Banks & Financial services companies – investment firms, commercial or retail banks – are awash in all sorts of documents, from customer onboarding forms to mortgage applications, ISDA Master Agreements, SEC 10-Ks and more. As part of their digital transformation efforts, they’re looking to automate the processing of financial documents – potentially millions of them. But financial firms are running into roadblocks when it comes to financial document automation 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 much like a human does. That changes the automation equation, enabling banks to move beyond robotic process automation and bring immediate value to financial services firms.

Download the Gartner Report: 2020 Market Guide for Text Analytics

Download Now

Financial Process Automation Key Benefits

image shows that indico's intelligent process automation can lead to a 4 times increase in process capacity

Capacity expansion up to 4x:

growing revenue without growing expense

image shows that Indico's intelligent process automation can lead to an 85% reduction in process cycle time

Cycle time improvements:

process documents 85% faster

image shows that indico's intelligent process automation can lead to an 80% reduction in the amount of total resources required

Increase Efficiency:

free up employee time for higher-value initiatives

Knowledge Capture:

Codify and streamline processes

Compete Effectively:

With stalwarts and fin-tech startups alike

Customer Satisfaction:

Improve turnaround times

knowledge workers automating manual document-based workflows

Problems with Early Attempts at Automation in Financial Services & Banking

Banks and Financial services firms for years have been trying to automate document-intensive workflows such as the mortgage origination process, but with limited success. The reason: because most processes deal with documents that contain unstructured content.

Early financial services 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 financial institution 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 so, as soon as a new document type came along, or an existing document format changed, the automation would break down. Money wasted.

Banks and Financial services firms for years have been trying to automate document-intensive workflows such as the mortgage origination process, but with limited success. The reason: because most processes deal with documents that contain unstructured content.

Early financial services 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 financial institution 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 so, as soon as a new document type came along, or an existing document format changed, the automation would break down. Money wasted.

Learn How Chatham Financial Increased Process Capacity by 400%

What About OCR and RPA?

Other early attempts at automation in financial services document 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 useful information, and all the issues that presents.

Robotic process automation in the banking and financial services 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, say, a mortgage processing system. That would be a process that’s ripe for automation using RPA.

But that’s not at all how processes tend to work, especially those dealing with unstructured content, which is most of them. With unstructured content, a human has to read the document and make judgment calls about it, including what data to extract. (RPA can, however, complement IPA in a financial services automation solution, as we discuss briefly below and in detail here.)

Other early attempts at automation in financial services document 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 useful information, and all the issues that presents.

Robotic process automation in the banking and financial services 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, say, a mortgage processing system. That would be a process that’s ripe for automation using RPA.

But that’s not at all how processes tend to work, especially those dealing with unstructured content, which is most of them. With unstructured content, a human has to read the document and make judgment calls about it, including what data to extract. (RPA can, however, complement IPA in a financial services automation solution, as we discuss briefly below and in detail here.)

Intelligent Document Processing for Financial Services and Banking

Intelligent document processing technology, as implemented by Indico, is fundamentally different from OCR, RPA and templated approaches because it can understand document context much like a human does. That’s because Indico’s Intelligent Process Automation (IPA) 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 digital process 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. But even the largest financial services company or commercial bank would be hard-pressed to come up with that much data and effectively train its own models.

Indico uses an AI technology known as transfer learning to create custom models that can tackle virtually any downstream task – including customer onboarding and other common financial services and banking processes. 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. (For a deeper dive on this point, check out our Intelligent Process Automation page.)

Intelligent document processing technology, as implemented by Indico, is fundamentally different from OCR, RPA and templated approaches because it can understand document context much like a human does. That’s because Indico’s Intelligent Process Automation (IPA) 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 digital process 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. But even the largest financial services company or commercial bank would be hard-pressed to come up with that much data and effectively train its own models.

Intelligent Automation in Financial Services: Solving Real Business Problems

Intelligent document processing in financial services and banking can address a number of common processes, including the following:

Meeting the LIBOR challenge

The LIBOR interest rate benchmark is due to be phased out at the end of 2021, meaning commercial banks and other financial institutions need to find any loans that reference it. They could have a team of humans pore over paper documents looking for relevant terms – or train an IPA model to do it for them. An IPA 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. (For more detail, read our blog post : “Don’t Labor over LIBOR: Meet the Looming Deadline with Intelligent Automation.”

Streamlining Financial Document Analysis

Investment firms rely on data in SEC earnings reports to inform their analysts’ investment advice. Analysts must study quarterly 10-Q and annual 10-K forms looking for actionable data, pull it from the reports and enter it into spreadsheets. An effective intelligent document processing tool could take on the job for them, freeing up more time to actually analyze the results. (See the blog post :“Bringing Intelligent Process Automation to Financial Document Analysis.”)

