In this post, we will look at the different join types and how you can use them effectively, but first, let’s do a little data blending recap.

What is data blending?

Data blending refers to combining data from multiple data sources to create charts, controls, and tables. Often, data blending is considered an “analyst-only” skill, since it used to require a deep understanding of SQL programming. However, Data Studio allows users to blend data using drag and drop features, and makes the entire process accessible to those without technical backgrounds.

If you haven’t explored data blending and are just getting accustomed to the concept, you can take a look at our Definitive Guide to Data Blending in Google Data Studio. The guide will help you understand the basics before you dive into advanced data blending.

Now, as of 2022, Google Data Studio has five join configurations that you can choose from while blending your charts and tables. A join configuration features an operator that defines how the matching and non-matching will be combined. It also consists of a condition that defines how the data sources are related. These join configurations include:

1. Inner Join (New*)

Inner Join is also referred to as Join. This join type will only return the matching rows from the left and right tables. It discards all data that do not meet its condition. It gets rid of all rows that do not match the join condition.   

2. Left Outer Join

The left outer Join returns both matching and non-matching rows from the left and right tables. This was the only join type available until recently. Its called left because it returns all the rows from the left even when rows on the right do not match. If a row does not match or join with a row on the right, it will return NULLs to the columns on the right. The left outer Join gets rid of rows on the right that do not join any left row.

3. Right Outer Join (New*)

The right outer Join returns both matching and non-matching rows from the right and left tables. It works asymmetrically and similarly to the left outer Join.

4. Full Outer Join (New*)

It only returns the matching rows from the right or left tables. Full in the full outer Join shows that all rows are returned even if they do not meet the conditions. Outer indicates Null to columns that do not join any row.

5. Cross Join (New*)

It returns all the possible combinations on the rows on the right and left tables.

What are Join Conditions?

Join conditions refers to a field or fields used to join two tables. For example, you can use Campaign name to join data from google ads and google analytics. Note that the field’s name doesn’t have to be the same for you to join two tables. As long as the data is the same, the fields can have different names.

How to blend data using Join Configurations in Google Data Studio?

In this example, we will use data from a group of students that take different classes. We will then use data studio to blend data and get meaningful visualizations that answer the following questions.

• Which students take which classes?
• What grades did they score?
• Who scored the highest grade in each class?
• What was the highest score in each class?

1. First, we will load the data into Google Sheets (Remember, you can use your favorite connector type or Supermetrics as well). Each sheet contains a unique table.

Sample Tables

Classes

Students

Grades

2. Launch Google Data Studio and click on Blank Report.

3. Click on Add data and select the connector that contains your data.

4. Add your tables or charts to the report one by one.

5. After adding all the required tables, click on Blend Data

6. Add dimensions of the tables using drag-and-drop.

Click join another table to add the next table.

Grades

• student_id
• class_id
• Grade

Students

• student_id
• Student_name

Classes

• class_id
• class_name

7. Rename the tables for easy identification.

8. Blend the tables by clicking configure joins.

For example, let’s join grades to students. These two tables are related because they both have a student ID. So set student_id as the join condition. Select left outer Join and then click Save.

Repeat the same procedure to join students and classes, but this time use class_id as the relationship. (Class_id appears on both tables)

9. After joining the three tables, click on save.

10. Click on view to zoom-in on the data.

That’s it! You have blended your first data using the left-outer join operator. You can use the other join operators to answer more questions or focus on specific areas of your business/data.

Once you blend the tables, you can convert the final table into charts to get more visualizations that will help you study the data with ease. You can also play around with the join operators to view the data differently. For example, the cross join operator will show you all the possible data combinations.

Differences between Data Sources and Blends

People usually confuse data sources and blends when using the Data Studio. Blending tables or charts create a new resource called blends. The main reason why people often get confused is that blends function the same way as data sources. However, there are still some distinct differences.

• Blends depend on multiple data sources for information.
• Blends are tied to the report that they are created. It means that blends can’t be used across different reports. The only way to use them in another report is by copying a report to another report.
• Blends inherit credential settings and data freshness from data sources.

Final Thoughts

The easiest way to study the relationship between two data types is by blending them. Before data blending, data analysts relied on SQL programmers to join multiple data sources to create unified charts and tables.

As tools like Google Data Studio are evolving and becoming more visually-driven, anyone can practice advanced data blending using the five join types. All you need to understand is how to create and edit data sources, and from there, you can start your data blending journey.

I hope this was helpful! If you have any questions on advanced data blending, feel free to subscribe to my newsletter or take a look at one of our courses.

Dynamic Dashboards and Data Analysis with Google Data Studio – 2022

If you want to learn how to build powerful data visualizations and unlock insights that can help you drive business results for your clients or employers, take a look at my full course on Udemy.

Google BigQuery for Marketers and Agencies

This course is designed to bridge the gap between data analysis and digital marketing, and help you become a data-driven decision-maker. It covers the more technical aspects of JOINs and data blending as well. You can access the full course here as well.

The post Advanced Data Blending and Join Operators in Google Data Studio – 2022 appeared first on Data Studio.

In this post, we will look at data blending in Google Data Studio (the 2021 Edition) and see how the functionality has evolved over the years. When Data Studio first became available, data blending was not its strongest suit, but things have changed for the better. Let’s see how.

What is Data Blending?

Data blending is the process of merging data from multiple sources to create one new dataset that can be processed, analyzed, or presented in a visualization tool such as Google Data Studio.

Large agencies and other businesses get data from many sources, and sometimes they want to bring this data together temporarily to answer a specific question or compare data relationships. Data blending tools help unify data from web analytics, spreadsheets, cloud applications, and business systems, among others.

