On October 16 Business Rules Excellence Awards (BREA) announced the winners of the 2018 Awards. Congratulations to the OpenRules’s customer, a major California bank, who became a winner! Here is the brief from the BREA’s presentation:
After successfully putting a new OpenRules-based system in production, the bank extended the use of OpenRules to another mission-critical application related to risk management. Below you can read the description of this OpenRules success story.
1. Executive Summary / Abstract
This case study describes a successful implementation of business rules within a large California bank with 80 billion in assets. The bank needed to modernize the existing onboarding system by adding over 600 new, dynamically-formulated KYC (Know Your Customer) questions. The new interaction logic was very complex, requiring the complicated UI to be dynamically modified for different types of customers, accounts, and planned account activities. The questions to be asked were dependent on previously-provided answers and already-known information about customers and their accounts. Hard-coding of such interaction logic into the existing onboarding system would be both cost- and time-prohibitive. Instead a radically new approach driven by externalized rules was implemented to significantly reduce implementation and maintenance costs as well as delivery timelines.
To add new capabilities and seamlessly extend the old system to match the existing User Experience a new, rules-based system became a necessity. The bank evaluated several business rules frameworks including major commercial and open source products and selected the one that best matched the bank’s needs. As a result, the new system was successfully developed, tested, and deployed on the bank’s production servers using rules-based web services. The new system addressed the core challenges and saved an otherwise failing project allowing the bank to meet regulatory commitments and deadlines.
The rules-based approach allowed the bank to:
- catch up on all time lost and still deliver on time;
- integrate the new system with the existing IT infrastructure and to provide seamless user experience and
- create an organizational environment where Business and IT worked together as one team with a minimal learning curve.
Along with essential cost and time savings, the bank listed the following benefits:
- very powerful yet intuitive rules and template architecture;
- short run/test cycles of building dynamic web applications;
- all rules defined declaratively, externalized out of the application code and can be understood and modified by new personal.
Now the bank uses the same business rules framework for another mission-critical application related to risk management.
2. Overview
This case study describes a successful implementation of business rules within one of the largest California’s banks for a highly dynamic web application that supports complex customer account management processes.
The bank has more than 6,000 business users working in branches creating new and modifying existing customer accounts. The existing onboarding system was in production for years and used a traditional (not rules-based) GUI technology. To comply with regulatory requirements the bank needed to revamp their onboarding process by incorporating over 600 new, dynamically displayed questions and sections to complete KYC requirements for different types of customers and accounts that were being on-boarded.
The bank faced major technical and organizational challenges. The interaction logic was very complex requiring the graphical views to be dynamically modified for different types of customers, accounts, and planned account activities. The questions to be asked were dependent on previously-provided answers and already known information about customers and their accounts. If such interaction logic was hard-coded again, then delivery of such a system (due to bugs and potential QA issues) and future maintenance costs would be time- and cost-prohibitive. So, the bank started to consider a rules-based approach that would allow externalizing the logic in the form of user-friendly business rules that could drive both interaction and composition of the user interfaces.
Another challenge that the team had to solve was a seamless integration with the existing onboarding solution for users. Specifically, the requirement was to have end users not to be aware that they are leaving their existing application from beginning to end of the onboarding process. That required matching the existing UI experience and providing a seamless integration between the old and new systems.
In the face of encountered difficulties with no solutions available on the one hand, and looming regulatory fines, on the other hand, the bank had to drastically rethink and seek for a solution that could still work. They opted for a solution that could utilize an off-the-shelf rules-based framework to meet complexity of the requirements. As a result, the new system was developed, tested, and deployed on the bank’s production servers using rules-based web application and web services frameworks.
The selected rules-based approach allowed the banks’ business analysts and software developers to work as one team by adding new functionality and validating all combinations of questions/answers to identify any inconsistencies and flag them to meet strict regulatory and compliance requirements. As a result, the developed system proved to be very successful: (1) bank was able to save approximately 6 months as the result of the pivot and therefore met the originally committed timelines; (2) the system was developed in-house with very little external professional help which reduced budget 3 times by saving on extensive customizations that would have been required to a vendor-based solution.
