sure way to get overwhelmed by the ocean of data that’s collected every second by Google Analytics. So, step 1) set up a measurement plan. Step 2) come up with effective naming conventions that make sense for YOUR business.
Naming conventions should be applied in many vital areas. But in this blog, I focus on the three key areas where naming conventions are most important in Google Analytics; creating accounts, properties, and views. If these areas are easily identified and distinguished by users, locating the desired metric and dimensions is much more efficient when connecting to Power BI Desktop. Let’s take a look.
Meaningful Naming of Accounts, Properties, and Views in Google Analytics.
After reading an article by JuliusFedoroviciousfrom Analyticsmania.com titled Google Analytics and Google Tag Manager Naming Conventions (2021), I learned that a good rule of thumb to follow when creating a GA account is that one (1) account is for one (1) company/client. For the project I was just on, we had the client create an account and added our team as a user. This is also mentioned in the article as a best practice, instead of creating a company account and adding containers for each client/company. For example, if I wanted to create a new account, I could use our company’s name, Withum. When I make the connection in Power BI Desktop, I can easily find and navigate the Withum account in the Navigator window.
The next level where naming conventions are equally as important is properties. Not to get too in-depth here, but from what I’ve learned, the biggest takeaway is that just because a website belongs to a particular company does not necessarily mean that website should get its own property. This helps decipher what metrics and dimensions we want to pull in when we get into Power BI Desktop. For example, suppose certain websites are all connected as part of a visitor’s or user’s single experience (e.g., home page, support page, blog, etc.). In that case, they could be tracked in a single property so when imported into Power BI, the visualizations and reports can be created using fewer queries (possibly one single query), improving performance.
The final level of the Google Analytics hierarchy is the views. The view allows you to view a subset of data belonging to a specific property. In the same project I mentioned before, every view was kept with the standard title of All Web Site Data. It was helpful that the property was appropriately named, and we were only working on one property and its corresponding view. However, if I had to use more than one view of the property, things could get ugly, even worse, time-consuming. A good rule of thumb is to include the property name when naming the view. For example, I could name my view, Withum – Main View. Or, Withum – Test View. When I am ready to connect to Google Analytics in Power BI Desktop, I can easily distinguish which property has the right view that contains the data I am looking to analyze.
My hope for this blog post is to keep the conversation going (or even spark new ones) on effectively leveraging the power of Power BI + Google Analytics to expand your data even further. Not to say the Google Analytics team didn’t do a great job with this tool, I just believe that Power BI is a much better tool for visualizing your website data into interactive reports and dashboards. Thank you for taking the time to read through this blog post.
If you have any insights you would like to add or share, please do so, as it helps us all learn and explore different options. If you see anything here that is inaccurate or incomplete, please comment back to correct it. Nothing is worse than leading someone down the wrong path. Thank you.
Author: JP Ovalle, PMP
InsightWhale Team. September 9th, 2020. References https://insightwhale.com/google-analytics-setup-best-practices-and-guide/
Technology and Digital Transformation Services