What I will show you in this article is how to use Power BI to connect to your favorite web performance monitoring service Google Analytics. It is true that this service already has its own web interface but sometime you might want to edit your data prior to visualization or create your own reports.
Connecting to Google Analytics using Power BI
First way of how to work with data is connect directly from cloud service to a Google analytics account using authentication protocol Auth2. This way, we get connected to Google API which will send us the data.
We need:
- Power BI account (free)
- Google analytics account with data history so we can analyze it
Steps:
- Connect to Power BI account
- Choose “Get data”
- Services
- Choose Google analytics
- Log in using OAuth2 protocol
- Look at premade dashboard and create own graphs
This is certainly the easiest way. Reason for it is that basic Google analytics dashboard will be generated automatically if we connect directly from cloud power bi service. This is not the case for the second option. But this goes in exchange for higher variability.
There is useful info on Google analytics connector on webpage powerbi.com or exceltown.com
Connecting to Google Analytics using Power BI Desktop
Second option of data analyzing from Google analytics is connecting through Power BI desktop app. Link to download here
Steps:
- Install and launch power BI desktop
- Get data
- Choose Google Analytics source
- Choose dimensions and metrics of our interest
- Choose visualizations
Downsides of Connecting to Google Analytics API
There is a set of restrictions (maximal number of sessions etc.) The most important restrictions concerning reporting architecture are those:
- High-traffic web pages can cause google analytics API to sample the data. In cases like that, only thing you can do is to shorten the observed period or to download the data daily to a data storage and follow that by reporting via Power BI or else.
- Restriction of dimensions (7) and metrics (10) which can be chosen in 1 API call. It is therefore important to think everything through in advance. This restriction can make reporting architecture very complicated. (you need to do more views)
