Disaggregation (Top-Down Distribution)

Disaggregation (also called top-down distribution) helps you to manually enter data within manual planning. It enables you to make manual entries within input-ready queries, also for aggregated values .

Input-ready queries are used with aggregation levels; without the disaggregation, a cell is only input-ready if all characteristic values of all the characteristics contained in the aggregation level are uniquely defined. Accordingly, all the cells that contain aggregated values with respect to the aggregation level, such as totals, subtotals, and inner hierarchy nodes, are not input-ready. In order to change aggregated values (with respect to the aggregation level), these values must be disaggregated on all the data records that contribute to the aggregated value of the cell. In the Query Designer it is possible to perform this disaggregation for the following elements:

• Key Figures
• Restricted key figures

Disaggregation is only available for basic key figures with aggregation type SUM, but not for calculated key figures, key figures with exception aggregation, local aggregations and time key figures.

Disaggregation in input-ready queries can have an effect on the application modeling of the BW planning, especially in the area ofcharacteristic relationships.

More information:Input-Ready Query

Some simple examples that show how the system distributes values according to the settings made in the Query Designer are given below. The sample scenario is as follows: The aggregation level contains the characteristics: Product, Version, Fiscal Year Variant, Fiscal Year, Currency and the key figure Amount.

Product is drilled down in the rows, and all other characteristics are limited to one value. As a result, all the data required for the disaggregation is visible in the query view.

Example 1

In the Query Designer, the key figure Amount is defined by the setting No Disaggregation vorgenommen.

No Disaggregation

This result row is not input-ready. Only changing the values in the input-ready cells has an effect on the result cell.

Example 2

In the Query Designer, the key figure Amount is defined by the setting Disaggregation for distribution type Equal Distribution.

Disaggregate Entered Value, Equal Distribution

This result row is input-ready. Its value changes from 30 to 36. Value 36 is distributed equally to all three affected cells (36/3 = 12).

Example 3

Disaggregate Entered Value, Analog Distribution (Self-Reference)

The value of the result row changes from 30to 36. 36 is distributed to the cells analogously to the previous cell values, that is weighted.

Example 4

In the Query Designer, the key figure Amount is defined by the setting Disaggregate Value Entered for distribution type Analog Distribution (With Reference to Following Structure Element).

Disaggregate Value Entered, Analog Distribution (to Following Structure Element)

The value of the result row changes from 30to 36. As a result of the selected distribution type, the values in key figure Amount (Reference) are used as reference for the analog distribution of the changed value. (In this example this produces the same results as Example 1.)

You have the most flexible type of distribution when the reference column is input-ready. In this case you can enter the weighting for the distribution there.

Example 5

Disaggregate Difference To Value Entered, Equal Distribution

The value of the result row changes from 30to 36. The new value is not distributed to the affected cells; instead, only the difference between the value before input and after manual input (6) is distributed. Since distribution type Equal Distribution was selected, all affected cell values are modifed by the value 6/3 = 2.

Example 6

Disaggregate Difference to Value Entered Value, Analog Distribution (Self-Reference)

The value of the result row changes from 30to 36. The new value is not distributed to the affected cells; instead, only the difference between the value before input and after manual input (6) is distributed. The 6 is distributed to the cells analogously to the previous cell values, that is they are weighted.

Example 7

Disaggregate Difference to Value Entered, Analog Distribution (with Reference to Following Structure Element)

Amount

Example 8

All the examples have in common that the reference data for the distribution exists in the query view. In practice, however, this is often not the case, which can cause results that appear to be incorrect.  An example that contains such a supposed error is given below. The following data is assumed; all characteristics that are not specifically mentioned have a fixed value.

Disaggregation with Hidden Reference Data

The same query view as above is shown, but the characteristic Fiscal Year is not drilled down. If the result row changes from 0 to 6 and the system uses distribution type Equal Distribution, the values are not distributed equally after aggregation over Fiscal Year. This is because each product has a different number of data records and the system reads all the data records that contribute to the result.