Intelligent Document Processing with MuleSoft RPA and AWS
In today's fast-paced business environment, companies are looking for ways to streamline operations and increase efficiency. Manual document processing — whether it be invoices, resumes, or customer data — can be time-consuming and error-prone, creating operational bottlenecks and inefficencies. With MuleSoft RPA and AWS, companies can streamline their operations by using intelligent automation to increase the efficiency and accuracy of their document processing.
Sadhana Nandakumar (00:11):
Organizations are looking to innovate every corner of their business to improve efficiency and reduce costs. Teams spend about 10 hours a week on repetitive tasks, mostly extracting and processing data from documents. As per Gartner, by 2025, 50% of the B2B invoices worldwide will be processed and paid without manual intervention. MuleSoft RPA provides the ability to automate routine workflows at speed and scale anywhere across your business, saving time and cost.
Meet Mary, an accounts payable person at Acme Corporation. Mary and her team spend a substantial amount of time every week processing invoices from vendors. Vendors email invoice documents to Mary's team, which they then extract, validate, and update on the legacy management system. Additionally, it is also possible for vendors to send in invoices in other languages during which her team has to work with the translator to extract and update the information. On average, this process takes about 30 minutes a day, and Mary and her team receive about 50 invoices per week while the team is very skilled at their work. The expansion of business by Acme Corporation has added pressure on Mary and her team leading to costly manual processes and human errors. MuleSoft RPA's intelligent document processing capability can help automate all of these repetitive tasks and let Mary and her team focus on other activities that need intervention. Let us now look at this use case in action.
A vendor sends the service invoice in Spanish via email to Acme Corporation. As soon as the email is received the automation downloads the image, extracts the data, and updates it on the legacy account management system. Notice that the invoice data is also translated into English before the update. The entire process, including validation on the invoice, was performed automatically in less than a minute.
Let us now look at how the data was ingested, extracted, analyzed, and updated on the target system. MuleSoft Composer can pick up the new email event and call our intelligent document processing capability to extract and update the information. This is hard to reach by APIs. RPA Builder's Mail Operation Toolbox item allows developers to quickly build and deploy an RPA bot that provides the ability to read the email and download the attachment for document processing once the data is available. The AWS document processing toolbox tools can be used to analyze and translate the documents contents.
The ERP docs with AWS toolbox element makes use of the underlying AWS text extract APIs to analyze the document and return the data in raw text or GS o format. Amazon Text Extract uses AI to extract text, handwriting, and data from scan documents. It uses multiple machine learning models to classify various documents and intelligently analyze them. Once the data is extracted, it can be translated to English using Amazon Translate. Amazon Translate is a text translation service that uses advanced machine learning technologies to provide high-quality translation on demand. It can auto-detect the source language and translate it to the target language. In our case, we are translating the invoice list from Spanish to English. Notice that we are adding a gateway condition to check if the invoice image is valid. An email is automatically sent to the vendor for correction if it is not a valid invoice.
Finally, let us look at how the data gets updated in the legacy management system. RPA Builder provides toolbox items to perform screen scraping and data entry. In our case, by using the appropriate export expressions, we are able to update the data extracted from the invoice to the legacy account management system interface. MuleSoft automation has provided the ability to automate all the steps of document processing workflow making invoice processing easy to manage and scale. Let us assume that after a few years, ACME Corporation decide to modernize its legacy systems and is looking to move to Financial Services Cloud as its account management system. With a new composer flow, this automation can easily be reused to extract invoice records and sync with Salesforce. Notice that a new invoice is created on Salesforce. With the extracted invoice details, the automation asset that's created can also be shared and reused in the scope of a larger business process.
With the flow RPA capabilities, it is possible to directly invoke the RPA bots from a Salesforce flow. For instance, it is possible to invoke the intelligent document processing capability from a service agent's screen to extract the invoice information. With a push of a button, it is now possible to extract the handwritten invoice data via the RPA process that we already created. Manually processing documents can be time-consuming, costly, and error-prone. Intelligent document processing solutions provide the ability to extract and use meaningful data to automate business processes rapidly. In this use case, we saw that the document processing capabilities reduced the time and effort to less than a minute per document. Mary and her team can now focus on the tasks that need more manual intervention. By bringing together the power of MuleSoft Automation and AWS, it is possible to automate document processing accurately, thereby saving operational time and cost.