Union datasets

You can combine two datasets into a single dataset by stacking rows from one dataset below another. Unioning datasets allows you to bring together similar data from multiple sources for consolidated analysis and reporting.

In This Topic

Union requirements
Add and manage data inputs
Union Type
Unique Rows (UNION)
All Rows (UNION ALL)

Union requirements

A Union combines two datasets based on one or more common columns. To configure a union, you must specify the following:

At least two data inputs. Select a pipeline node to add its dataset to the union.
Column mapping for each dataset. Use Manage Columns if necessary.
Union Type

Add and manage data inputs

Select the datasets and specify the column matches to create a successful union. All datasets should contain matching or compatible columns so the data aligns correctly in the combined result.

Select the + button to add a new input.
Select the pipeline node that generates the dataset to use as an input. You can have between two and four inputs.
Select Manage Columns to specify which columns to include and how to match them.
1. Select the columns to include in the result dataset.
2. Map the columns for each dataset.
Note
If data types do not match, the result column is converted to text. Numeric formats such as percentages and currencies are converted to automatic numeric.

The Data Center maps columns automatically when the column names match. But when the column names do not match, you must choose the result column to use and map the columns that do not match.

The two inputs have different column names for email addresses and postal codes.

You decide to use the result names of *Email Address* and *PostalCode*, so you remap the second dataset accordingly.

Tip

Use the filter option that focuses your column selection. Select All Columns, Mismatched Columns, or Enabled Columns.

Union Type

Union Type	Results
Unique Rows (UNION)	Rows from all inputs and removes duplicates when all column values match.
All Rows (UNION ALL)	Rows from all inputs and keeps duplicates. Use when repeated rows are expected.

Unique Rows (UNION)

Combines rows from all inputs and removes duplicates when all column values match. Use this option when you need a distinct list of records.

In this example, the rows that appear in both inputs appear only once in the result.

C00393 / #0119 / Basic

C00438 / #0122 / Pro

All Rows (UNION ALL)

Combines rows from all inputs and keeps duplicates. Use this option when duplicate records are valid.

In this example, the duplicate rows appear twice in the result because they exist in both inputs.

C00393 / #0119 / Basic

C00438 / #0122 / Pro