Processing ISDA Master Agreements

Processing over-the-counter derivatives transactions requires examining the ISDA Master Agreement that defines the terms between the two parties involved in the trade. It’s a herculean task, given the ISDA document weighs in at 28 pages and is really just a template; different variables will apply to each transaction. It can easily take a human 2 hours to process a single one and large banks may process hundreds of thousands per year – making the processing of ISDA agreements a prime candidate for intelligent automation in banking. (For more, see the post: “Process Automation Comes to ISDA Master Agreements.”)

Automating Commercial Mortgage Processing

Assessing the creditworthiness of an applicant for a commercial mortgage means examining reams of documentation, including W-2s and bank statements to tax returns and business plans. It’s a labor-intensive process that involves a human extracting key data points and entering them into spreadsheets or another downstream system for processing and analysis. Given the varied nature of the documents, templated tools or RPA won’t get you too far, but an intelligent document processing tool will be able to dramatically reduce processing time. (For more about this and other use cases, see: “3 Use Cases for Document Processing Automation in Commercial Banking.)

Anti-money Laundering

Commercial banks in the U.S. must comply with various regulations intended to detect money laundering. In practice, that means collecting numerous documents from clients intended to prove they’re legitimate and ongoing monitoring for any negative news about clients. Traditionally these were manual processes, but today tools like Indico’s IPA platform can automate large portions of anti-money laundering programs. (This topic is also covered in the post “3 Use Cases for Document Processing Automation in Commercial Banking.”)

Processing Trade Order Confirmations

Financial firms involved in any kind of trading know that the confirmation process can get complex and time-consuming, which is why many of them are now looking at intelligent document processing as a way to streamline the process. Any kind of trade – including over-the-counter stocks, stocks traded on an exchange, and derivatives – requires a settlement process and, ultimately, a trade confirmation. It’s an important step because the confirmation spells out the terms under which the trade was executed, so both sides can see whether the trade matched their expectations in terms of price, quantity and timing.

How IPA Complements RPA

While intelligent process automation often works well on its own, some financial and banking industry processes can benefit from the combination of IPA and robotic process automation.

RPA is effective at automating deterministic business processes that involve highly structured data. So, it’s useful for automating repetitive, labor-intensive tasks, often with greater accuracy than humans – because software robots don’t get tired. IPA, meanwhile, is able to automate processes that involve unstructured data, making it complementary to RPA.

One example of how this can play out in practice is in automating lockbox processing, which involves matching payments, in the form of checks, to invoices. An IPA tool is used to “read” invoices and checks, extract relevant data and translate it to a structured format, then hand it off to an RPA tool for input to a downstream processing system.

Financial and Banking Process Automation: Key Benefits

In this age of digital transformation, it’s imperative that financial institutions and banks take steps to achieve efficiencies wherever they can. Indico’s intelligent document processing solution is one solution that can help in that effort while delivering significant benefits, including:

Capacity Expansion:

Automation increases employee productivity up to 4x, enabling the company to increase revenue with existing headcount.

Cycle time improvements:

Automating manual processes enables companies to get work done faster, even while increasing accuracy.

Increase efficiency:

Taking mundane tasks off an employee’s plate frees up time for more rewarding work that’s also more valuable for the institution.

Knowledge capture:

Automation requires companies to codify processes that may have been performed for years with no formal agreement on how it should be done. In the process, companies are often able to streamline the process and make it more efficient.

Compete more effectively:

Automation makes organizations more competitive, taking a page out of the book of the most nimble fintech startups.

Automation Augments but Doesn’t Replace Humans

A common misconception about automation is that it will displace lots of human 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.

A recent report by the McKinsey Global Institute supports this notion. It said only about 5 percent of occupations “could be fully automated by currently demonstrated technologies.” The more likely scenario is that portions of jobs will be automated – about 30 percent of the activities in 60 percent of all occupations, according to McKinsey.

In the financial and banking world, that means your employees will have more time to dedicate to strategic efforts that can give you a competitive edge.

What’s in a Name?

What Indico calls intelligent process automation can go by various other names. Analyst firms including the Everest Group use the term “intelligent document processing” while others go with just “intelligent automation.”

Gartner lumps various automation technologies under the term “hyperautomation,” which it defines like so: “Hyperautomation deals with the application of advanced technologies, including artificial intelligence (AI) and machine learning (ML), to increasingly automate processes and augment humans. Hyperautomation extends across a range of tools that can be automated, but also refers to the sophistication of the automation (i.e., discover, analyze, design, automate, measure, monitor, reassess.)”