Managing Post OLAP Configuration Tasks

Upon completion of the measure and dimension configuration, the WFA on HANA instance is ready to navigate and test. Return to the Workforce Analytics page, select the Analytics tab on the left navigation bar to see the measure catalog.

1. Open the measure catalog to check that measures are available here.

2. Check results across time.

3. Check results with dimensions.

4. Confirm there are no redundant measure/dimension combinations.

1. If you receive an application error when selecting any measure – check that Measure Templates have been installed on the instance via Measure Templates in the internal SAP Admin menu. If you are not able to access this menu, raise a Support case with the instance details and request a check for Measure Templates in the instance.

2. If the measure page loads, but the tables/charts do not render, check RBP settings for your user and confirm you have the required permissions to access that measure.

Data validation will help to prove that the results displayed in metrics and dimensions are correct and require auditing aggregated counts from the "raw" data against selected measures. Data validation can also highlight data quality issues with the input data and may surface incorrect or incomplete data stored. Input measures are generally the ones chosen for auditing as they provide the basis for derived and calculated measures.

Data validation can surface data or configuration corrections that are required in different areas:

This section provides one suggested method for data validation – it will not cover all possible use cases, but the concepts can be extrapolated as appropriate for the individual WFA on HANA implementation. WFA on HANA implementers can adjust and adapt this process according to what works best for their implementation. This section is an excerpt from the Validation Guide.

1. Use Drill to Detail on the measure page to extract list of data included in metric.

   One of the myths is that the number of records returned in drill to detail should be equal to the number you drill into. This is not true. Drill to detail is used to show the data from the fact table that is behind the current measure. There could be more records in the fact table, or less. When looking into drill to detail, rather than focusing on the number of records, you should be focusing on the column that is behind your current measure.

   

   For example, if your measure is a sum of column A, you should have column A in Drill to Detail. Export it to excel and sum the number from that column. The sum should be equal to the number you drilled into.

   If your measure is a distinct count of column B, you should have column B in Drill to detail. Export it to excel and apply a distinct count algorithm on that column. You should be able to find that distinct count in excel matches the number you drill into.

   

2. Construct an Advanced Reporting/Live Data (Ad Hoc) query to extract employee (or other counts) for the same time period, using the same logic, for example, including employee statuses of A, U, P, as is configured for WFA on HANA.

   

3. Compare the two exports and create a list of discrepancies, showing which employees are in the Drill to Detail export and not the Advanced Reporting/Live Data export (and vice versa).

4. For the discrepancies, attempt to match employee’s data in Employee Central with WFA on HANA configuration and the Advanced Reporting/Live Data query.

   For example:

   • What timeline does the employee have in Employee Central blocks?
   • What data does the employee have in Employee Central blocks?
   • Does the employee data in Employee Central blocks match the Advanced Reporting query results?
   • If they don’t match, at what point does the data between Employee Central, Advanced Reporting, and WFA on HANA stop matching?

   If they don’t match, it might indicate something in WFA on HANA or Advanced Reporting query configuration needs to be adjusted. It MIGHT also indicate a discrepancy between Employee Central and Advanced Reporting data that will need to be resolved, though this is less common.

5. Correct/change Employee Central data, update WFA on HANA configuration and/or Advanced Reporting/Live Data query logic as necessary and repeat the process until all discrepancies are removed or explained.

If there is a concern with inaccurate results, try to get back to the basic, raw data. This may require you to create a lot of new columns and add them in Drill to Detail in order to see data behind your logic. Try to focus on a few selective discrepancies so you can dig into all details behind that person and analyze if the result truly reflects the data and logic. There is no quick way to investigate discrepancies.

If a base input measure doesn’t work correctly, you will need to break it down to more fundamental elements: the fields that you use in the formula.

Add all fields in Drill to Detail and see if they return some values. Verify if that value matches what you have in the logic for the base input measure. Run through the logic based on the data you can see and simulate what the Data Factory would do if you attempted to build the measure. This may require you to remove all conditions, exclusions, and simply to start with the field that you want a sum or distinct count.

Workforce Analytics Drill to Detail (DTD) enables appropriately permissioned users to 'drill through' to the individual employee records that comprise the selected result.

