Summary: You can link Datasets together by creating relationships between their columns. This lets you able to search through them together in the explore page and generate charts that combine columns from two or more datasets.
The typical need for joining datasets together is to lookup a primary key value in the primary “fact” table, from another “Dimension” table. For example, sales dataset can contain the ID of a shopping category, but the actual name is in another categories table that contains the IDs of all categories and their names. In this case you would want to generate a more useful chart such as “Sales By CategoryName” rather than “Sales by CategoryID”. The way to do this is by joining the two datasets together using the CategoryID column.
Adding a Dataset Relationship (i.e. Join)
To add a dataset relationship, do the following steps:
1- In the Datasets list page, click edit next to the dataset for which you would like to add a new field
2- Click on the relationships tab
3-Click on “Add new relationship”
4- In the datasets relationships popup, select the join type. Inner joins will bring all the records that exist in both datasets. Left joins will bring records that belong in the original dataset, even if there is no match in the joined dataset
5- Select the second dataset that you would like to create a join with
6- Click on +Add Key and add the columns to be joined from both datasets. You can add up to 5 key joins in the same dataset relationship
7- Click Save
8- The relationship will now be saved in the list.
Using Joined Datasets in the explore page
Once a relationship is defined, any column in the joined dataset can be looked up or searched while exploring the original dataset. Below example shows columns from the joined dataset (SuperStore) while exploring the dataset (invoices list)
Editing and Deleting Datasets Relationships
You can edit the defined relationships by clicking on the edit icon in the list. Similarly you can delete a relationship by clicking on delete