3. Business Context
Prior to this implementation, the existing onboarding system was in production for years and used a traditional (not rules-based) GUI technology to support more than 6,000 business users working concurrently on creating new and modifying existing customer accounts. The existing system was quite efficient, but the entire interaction and presentation logic was hard-coded, controlled by experienced programmers, and it was difficult to make any changes. To comply with new KYC (Know Your Customer) regulations and address them for each of their unique channels – each servicing quite diverse customer types. It was necessary to implement complex, dynamic flows with overlaying UI interactions including different operating modes, new complex rules which drive system behavior in response to many of the customer types, user actions, ability to pause, save, resume, restore, or cancel different branches of the dialogs, hard stops, etc. The bank needed to add over 600 new, dynamically formulated questions depending on customer or account types, planned account activities and previously provided answers. The existing solution was not able to accommodate new requirements neither in time for delivery timelines nor from the cost perspective. A new approach – using rules to drive the complexity and reduce the cost of development became the only viable option.
4. The Key Innovations
4.1 Innovative Solution
The bank quickly understood that applying the traditional business rules approach “as is” was not going to be successful either. To solve the business issues described above, it was not enough to use generic decision tables and other traditional forms of business rules. To create a new, rules-based graphical interface with a very complex interaction and presentation logic, the bank needed a business rule platform that already provided a rich set of the predefined rules for building (1) web-based layout templates and (2) rules-based GUI navigation. The bank found an open source platform which provided such functionality out of the box without requiring a long learning curve. The new BR platform supported the integrated use of the rule engine and rendering engine that became the foundation for the successful and quick implementation. The following capabilities essentially simplified the development of the required functionality:
- The predefined rule templates already dealt with such GUI concepts as pages, sections, and various types of questions that could be used to easily configured dynamic web pages. The new product already included page navigation and updating rule templates which become critical for the implementation of the new interaction logic. The predefined decision tables for building dynamic dialogs allowed the bank specialists to naturally represent such pieces of the interaction logic as: if the answer to a question “A” is X and the answer to the question “B” contains Y and Z, then hide the question “C” and display the multi-choice question “D” with the possible values V1, V2, and V3
- Below is an example of an Excel-based decision table with interaction rules that allow a user to define how the dynamic content should be modified in real time based on the user answers:
- And here is an example of a typical dialog with dynamically modified content:
- Using this type of business rules, it was easy to extend and customize the provided rule templates when the interaction logic specific for this application required adding new types of rules
- The provided templates for web forms allowed the bank to present the major graphical components required by this web application. However, to support the same look and feel as the old GUI the bank needed to customize and to extend the provided representations for different types of pages, sections, and questions. It became a relatively simple task to add custom CSS, HTML, and JavaScript directly to decision and layout tables presented in Excel and supported by the selected business rules product
- The supported Java API and web service deployment capabilities allowed a quick and simple integration of the rule and rendering engines with existing system including its data warehousing system, ESB (enterprise service bus) and the application server
- The new rules-based approach essentially improved integration across business areas by bringing together business people, developers, QA, and management. It allowed business analysts and developers to work together as one team by not only specifying new rules and web forms, but representing them in Excel tables, creating and running tests, visualizing the resulting GUI, and iteratively improving the integrated system
- The new approach essentially improved the flexibility of decision making and expedite the development and testing processes.
4.2 Impact and Implementation
The new rules-based approach had positive impact on the bank’s employees giving them the common goals and tools to achieve them. Previously business people specified what to do and written specifications were given to developers for implementation. Then the implemented software was given to QA engineers for testing. With a new approach, all these 3 categories of employees were combined in one team constantly working together under the same highly challenging deadline.
The rules-based approach provided the following implementation advantages:
- Instead of creating intermediate specifications the team gradually implemented new pages, sections, and questions using rules and layout tables in Excel, which were intuitive enough to be understood by all involved specialists
- The important part of this development approach was the use on the centralized business glossary that provided terminology, uniquely named decision variables and business concepts common for business and technical people
- The test cases prepared by business specialists in Excel became an essential part of the rule repository maintained along with the rules to support future modifications
- As a result, the team (BA+DEV+QA) added new functionality feature-by-feature following immediate testing, correction, and integration. Even the managers could see, control development and testing progress every day as new functionality being added and available to all involved people all the time.
