Skip to main content
Documentation
Documentation
Administrator Guide
Last Updated: 2023-06-23
Concept: Dataset Example Data

Concept: Dataset Example Data

When you're editing dataset transformations, you can view how your changes affect an example of the data. This view enables you to see how different stages impact your data.
You can change the data displayed by selecting any of these example options:
Example Option
Description
Security Requirements
No Example
No data displayed. This option has the fastest load time.
Dataset Editor
 permission or better on the dataset.
Default Example
Workday displays a subset of the data, enabling you to assess how different stages affect a subset of your data.
Dataset Editor
 permission or better on the dataset.
Custom Example
This option is only available for derived datasets. You can create rules with different filter criteria for different subsets of your data. These rules enable you to curate what data you see and how each stage impacts it.
You can add 1 rule per base dataset or table used in the derived dataset. The rules you define are specific to this derived dataset and don’t affect other datasets with the same base dataset or table. When adding a filter stage, you can add more complex filter criteria similar to the
Filter
 stage.
  • Dataset Editor
     permission or better on the derived dataset.
  • Any of these security requirements:
    • Dataset Viewer
       permission or better on the base dataset.
    • Table Viewer
       permission or better on the table.
If you have permissions to view the
View Dataset Transformations
 report, then you can change example data but won't be able to save.
Workday displays up to 20,000 rows when you select either Default or Custom Example.

Comparing Example Data to Published Data

Workday displays example data on a limited number of records, which typically is less than the number of records published in a dataset. If the dataset uses a Join stage and 1 or more Prism calculated fields after the Join stage, the example data displayed might be different than the published data. The published data is correct because it works on all data from the sources.
Example: Your dataset includes a Join stage using a right outer join and a Prism calculated field with a CASE function that evaluates the value in a field from the secondary (right) pipeline. The example data might not find a match and return NULL for the field value, whereas the published data finds a match and returns a non-null value. As a result, the CASE function returns a different value in the example data than in the published data.