Validation with Grouping Characteristics

Use

In certain situations, it may be useful or even necessary to not apply the rules you have defined in a validation method per diem to all of the data, but rather to first split up the data according to specific criteria and then to execute the validation for each data subset.

Example Example

One case in which this could be conceivably beneficial is when a validation confirms that the assets and liabilities of the balance sheet of a consolidation group have a zero balance even though the detailed validations of individual consolidation units uncover remaining amounts that, when totaled, also balance to zero. Of course, in such a case you first would have to balance the balance sheets of all consolidation units to also balance the assets and liabilities at the consolidation group level.

End of the example.

Thus, validation of data subdivided according to certain grouping characteristics allows you to perform detailed checks with regards to technical consistency and accounting consistency.

Prerequisites

The grouping function is supported only for validation methods with the type Data Rows are Totaled .

Features

You can use the grouping function in individual validation rules within a validation method. In the list of available characteristics, you choose the characteristic by which you want to group the data. If attributes are assigned to the characteristic, you can use the attribute values for the grouping, as well.

You can choose more than one grouping characteristic. If you do this, the system splits the data into as many subsets as there are unique combinations of all values that exist for each grouping characteristic. Then the system consecutively performs the validation rule for each data subset.

The system shows a positive overall check result only if all of the subset checks produce positive results. If one or more subset checks produces a negative result, the system regards the overall result as negative. If no data or only zero values are found for a given grouping characteristic value (or a combination of multiple characteristic values), this returns a positive check result.

The following applies when validation rules have one or more grouping characteristics: Up to four variables are available (as is also true otherwise) for the different messages issued according to the check result. However, here the system always substitutes the fourth variable with a concatenated string of the values of the grouping characteristics used for the current check. For more information, see Use of Message Variables .

Note Note

As described above, using grouping characteristics in a validation rule causes the system to execute that rule once for each unique characteristic value (or unique combination thereof). This increases the runtime of the validation accordingly. As a rule of thumb, the validation check for each characteristic value requires as much processor time as the processing of a complete validation rule without grouping characteristics.

End of the note.

Example

You want to use validation to check whether the cumulative value of balance sheet item A is greater than the cumulative value of item B - both in local currency respectively. In addition the check result across all of the data, you also want to see the results differentiated by the country of each consolidation unit. The data basis contains the following totals records to be checked:

Totals records

Country	Item	Value in Local Currency
USA	A	700
USA	B	900
Germany	A	500
Japan	B	0
Japan	C	200

You have defined the following rule:

LC_YTD (item A) > LC_YTD (item B)

You choose Group and add the characteristic “Country” to the list of grouping characteristics that has appeared. Now, the rule is executed once for each group of data records that have the same value for characteristic “Country”, rather than across the aggregated values of all five data records.

Since the data records have three different countries, the system returns the following three results:

Checks with grouping by country

Country	Value of A	Operator	Value of B	Result
USA	700	>	900	FALSE
Germany	500	>	N/A	TRUE
Japan	N/A	>	0	TRUE

The negative partial result for the grouping “USA” causes the entire validation method to return a negative result.

In contrast, the same data would produce the following result without grouping by country:

Check without grouping

Value of A	Operator	Value of B	Result
1200	>	900	TRUE

This means that without the grouping, the validation method returns the opposite result.