In yet another example of an automation use case where robotic process automation (RPA) alone cannot get the job done, this week we bring you an example from the financial services realm: lockbox processing. It’s an example of a process that requires an intelligent automation tool to “read” documents that include unstructured content, in this case invoices and checks, before handing off to an RPA tool for additional processing.
Similar to a post office box to which companies have customers send correspondence, a lockbox is a service offered by commercial banks whereby companies can have their customers send payments to the bank. For a fee, the bank takes care of matching each payment to an invoice, helping to streamline the accounts receivable process for its client company.
RPA alone is not enough
The lockbox process involves two basic types of documents: invoices and checks. Invoices from different companies usually look quite different from one another, of course. The same goes for checks coming in from customers. Lockbox processing involves taking the incoming checks and matching them to one or more invoices.
With thousands of checks and invoices, the process is highly time-consuming and tedious for a human. That’s why commercial banks have been keen to implement automated lockbox processing.
But given the varied nature of invoices and checks, banks have quickly found that RPA software, or tools that take a templated approach to automation, are not effective. Such tools only work with highly structured content with no differentiation from one document to the next.
Lockbox automation: intelligence required
To effectively automate lockbox processing requires a tool with enough intelligence to be able to “read” the invoices and checks like a human would. It must be able to understand and extract unstructured and semi-structured data from invoices and checks, normalize the required data, and map it to a downstream processing system so payments can be reconciled.
An intelligent process automation tool uses artificial intelligence technologies including machine learning and natural language processing to be able to identify the relevant elements to extract from each invoice and check. It then puts them into a structured format.
Once the required data is in a structured format, then an RPA tool can be used to enter the data to a downstream processing system. That makes automated lockbox processing an example of a process where IPA complements RPA.
Effective tools deliver big payoff
One of the keys to success with financial services automation projects (or for any other vertical) is having a tool that’s accessible enough for business people to use, those who are on the front lines and understand the process inside and out. Indico’s IPA platform is one such tool. It’s based on a model that includes more than 500 million data points, enough to provide context to most any document. It’s a simple matter for business people to use the tool to label the various data points that need to be extracted. It takes maybe 200 documents to train the tool; after that you can expect extremely high accuracy and acceleration.
The payoff is immediate. Financial services companies can accelerate cycle times by as much as 85% using IPA and increase the throughput of their teams by 4x, freeing up significant cycles to be spent on higher-value activities.
To learn more about how Indico can help you automate lockbox processing, download this white paper from experts at the Everest Group, “Unstructured Data Process Automation.” And feel free to contact us if you have any questions or want to arrange a quick demo.