Australian dairy manufacturer simplifies invoice matching with intelligent automation
Australian dairy manufacturer simplifies invoice matching with intelligent automation
A leading Australian food and beverage manufacturer was overwhelmed by a high volume of remittance payments which were being manually processed. The manufacturer partnered with Fortude to automate the end-to-end invoice matching and posting process. This was achieved via an enhanced credit-cash application UiPath-powered bot that integrated seamlessly with the manufacturer’s SAP ERP.
The Challenges
The customer credit department of this Australian food and beverage manufacturer dealt with a high volume of remittance payments each day. A staggering 70,505 transactions were handled manually each year out of a total of 231,809. These payments must be accurately posted against the corresponding customer and exact payment in SAP. Similarly, the team had to carefully review payments on the customer’s bank statement that were below $10,000 and post them to the correct customer account.
The customer initially used a different automation platform to automate both procedures. However, they faced challenges with high license costs and difficulty identifying payers, leading to a manual approach to posting bank statements. As a result, the exception rate for posting had steadily increased over time, requiring the credit team to do more manual work.
Solution
Document Understanding
The Fortude team leveraged a new UiPath-powered bot which integrates seamlessly into the SAP ERP to automate the end-to-end invoice matching and posting process. This significantly improved the productivity of the credit team agents.
Remittance files and invoice data come in many different layouts and templates. They can be structured documents or unstructured PDFs or image files. Using IA, the bot can process these different formats by training a ML model.
- Extract data from remittance files (Email)
- Download Bank Statements from SAP
The enhanced logic can handle multiple variations of invoice details, including customer number, invoice number, invoice amount, payment amount, and payment dates.- Invoice matching
Instead of manually determining which SAP accounts to post each transaction, the bot
automates matching invoice numbers and amounts by processing SAP bank statements and
extracting invoices sent by customers. The bot automatically posts these transactions to SAP and generates a summary output file. - Re-train the bot using Machine Learning
Continuously improve the bot to process new templates based on human input.
Impact
- Reduced incorrect payments and non-payments by matching invoice numbers and line amounts for each invoice.
- Reduced the manual effort required, which helped the resources to focus on higher value tasks.
- Enabled the extraction and interpretation of a variety of documents – This allows for the processing of structured and unstructured documents with varying formats and components.
- Improved accuracy over time by retraining bots to read and learn new document templates.
Download Case Study
Subscribe to our blog to know all the things we do
