Google Data Studio: 5 Advanced Techniques to Boosts Your Visualisations ( Part III)
In my last articles, I talked about Filters, Control, Segment and Calculated Fields. So for this article, we're going over the Blend to see what it is all about, as well as some bonus tips!
What to expect:
This article will show you how to join or "blend" data in Data Studio. You'll learn which type of "join" Data Studio uses and how you can use it to combine multiple data sets. And even how to power more complicated queries out of the same data set.
Blends
Blends (a.k.a joins) is Data Studio's version of a Join. You can blend up to five sources. Blends are always a Left Join, and you can use these for various purposes, i.e. combine two different data sources ( Google Analytics and Google Ads).
Join key
When looking at your join key, you need to think about what fields are in both your data sets that Data Studio can use to connect them? And it could be something like DATE, where it could be the only commonality used as the join key.
Blends are always a LEFT JOIN. If it doesn't find data in the far-left data set, it won't exist. You'll have to keep in mind that the most complete data source will place on the left so that the data is available across the joined data set. If your data is not available on the left, the Blend will not work in this case.
Creating a Blend
You can create a blend by either:
- Select multiple charts, right-click "Blend data". For example, blend the session from GA and GAds or blend the cost from GA with Cost from GAds.
- Or + Blend Data under Data Sources from scratch.
Name your blends
We would highly recommend making sure you name your Blends. Having an organised and systematic naming convention helps you to understand and remember your components of blend data easily, especially when there are dozens of Blends in place. You can name it as BL: ( Source 1 ) + ( Source 2).
One thing that you want to be careful with is your join keys. It would be best if you were intelligent with the design of your Blend. If you are trying to join any information in your thought, i.e. the date and Source Medium, you'll end up receiving a wacky chart. ( see below) So make sure you combine the data logically. If you spot something weird, the first you need to look at is your field in your join.
Besides your join keys, you need to think about what data you want to display and design your Blend accordingly because there may have multiple blends in Data Studio.
If you want to display a field, you need it included in your blend. Think of your blend like a new data set — you can’t create a visualization if the data set doesn’t have the field you need!
It is also always recommended to join raw numbers and then calculate the conversion rates.
Benefits of Blends
Remember how your Calculated Fields don't let you mix dimensions and metrics? Blends (kind of) does. To give you quick examples:
- Unique Events where Event Action = Contact Sales
You want to know the percentage of site traffic click contact sales, and you have a goal for contact sales. You would divide your goal with the session. But if you don't have a goal, you can use Blends for this calculation by having one chart with total sessions and another chart with sessions filtered by contact sales action. The blend is powerful because you can segment or filter one data set and not the other and combine them.
- Blend with Sessions…
The examples given above are what we would call a Self-Blend by blending the same data source with itself to create that calculation.
Another great use of Blend is the ability to create a chart of multiple timeframes. So here, you have three different data sources coming from data sets. Column one gives me all of the data within the last 30 days, column two pulls in the data over the previous seven days, and column three gives you yesterday's data. You can build a little funnel visualization of 30 days, seven days and one-day active users and create calculations of the percentage between them.
Bonus tips — Changing Data Sources.
Learning how to change the data behind a data source ( without losing all your calculations) is quite essential, where it will save you a lot of time and pain when you suddenly need to do so. There may have various reasons why you want to change the data source:
- The GA view you're using.
- The BigQuery table you're pointing to
- The spreadsheet you're pulling from
There is no right or wrong if you want to create a new data source from scratch. But, keep in mind that if you have dozens of calculated metrics and custom metrics, you'll have to recreate them ALL. So instead, we'll genuinely suggest you copy the data source and edit the connection, which will preserve all calculated fields you have created.
First, click the copy icon on the left of your create report, confirm you want to COPY DATA SOURCE, then the page will load to bring your data collection point. Now, you get the choice to choose what view you would like to take and hit RECONNECT and APPLY.
And now, you will get a copy like what's shown below. Make sure you rename it to understand which raw data is coming from. Then hit SAVE, and you're done!
Personalize the charts and tables with your own data. We hope you found this article helpful, and we can't wait to see what you create! If you have any questions or need some help getting started, please don't hesitate to reach out or visit CXL for more.
