Intelligent Document Processing for Commercial & Retail Banking - Indico Data

Intelligent Document Processing for Commercial & Retail Banking

IDP automation: Banking use cases and benefits

Chatham Financial increases process capacity by 300%. Read the case study

On this page, learn about commercial & retail bank process automation topics including:

Commercial and retail banks are awash in all sorts of documents, from customer onboarding forms to commercial and retail loan applications, mortgage origination and refinance applications, and more. As part of their digital transformation efforts, banks are looking to automate the processing of these documents – potentially millions of them.

But banks are running into roadblocks because most of the documents contain unstructured content, which automation tools that rely on keywords, rules-based methods and templates simply can’t handle. Intelligent document processing systems, however, use artificial intelligence technology that enables them to “read” unstructured documents just like your employees do – except far faster. Employing AI in commercial and retail banking changes the automation equation and brings immediate value.

Early Attempts at Automation in Commercial & Retail Banking

Commercial and retail banks for years have been trying to automate document-intensive processes, but with limited success. The reason: because most processes deal with documents that contain unstructured content.

Early bank process automation attempts relied on systems based on keywords, rules, or templates, all of which fall short when confronted with unstructured content. Such approaches would have to account for every possible document type a bank may come across – an unlikely scenario given the range of documents and content involved in many processes.

Consider customer onboarding. It requires all sorts of documents to get a new customer set up correctly, including the bank’s own application forms, perhaps tax returns, statements from other institutions, credit reports and the like. You would have to spend millions paying consultants to come up with templates or writing rules for every possible document type you may encounter. Even if you succeeded for a time, as soon as a new document type came along, or an existing document format changed, the automation would break down. Money wasted.

The shortcomings of OCR and RPA

Other early attempts at commercial and retail banking automation included optical character recognition (OCR). OCR is machine learning technology that can be used to convert documents such as PDFs into a machine-readable format. That’s useful, but it still leaves you dealing with templates to actually extract pertinent information, and all the issues that presents.

Robotic process automation in banking is another possibility but it, too, has severe limitations. RPA is great at automating processes that involve the exact same steps each time. Say, for instance, a bank data entry clerk entered the exact same keystrokes in the same order time after time into a mortgage processing system. That would be a process that’s ripe for automation using RPA.

But that’s not at all how bank processes tend to work, especially those dealing with unstructured content, which is most of them. With unstructured content, an employee has to read the document and make judgment calls about it, including what data to extract. (RPA can, however, complement IDP in some retail and commercial banking automation solutions.)

Intelligent Document Processing for Commercial and Retail Banking

As implemented by Indico Data, intelligent document processing technology is fundamentally different from OCR, RPA and templated approaches because it can understand document context much like a human does. That’s because the Indico Unstructured Data Platform is based on a model that incorporates some 500 million labeled data points, enough to enable it to understand human language and the context of a document.

Building an effective automation model requires having a large set of data to “train” on. It’s that trove of data that brings intelligence to any artificial intelligence solution.

Indico uses an AI technology known as transfer learning to create custom models that can tackle virtually any downstream task. The end result is it takes a relatively small number of documents to train the model – usually just a few dozen. What’s more, you don’t need data scientists to make it all work. Rather, business professionals on the front lines train the automation model – those who know the processes best.

Intelligent Automation in Banking: Solving Real Business Problems

Intelligent document processing in commercial and retail banking can automate a number of common processes, including the following.

Automating customer onboarding

Whenever a commercial or retail bank gets a new customer, it requires reams of documentation to get the customer set up correctly in its systems. In addition to the bank’s own account or loan application forms, it may also require tax returns, identification, proof of address, statements from other institutions, and the like. Banks may also have to reach out to other institutions to pull credit reports or transfer funds, for example.

While these forms may be in a standardized format for whatever institution they come from, the fact that a given bank will be dealing with dozens or hundreds of third party companies effectively makes all of this documentation unstructured.

With that kind of capability, financial services firms can largely automate the customer onboarding process, perhaps requiring only a supervisor to check for accuracy as a final step.

Meeting the LIBOR challenge

The LIBOR interest rate benchmark has been phased out as of the end of 2021, meaning commercial and retail banks need to find any loans that reference it. An intelligent automation model could search thousands of documents looking for LIBOR-related terms, extract relevant data from any documents it finds, and enter the data into a downstream tool to gather all the LIBOR loans in one place.

Automating Mortgage Processing

Assessing the creditworthiness of an applicant for a commercial or retail mortgage means examining reams of documentation, from W-2s and bank statements to tax returns and purchase and sale agreements. It’s a labor-intensive process that involves having employees extract key data points and enter them into spreadsheets or another downstream system for processing and analysis.

How IDP Complements RPA

While intelligent document processing often works well on its own, some commercial and retail banking processes can benefit from the combination of IDP and robotic process automation.

Automation Augments but Doesn’t Replace Humans

A common misconception about automation is that it will displace lots of humans from their jobs. In our experience, that is not the case. Rather, automation is used to augment employees, relieving them of the most repetitive, boring aspects of their jobs so they can focus on more important matters.