4.3 Business and Operational Impact
This successful implementation of advanced business rules and decision concepts demonstrated the advantages of the selected approach. As a result, the new system was developed in time within the dramatically reduced budget. According to the bank, business and operational impact include:
- Increased revenues: while we do not have hard numbers, but the newly introduced business rules framework has much smaller and simpler licensing requirements and allowed the bank to minimize a number of the involved consultants
- Establishment of compliance capabilities in accordance with the government regulations
- Improved maintenance and transparency of business rules or decisions used
- Better synchronized system development and maintenance work of different categories of the bank’s employees and consultants from subject matter experts, to developers and QA
- Ease of learning the system implementation by new staff members.
The best proof of the positive impact is the fact that after successfully putting a new onboarding system in production, the bank now uses the same business rules framework and a similar development approach for another mission-critical application related to risk management.
5. Hurdles Overcome
The bank faced the following challenges:
- The old system was not able to support new functional requirements
- An immovable delivery deadline as the bank was facing regulatory sanctions if the new features are not delivered in time
- The bank also needed to take into account a lead time for development of training materials and provide necessary training in thousands of branches that required to minimize necessary changes in the user experience.
- Complexity of the interaction logic would make writing the exact specification of new features a lengthy process. The alternative process allowed business analysts to be directly involved in the rules-based development and testing and reducing SDLC time.
With the new rules-based approach, the last challenge was converted to an essential advantage allowing business people to work in concert with developers to dramatically speed-up development and testing processes for new business rules and graphical components.
6. Benefits
The key benefits of the new, rules-based approach were based on two major factors:
- It provided off-the-shelf business rules and decision management framework that already included predefined templates for complex interaction rules and web forms framework which could be easily customized
- It incentivized the bank to combine business analysts, developers and testers in one team working together by gradually adding new functionality without the necessity to create intermediate specification documents.
6.1 Cost Savings / Time Reductions
The new business rules platform licensing fees are many times less to compare with the existing platform that caused an essential cost saving for the bank.
The selected approach reduced time-to-market twice by 6 months (twice less) and essentially simplified specification-implementation-testing cycle allowing the bank to meet a challenging deadline.
6.2 Quality Improvements
The newly developed system not only added a new required functionality but improved the quality of the solution by making it easier to modify, extend, and to learn for new employees. All of the UI logic was documented in a simple and intuitive, Excel based decision tables like the one shown above.
7. Best Practices, Learning Points and Pitfalls
7.1 Best Practices and Learning Points
- Externalized, rules driven UI logic, was easier to document, develop and test
- Minimized documentation while improved communication and collaboration
- Business specialists and developers working in concert.
7.2 Pitfalls
- Don’t try to develop a rules-based product in house and investigate and compare different off-the-shelf business rules platforms using a clearly defined POC
- General-purpose business rules platforms usually require the development of rules repositories from scratch. Find a platform that already supports domain-specific templates and rules repositories with an ability to be extended and customized.
8. Competitive Advantages
The bank originally implemented the approach only for its Retail Channel. Other channels used different platforms. By switching to a rules-based approach and by selecting the right business rules platform allowed the Retail channel to complete the delivery of the KYC requirements in time in its complete scope and without any performance or quality issues compared to other internal channels.
9. Technology
The following picture describes the underlying technology infrastructure:
This architecture allowed the existing and new onboarding systems to co-exist (at least initially) and to exchange the customer’s data because they share the same Enterprise Data Repository and Enterprise Service Bus (ESB).
10. The Technology and Service Providers
To develop a new on-boarding system the bank selected OpenRules®, an open source Business Rules and Decision Management system (http://openrules.com). It comes with a built-in component, called “OpenRules Dialog” that was specifically designed for building dynamic web-based questionnaires. The predefined and easily customizable templates for interaction rules and for page, section, and question layouts played the key role in the success of this development. The bank also utilized OpenRules consulting, training, and technical support services.
OpenRules fulfilled the bank’s needs with a quickly-competed POC to prove the ability to meet the business requirements; positive reference check; straight-forward Excel-based UI for creation and maintenance of business rules and graphical forms; simple rules configuration logic with predefined and extendable rule templates; minimal learning curve; simple licensing requirements; low cost and a quick development schedule; very responsive technical support.
=====
A few images from BBC-2018 (added on Nov 10, 2018):
OpenRules - Decision Intelligence Software 