SAP Analytics Cloud Update of Actuals in Rolling Forecast

With the rolling forecast, we would like to supplement the annual planning and bring flexibility into the planning and be able to react faster to changes in the business. In order to optimally support this process with a software application, various functions and options are required, which we can find in SAP Analytics Cloud. These include:

– Tables with special functions for a rolling forecast

– Setting the time frame (cut-over date)

– Automatic filling of the ACTUAL data

– Extrapolation functions

– Management of scenarios and versions

To reduce manual interventions and errors, it is necessary to automate tasks and process steps. Therefore, in this article I would like to show step by step and exemplarily the improvements in the process through SAP Analytics Cloud and the standards included in the product.

I would like to start with the possibilities of ACTUAL data replenishment or provision (loading). This process step can be laborious and cost a lot of time without technical support.

In SAP Analytics Cloud we have the possibility to include many systems, data sources (Connections) or files in our planning or forecast.

Picture: SAP Analytics Cloud Connections Example

Once this data is incorporated, SAP Analytics Cloud can automatically populate the forecast with the running ACTUAL.

In a departure from a plan or projection to the end of the year, our rolling plan in SAP Analytics Cloud is deployed across the year boundary. Here, we look at a rolling 14-month period into the future. The forecast goes into a re-forecast every month (RF1, RF2, … RF[n]) and is set up on a monthly basis.

The concept and a basic structure for the rolling forecast can be as follows:

Concept for rolling forecast with ACTUAL (Actual), RF1 (Forecast 1) and PF (Projection)

In each month the running ACTUAL is maintained and the forecast period is set one month further (Cut Over Date). Without application support, this requires many manual actions. SAP Analytics Cloud can help with standard functions.

In SAP Analytics Cloud it looks like this (appearance and design is up to the artist):

Rolling forecast implemented in SAP Analytics Cloud (exemplary)

The provision of the ACTUAL data and the connection represent a first and often difficult problem to solve. As seen above, SAP Analytics Cloud offers many possibilities to connect systems. These systems can of course be SAP systems or others. In this example I show the connection to our rolling forecast to a Google Spreadsheet (see also diagram: SAP Analytics Cloud Connections).

Table with new ACTUAL data (in a Google spreadsheet connected to SAP Analytics Cloud)

To load this table, we make the structure available in SAP Analytics Cloud in our rolling forecast model.

The information in the table (columns) above, is mapped to the dimensions in SAP Analytics Cloud.

The assignment of a data source, such as the Google spreadsheet in the example, and the SAP Analytics Cloud Model is only performed once. After that, new data can flow into our model for the rolling forecast continuously. This process can be triggered time-controlled or be ad-hoc.

Screen for managing the “incoming data streams” into the rolling forecast (data management in SAP Analytics Cloud).

After new ACTUAL data has been provided, the input screens and reports in SAP Analytics Cloud are automatically adjusted and display the new data and also the new time window.

Report or input schedule as of December 2020

If we go forward one month in time and ACTUAL data from January 2021 is available, it will be taken into account. This means SAP Analytics Cloud automatically adjusts the views in our input masks and reports.

Report or input schedule as of January 2021 – The cut-over date has shifted by one month.

With the now automated provision of actual data from one or more source systems, the first step for a more easily executable rolling forecast is given.

The handling of the different forecast versions, the reporting adapted for a rolling forecast and the extrapolation of data are topics for another time.