Drill to Detail is available in table/pivot views by default on standard Measure pages and can be made available by report designers in table views on custom report pages. When viewing a measures result in a table, Drill to Detail is available if the cursor changes to a hand (indicating a clickable link) when hovering over an individual result. In the example the Drill to Detail results will show the employee records that comprise the Average Age of 34 for Physically Challenged employees in Australia.

Drill to Detail is inherited from configuration that has been imported from another instance. This means that if a user makes changes to the DTD configuration and then imports configuration via the WFA on HANA import/export function, DTD configuration will change to reflect the DTD configuration that is in the template. If you import configuration from another instance (or from a template), check the DTD configuration after the import to ensure that it remains configured as required for the instance.

Drill to Detail will automatically exclude columns from the view if they are not applicable for the selected measure. For example, if the Recruitment Source dimension columns are included in the DTD configuration, they will only be displayed when using DTD on Movement In type measures. They will not display when DTD is used on certain other types of measures like Headcount measures.

To open the WFA on HANA Admin tool, navigate to Analytics Admin and select the tool.

1. Select whether to configure the DTD display for Measures.

2. Key columns that can be included in the DTD display.

3. Dimension columns that can be included in the DTD display.

4. Attribute columns that can be included in the DTD display.

5. List of columns currently included in the DTD display.

6. Edit (relabel and/or set BizX User column) and Remove column(s).

7. Change the order the columns appear, by using Move Up or Move Down.

8. Choose how the records in the DTD results are sorted.

Mark a column as a sensitive field for read logging.

Select from key columns, integral to the measure results.

PERSON_ID: Displays the column chosen as the Primary Person ID in the WFA on HANA implementation. In a typical implementation, this is the employee's Person ID from their Employment Information block.

USERS_SYS_ID: Displays the column chosen as the Secondary Person ID in the WFA on HANA implementation. In a typical implementation, this is the employee's Users Sys ID from their Job Information block.

Effective From Date: The date at which the employee’s record starts. Typically, this is sourced from the Effective Start Date in the relevant table.

Effective To Date: The date at which the employee’s record ends. Typically, this is sourced from the Effective End Date in the relevant table.

Current Measure: The results for that employee record for the measure that the user is displaying Drill to Detail results for.

Select from columns related to each dimension. The list of dimensions available will change in accordance with what is currently implemented in the instance.

ID: Displays the ID/Code applicable to the employee record for the dimension. In a typical implementation, this is the object's External Code.

Name: Displays the Name/Label applicable to the employee record for the dimension. In a typical implementation, this is the object's External Name.

Parent ID: Displays the ID/Code of the parent for the ID (if the dimension ID is part of a multi-level/parent child structure).

Parent Name: Displays the Name/Label of the parent for the ID (if the dimension ID is part of a multi-level/parent child structure).

Level X ID: Displays the ID/Code of the selected Dimension Level for the ID (if the dimension ID is part of a multi-level/parent child structure).

Level X Name: Displays the Name/Label of the selected Dimension Level for the ID (if the dimension ID is part of a multi-level/parent child structure).

The Level X ID/Name columns in a dimension allow you to display that dimension’s structure as part of the DTD results.

For example, you can see that the employee in the first record (User Sys ID = 100095) is in the Philadelphia location, which rolls up through Newtown Square > Pennsylvania > United States.

The list of columns that are attributes to related dimensions. The list of available attributes will change in accordance with what is currently implemented in the instance.

Code: Displays the ID/Code of the attribute for its related dimensions.

For example, you see that these two employees (User Sys IDs = 107026 and 107023) have Pay Component code BASE_CN and Frequency code of MON for their Total Workforce Annual Salary.

Drag and drop to add new columns to the DTD display.

If adding a Level X column, choose which level of the dimension you wish to use.

Select a column and choose edit columns to change its display label.

Relabeled columns will show a Column Display Name.

Select the column that contains the employee's User Sys ID, select Edit Columns and check Is BizX User ID to enable the Profile Preview on that column in the DTD results.

This column will now appear with the Profile Preview link in the DTD results.

Change the order of columns in the DTD display using Move Up and/or Move Down.

Choose which columns the DTD display will be sorted on in Order Drill to Detail Results.