Data Blending in Looker Studio (formerly known as Google Data Studio) Part 2 – Types of Blends

Sergei Kogut - February 15, 2023

In our previous post on data blending, we covered the fundamentals of data blends for Looker Studio, described what blending is all about, what it can and cannot do for you, and pros and cons of blending. In this post, we are going to learn about each of the five types of blending available in this tool.

Five Types

All data blends essentially do the same thing – they join records from two or more tables into a combined table by using one or more matching rules or “join conditions.” Join conditions are also sometimes called “keys.” All this might ring a bell if you ever encounter SQL queries. But what makes each of the blend types unique is exactly how join conditions are applied.

Left Outer – the most common join that adds matching records from the table on the right to the entire table on the left, but only when the join condition value on the right (for example, the User ID column) matches that on the left. Any rows from the right table with key values that aren’t found in the table on the left are skipped.

Right Outer – same as Left Outer but moves data in the opposite direction. Only the records from the left table that match all the join conditions are added to the entire table on the right.

Inner – most restrictive join type that returns only the rows where join condition values match in both tables.

Full Outer – all records from both tables are merged, all unique rows that only exist in either of the two tables, and all the rows that are matched by their key values.

Cross – this is the trickiest join type to use properly. Cross join doesn’t use any join conditions and doesn’t require you to set any key fields. Instead, it creates a much bigger table that consists of every possible combination of all rows. This can result in extremely large datasets, so cross join should be used with caution. But in some cases, this can be a valuable tool, for example building an annual sales report that compares several metrics to their previous year’s values.

Conclusion

Data blending or joining is one of the most valuable and versatile tools in a data analyst’s toolbox. If you learn to apply it effectively to your data needs, it can unlock additional insights not otherwise accessible without a lot of manual work. It’s especially valuable when creating reports that will be used to make decisions based on real-time patterns and trends from multiple data sources. In upcoming posts, we’ll cover creative reporting solutions that make use of blending. If you are looking for help with creating new Looker Studio reports or maintaining your existing dashboards, please reach out to info@morevisibility.com.