Concept: Prism Analytics Data Management Workflow
With Workday Prism Analytics, you can analyze your Workday and non-Workday data together without having to export it into a separate data warehouse and BI (business intelligence) application. This makes analysis faster, easier, more secure, and performed in a location where users can take action on it.
What are the steps involved in going from raw data (both internal and external to Workday) to visualizations and reports in Workday? What skills do you need to perform each step? This section explains each phase of the data workflow from transforming data to analyzing it.
Phase 1: Create Tables and Data Change Tasks to Bring in Data
The first phase in the data management workflow is to bring data into the Prism Analytics Data Catalog. You can bring in external data or Workday transactional data. The Data Catalog is where data analysts can see what data is available to them.
You bring in data by creating a table. A table is a Prism Analytics object that stores (materializes) data and represents it in a tabular format. A table has a schema and contains data that's valid against the schema.
After you create a table, you create a data change task to load or change data in the table. You can create connections from Workday to external servers and data warehouses to bring in data as the source of a data change task.
Example: You create a table to contain external operational data, and then create a data change task that loads that data from a delimited file on an external server into the table. You create a schedule for the data change task to run on a recurring basis.
Example: You create a table and data change task using data in a Workday custom report.
If you're familiar with ETL workflows (extract, transform, and load), tables and data change tasks together encompass the extract logic.
Who does this phase?
| Data Administrators or Data Analysts. |
Where can I read more details?
|
Phase 2: Create Derived Datasets to Transform the Data
After you've brought in data into the Data Catalog, you create derived datasets based on tables and other datasets to transform, enrich, and join the data. Create a derived dataset to describe the processing logic that you need to prepare the data for analysis.
Derived datasets contain information on how to process, blend, and transform the data you import into them. Transforming data is an iterative process and typically involves creating multiple derived datasets based on other tables and derived datasets. You can:
- Create new fields by creating a Prism calculated field.
- Add 1 or more stages in a derived dataset that perform different actions, such as aggregating, joining, or filtering. Some stage types can change either the number of records or fields in the dataset.
Example: You can create a derived dataset that aggregates data into groups using a Group By stage, and then create another derived dataset based on this dataset that joins the aggregated data with another dataset using a Join stage.
Example: You can create a Prism calculated field to calculate the point of sales revenue per store.
From the perspective of an ETL workflow, the derived dataset encompasses the transform logic.
Who does this phase?
| Data Administrators or Data Analysts. |
Where can I read more details?
|
Phase 3: Apply Security to the Data
Workday applies no security to the objects inside the Data Catalog. External data doesn't have any security applied to it, and when you bring Workday data into the Data Catalog, Workday removes the existing security from the fields in the tables and datasets. As a result, any user who has access to a table or dataset in the Data Catalog can see all data in the objects they have access to.
Before you make any data in the Data Catalog available for analysis, you need to apply security to the data using the existing Workday security domains. Applying security to the data you plan to make available for analysis enables you to take advantage of Workday's strong, configurable security model for external data as well as Workday data.
You configure the data security by editing the data source security on the dataset or table, but Workday applies the security to the data in the form of a Prism data source (see next phase).
Who does this phase?
| Security Administrators or Data Administrators. |
Where can I read more details?
|
Phase 4: Make the Data Available for Analysis
You can decide which data in the Data Catalog to make available for analysis by creating a Prism data source from either a table or dataset. When Workday creates a Prism data source, it loads the data source with the data from the table or dataset, and applies the appropriate security restrictions to the data source, fields, records, and field values.
The way you create a Prism data source depends on the Data Catalog object:
- Table. Edit the table schema and select theEnable for Analysisoption.
- Dataset. Publish the dataset. On the View Dataset Details report, select .
Workday applies the security domains configured in the data source security for the table or dataset, or it applies the
Prism: Default to Dataset Access
security domain if no data source security was configured. The Prism: Default to Dataset Access
domain provides contextual access to a Prism data source based on your access to the underlying table or dataset.You can use Prism data sources in visualizations and reports like any Workday delivered data source.
From the perspective of an ETL workflow, enabling a table for analysis and publishing a dataset are the load part of the process.
Who does this phase?
| Data Administrators or Data Analysts. |
Phase 5: Analyze and Visualize the Data
After a Prism data source is available, Report Writers can use it to create discovery boards and reports.