Outlining the Cash Application Add-On for Financials Contract Accounting

To train the Machine Learning model before productive use, historic payment data needs to be sent to the platform.

• Based on this data, the intelligent Machine Learning logic will be set up and matching proposals will be made.
• The Machine Learning platform permanently learns based on the available data and automatically adapts the Machine Learning model.

Thus, the system is able to react to changing conditions in the system and to keep the model up to date. Additional configuration effort to adapt the existing logic is not necessary anymore and the automation rate can be kept on a high level. This improves days sales outstanding on a sustained basis without the need for further investments.

Cash Application is running on the SAP Business Technology Platform and can be connected to any existing SAP S/4HANA system.

• Existing configuration and payment processes will continue to be used and enriched with Machine Learning functionality.
• This allows the implementation of this tool at a reduced TCO and time for value.

The cloud based architecture allows shared services to scale according to the business growth and financial organizations will be empowered to focus on more strategic tasks and service quality instead of manually clearing incoming payments against invoices.

It is used in addition to the standard clearing process in FI-CA using the clearing rules. Payment processes will be realized in Contract Accounting based on the configuration and the clearing logic that has been set up in the system. Items which cannot be cleared based on this logic, result in clarification cases. These clarification cases usually require manual effort and analysis for clarification.

These cases will be submitted to SAP Cash Application and the Machine Learning logic will evaluate the data in order to find appropriate open items for clearing. Depending on the result and level of confidence, items can then be cleared directly or a clearing proposal for manual review will be submitted. The confidence level can be customized dependent on the individual need.

Business Benefits of the SAP Cash Application Add-On for FI-CA are the following:

• Existing solutions require substantial implementation and maintenance efforts to keep good automation levels. The SAP Cash Application uses Machine Learning to keep the model up to date and therefore does not require maintenance efforts.
• Applying Machine Learning boosts process efficiency by automating tedious and repetitive manual tasks, improving both service quality and accountant productivity.
• SAP Cash Application increases the automation rate in addition to the classical approach with no additional configuration effort.
• Existing rule-based approach is still used but Machine Learning can be used in conjunction to reduce manual cases to an absolute minimum.
• The high automation rate to which customers in FI-CA are already used to, will even be improved and provides even higher process automation and excellence compared to the classical approach.

IMPORTANT: From 15 October 2023, onwards, Cash app will only be deployable for RISE and GROW customers. Existing customers are not affected by the commercial change.

SAP Cash Application is a cloud service running on the SAP Business Technology Platform that integrates with S/4HANA BRIM On-Premise or S/4HANA Cloud for Contract Accounting and Invoicing.

The historical clearing information is sent from BRIM to the SAP Cash Application in the Cloud to train the model and derive matching criteria. This training of the model happens regularly to ensure changing behavior is captured so the model can adapt without configuration changes.

When new bank statements are received - most times on a daily basis - the ones that are not processed by the standard clearing rules will be sent to the service along with the open receivables and the Machine Learning model can infer matching proposals.

The Machine Learning approach can capture much richer detail of customer and country-specific behavior, without the costs of manually defining detailed rules.

Proposals are returned to SAP S/4HANA and those that meet the configurable confidence threshold are automatically cleared for full automation. When there are multiple proposals for a payment, or the confidence threshold has not been met, they are presented to the AR accountant within the standard Fiori app that they use today for clarifying the clarification cases already.