Johannesburg – August 17, 2016 – A global bank headquartered in Western Europe began automating its lending business in the mid-1990s. In the beginning, it focused on one aspect of the overall process—generating loan contracts and collateral documentation. The bank first deployed a DOS-based system, but by the year 2000, had upgraded to the industry-leading HotDocs. In stages over the next several years, the bank made HotDocs available to 9,000 global customer representatives as a web application.
The HotDocs factor
From an efficiency standpoint, HotDocs offered the bank a dramatic improvement over its previous, manual approach, which involved a loan expert creating a list of document specifications, which were then passed along to the word-processing pool. The pool would manually create the documents, which were then passed back to the loan expert for editorial review, following which, the documents were sent back to word processing, and so on. While the approach eventually yielded transaction-ready documents, it was expensive, time consuming and prone to human error. In contrast, HotDocs enabled any customer representative to simply enter the transaction-specific data into a single interview and then, in seconds, generate the contract and collateral, which could then be reviewed by an expert before execution. Not only was this process much faster and less expensive, but it also yielded higher-quality documents with less risk, a benefit that correlated with the reduction of human interaction in the process.
Automating the rest of the commercial loan process
In addition to the reduction in risk of having a controlled, automated process, along with improvements in efficiency and quality of documentation that HotDocs afforded the bank, bank management wanted to automate other aspects of the lending business. Data routing had yet to be automated, and the approval process was difficult, involving the manual synthesis and analysis of hundreds—or in some cases, thousands—of discrete data items. To solve these and other issues, the bank created and deployed a proprietary workflow platform. This platform provides a framework for the bank’s lending business by organising it into a structured process (workflow) that gathers data in stages, routes the data for management review, provides essential data analysis and flagging, and generates transaction-ready documents.
The bank’s workflow begins with the opening of a case file and proceeds with the gathering of basic case information—the name of the business requesting the loan, the amount of the loan, etc. The workflow also references both internal and external credit systems to build the profile of the loan applicant. The next stage of the workflow involves the gathering and analysis of case-specific data. For this functionality, the bank built another proprietary technology that it dubbed Expert Loan, a system that gathers highly structured, loan-specific data and analyses it for compliance to predefined standards.
Built on the HotDocs platform, Expert Loan is an analytical, data-gathering sequence that branches into vertical industries—for example, hospitality, dairy, automobile dealers, etc. Beyond data gathering, Expert Loan provides analysis of data sets, designed specifically to flag potential problem areas with a particular applicant.
Interactive data gathering
Altogether, Expert Loan includes fields for more than 6,000 discrete data items. However, a commercial loan for a particular type of business—say, a hotel—might require only a few hundred data items. Asking thousands of unnecessary questions during a loan application would, no doubt, be inefficient, but it could also be confusing and misleading. By using HotDocs to build Expert Loan, the bank was able to take advantage of HotDocs interviews, which are highly interactive. Questions that should be asked under any circumstances can be grouped together in forms that are displayed for each new applicant. Conditional questions may or may not be displayed within a form, depending on the answers to other questions, or they can be grouped with questions having the same conditional profile into forms that are only displayed if certain conditions exist.
Loan approval analytics
Because commercial loans potentially involve very large amounts of money, the bank applies stringent approval standards. One of these standards involves comparing and analysing data items for internal consistency. For example, a hotel applying for a loan of £2 million to upgrade facilities would be consistent if the hotel has, say, 200 rooms. However, if the hotel has only 20 rooms, a loan of £2 million might be inconsistent. While this is a simple example, Expert Loan includes highly sophisticated business rules that can flag any of thousands of different obscure inconsistencies that an actual loan officer might miss. By scripting the requisite business logic into the Expert Loan interview, the bank is able to colour-code answers as a means of flagging possible inconsistencies. For example, an answer that is consistent with guidelines will be coded in green, an answer that may be inconsistent with guidelines will be coded in amber, and an answer that is inconsistent with guidelines will be coded in red. Should a customer representative recommend a loan to an approval officer despite the occurrence of amber and/or red answers, the customer representative can provide a written explanation of why the loan should be approved, regardless of inconsistencies with standards. Expert Loan’s built-in logic ensures that obscure inconsistencies in a data set won’t be missed, a possibility that could yield a catastrophic outcome.
Improved efficiency for loan approval officers
Prior to the deployment of Expert Loan, customer representatives would write a report explaining why a loan was being recommended for approval. In an effort to avoid a loan officer needing further clarification of data points in a loan application, customer representatives would make these reports comprehensive and extensive. The challenge for loan approval officers was to extract relevant facts from these lengthy, written reports. Expert Loan solved this problem by enabling approval officers to review only the data points of a loan application. Color-coding individual data items further streamlined and safeguarded the approval process.
One of the key functionalities of the bank’s workflow system is fraud prevention, particularly in the form of collusion between a borrower and a customer representative. One way that the workflow prevents collusion is through rights management. While a customer representative is able to access a particular loan interview in Expert Loan throughout the data-entry process, once the completed interview has been submitted for approval, the customer representative is locked out of the interview. After a loan has been approved, the customer representative can re-enter the interview; however, the data set on which the loan has been approved is locked down. Any additional fields that require answers can be accessed by the customer representative.
With the complete data set approved, the workflow allows for the documents to be generated, a process that takes just seconds. Documents are saved directly to the PDF file format and are profiled back into the workflow, which provides versioning control for HotDocs templates, documentation and data sets.
The bank has utilised HotDocs as an integrated part of its credit processes for more than a decade and has produced millions of loan and collateral documents for its customers. HotDocs provides the bank with several integral pieces of its loan approval workflow, including complex data-gathering, sophisticated data analysis and automated document assembly.