In the past (a painful experience for the analysts out there), individuals and businesses that needed to compare data from different sources had to export each raw file from the respective source platform and then combine it in Google Sheets or Excel. Every time they needed to compare newer data, they had to start the process all over again.

Not ideal…

Steps for Data Blending

Whether you are using Google Data Studio or another tool, the core steps for data blending are mostly the same.

1. Data Acquisition

The first step in data blending is data collection. You need to collect data from various cloud and non-cloud databases, social media tracking apps, cloud, and spreadsheets.

2. Joining Data

After getting all the necessary data, now it’s time to join it together. Here is where you will use the data blending functionality in Google Data Studio.

3. Data Cleaning

You can clean and refine your data by removing incorrect and incomplete information. Then modify the remaining data, so it’s properly optimized and formatted for an accurate analysis.

Get the Keys

Nope, not the door ones. The keys are super important when it comes to data blending. Your data sources need to share a common aspect. This acts as the join key. For example, it can be a user id, product name, or page URL, among many others.

The date is the most used join key because it’s very simple to compare two things over time. Besides, date helps you to spot correlations in datasets easily.  However, choosing the join key depends on the data that you want to compare. The first thing you need to do is form a hypothesis to help you match the sources.

UPDATE: As of February 17, 2022, Google Data Studio has added the remaining data blending options. You can read more about them in our article on Advanced Data Blending in Google Data Studio.

It is also important to note, that Google Data Studio supports only the LEFT (OUTER) JOIN. In other words, in a LEFT (OUTER) JOIN of Table 1 and Table 2, the result is all the records of Data Source 1 and those records in Data Source 2 that share the same key values. Here is a quick illustration.

Why is data blending important?

Rapid Results. Non-technical users can get rapid results in key areas such as finance, sales, and marketing. For example, marketers can blend data from a spreadsheet with the profitability of a product and CRM software. Then they can easily view the products that attract most customers purchasing interest besides the ones that make the most money.

Data-Driven Decision Making. Senior management can make data-driven decisions faster using the data blending tools. This will contribute to more sales and profits because it lets them focus on products that customers like.

How to use Google Data Studio for Data Blending?

Let’s look at two ways that you can blend data using Google Data Studio.

Method 1 (Quick and Dirty)

This is a super fast and visual way to blend data. If two charts share a common joint key, then Google Data Studio will combine them into one automatically.

Method 2 (More Complex)

This method involves more steps, but it gives you more control. In summary, these are the key steps, which we will explore with an example later on:

1. Click “Resource”, then “Manage Blended Data”.
2. Click “Add a Data View”.
3. Select the data you wish to compare in the pop-up.
4. Click on “Add Another Data Source”.
5. Select a join key in both sources.
6. Choose the dimensions you want to compare.
7. Optimize your settings and click “Save”.
8. You can view your blended data by clicking on the data source.

You can start with the first method for one-dimensional blends before trying the second one.

How does blending data sources with multiple dimensions as join keys work?

I have created two spreadsheets with sample data for this exercise. Of course, you can use any data source connectors available on Google Data Studio. Follow these steps to blend your data.

Step 1: Open Google Data Studio and click on “Blank Report”.

Step 2: Choose the data source that you want to blend. We will choose Google Sheets for this example.

Step 3: Now, select your data from the list. Use the panel to navigate and find your file.

Step 4: Read the prompts first before adding the file. In this case, we will use the first row as the headers, so we’ll make sure the option is ticked.

Other data connectors may have different options to make sure that you have ticked appropriately.

Step 5: Your data will be loaded on a new page. Now click on the “Resource” button on the top, and select “Manage Blended Data”.

Step 6: A new window will appear, where you can add a new data blend. Click the “Add a Data View” blend button.

Step 7: A new page will load to add a new file or data from the same file to blend. You can add many data sources to initiate the blending. But in this example, we will only use two because we have two Google Sheets.

Step 8: You need to set up the metrics and dimensions for your data to blend automatically. But first, you need to select the “Join Keys” if Google has not selected them automatically. Simply drag and drop the join keys.

Step 9: Rename your blended data. Review your inputs once more before saving. Click the blue-button on the bottom right corner to save your work. And that’s it, and you have created your first data blend.

Optional: You can create a calculated field in the data blending configuration. 

Creating calculated fields on a blended data source can be convenient when configuring join keys if the columns in your data sources aren’t perfectly compatible

Challenges of Data Blending in Google Data Studio

This section is based on feedback from the community and includes questions that I have encountered in the workplace.

What is the difference between data blending and data Joining?

Data blending merges data from different sources, whilst data joining uses data from the same source only.

How many sources can you blend in Google Data Studio?

Google Data Studio allows you to blend up to 5 data sources in one chart. That being said, if you are using a LIVE connector, the loading time of your report can be fairly high.

Can I use data blends across reports?

Unfortunately, blended data sources belong to the report in which they were created. Also, you cannot reference the calculated fields you create in other calculated fields in the same blended data source.

Final Thoughts

Data Blending is one of the most underrated techniques when it comes to data automation. If you pay close attention to it, you can unlock some valuable insights and save yourself a lot of manual work. It can make work easy for both new and advanced analysts when identifying sales patterns or data trends. Besides, it helps senior management to make decisions faster based on real-time data that combines multiple data sources. Although there are many platforms with data blending capabilities, Data Studio stands out to be one of the best, because it’s free, easy to use, and is constantly evolving.

I hope this was helpful! If you have any questions, feel free to DM me on LinkedIn or subscribe to my newsletter for more updates ?

Dynamic Dashboards and Data Analysis with Google Data Studio

If you want to learn how to build powerful data visualizations and unlock insights that can help you drive business results for your clients or employers, take a look at my full course on Udemy.

The post Data Blending in Google Data Studio – The Definitive Guide 2021 appeared first on Data Studio.