Home > Industries > Consumer Lending > Case Studies
|
Faster turnaround time and improved loss mitigation process for delinquent customers |
| |
| Overview |
The customer is a Fortune 500 and diversified financial services company headquartered in Cincinnati, Ohio. It has been providing quality financial products and services to its customers for the last 150 years. One of the largest banks in the United States in terms of asset size and market capitalization, the customer operates five main businesses: Commercial Banking, Branch Banking, Consumer Lending, Investment Advisors and Processing Solutions.
The customer needed testing support for its six applications and chose to partner with Tavant owing to its deep domain expertise. Tavant executed test process consulting, 3 tier reviews, domain-focused Functional Testing, Smoke Testing, XML Testing, Performance Testing, Integration Testing, etc., for all the applications. |
| Business Context |
The customer needed testing support for its following applications:
MB2C: An online mortgage application that enabled effective interaction between the bank's staff and potential borrowers.
CLB2C: A web-based consumer lending application that extended the bank's operational and technological capabilities by enhancing prospective borrowers' online experience with home loans and auto loans, and enabling technological best practices for the bank's call center and financial center sales teams.
Home Connect: An intranet Point-of-Sale (PoS) application that enabled effective interaction of Loan Officers with potential borrowers.
Collection Web Portal:A web-based application that augmented the bank's operational and technological capabilities by enhancing prospective borrowers' online experience with equity and auto loans, and facilitating the adoption of technological best practices for the bank's call center and financial center teams.
Mortgage Loan Officer: A web-based application with storefront capabilities for Mortgage Loan Officers (MLOs), allowing them to establish a direct link with their books of business via the web and to enable prospective borrowers to self-initiate sales relationship through online application submission.
Dealer Reserve: A backend financial application that considered 12-15 parameters defined by the bank to calculate the price of a vehicle for customers. |
| Solution |
The bank chose to partner with Tavant for testing of all the six applications. Tavant and the bank's testing teams worked closely as a Global Testing team for testing of the applications. A key highlight of this engagement model was unambiguous demarcation of Tavant and the customer's testing stakeholders' responsibilities.
Tavant's scope of work included, but was not limited to areas of project planning, execution and monitoring; in fact, it encompassed: |
> Defining project timelines, scope of testing and schedules for testing
> Identifying the required resources, and their roles and responsibilities
> Defining the strategies, test methodologies and test deliverables
> Identifying the nature and extent of tests deemed necessary
> Identifying potential risks and creating mitigation plans for the same
> Determining management and technical effort required to support testing throughout the system development lifecycle
Working in an on-site/offshore model, Tavant carried out the following testing activities for the customer:
> Test process consulting
Tavant designed and implemented test methodologies for functional and performance testing of the bank's applications. These methodologies detailed out the approach to be followed for testing a release (e.g., identifying testable requirements for functional and performance testing, designing test suites, test execution approach, etc.). End-to-end traceability was achieved by using HP Quality Center.
> Adoption of industry best practices for test case designs, including:
-- Pair-wise Testing
-- Equivalence Partitioning
-- Boundary Value Analysis
-- Functional Flow
> 3 tier reviews to ensure 100% test coverage
> Introduction of Build Acceptance Testing (BAT) prior to System Testing
> Incorporation of Layered Modular Testing into System Testing
> Additional Smoke Testing prior to UAT promotion
> Domain-focused Functional Testing
Extensive focus was directed on domain by organizing regular workshops for business analysts, which were aided by on-the-job training. Before test cases were designed, requirements were discussed and analyzed within the team for a clear business understanding and sharp focus on end-user requirements.
> Integration Testing
Stubs and drivers, developed in-house, were used for testing third party vendor integration and the bank's legacy systems. Approx. 80% Integration Testing was automated for validation of request/response XMLs and field mapping documents, resulting in 30-40% increase in productivity.
> XML Testing
This involved modifying XMLs, and validating them with XML schema, DB and UI.
> Pricing
In-house developed Test Harness was used to test high precision and complex calculations for pricing.
> Performance Testing
Performance tests were run using HP LoadRunner to detect critical performance bottlenecks.
> Planned Regression Testing
End-to-end scenarios, cutting across critical functionalities of the applications, were prepared and used as one of the bases for Regression Testing.
-- 14 business critical scenarios were created as a part of regression test suite for MB2C application
-- 10 business critical scenarios were created as a part of regression test suite for CLB2C application
-- 34 business critical scenarios were created as a part of regression test suite for HC application
> Statistical approach toward test management, execution and monitoring
-- Strict adherence to Entry/ Exit criteria, as defined for each phase of testing
-- Pre-QA Doneness Analysis
-- Unit Test Coverage
-- Strict adherence to QA gate criteria through Defect Prediction and Defect Trend, and by defining UAT Acceptance Criteria
Besides these activities, Tavant also:
> Adopted best practices to reduce test cycle time
> Enhanced integration test coverage upfront
> Streamlined the release process
> Executed effective "error-proofing" in early phases of the project
> Followed collaborative engagement and execution
However, in meeting the customer's requirements, Tavant faced a number of challenges, such as:
> Aggressive release dates and testing timelines
> Functional testing of business-critical mortgage workflows
> Usability, accessibility, security and performance of web applications
> Preparation of test data through retrieval of data from multiple systems
> Testing of high precision and complex figure calculations, which were to be used across the bank for pricing loans
> Need for reduction in testing cycle time without compromising on the test coverage
> Dependency of the core applications on third party systems, such as Credit, DU and LP, and the bank's legacy systems like ACAPS, STRATA, UniFi, Home Value BOT, LDAP, eStara, Nautilus, CACS, PeopleFinder and DocuStar
> Limited documentation on the bank's legacy systems
> Continuously evolving requirements
But these challenges were overcome with effective plans and strategies.
Metrics:
|
Project Title
|
Total Number of
Test Cases
|
Test Efficiency
|
Test Coverage
|
Post Delivery
Defect Rate
|
|
MB2C
|
2147
|
95%
|
98%
|
0.01
|
|
CLB2C
|
2806
|
95%
|
99%
|
0.12
|
|
HC
|
8677
|
-
|
97%
|
-
|
|
CWP
|
3064
|
98%
|
99.5%
|
0.29
|
|
MLO
|
2260
|
-
|
99.5%
|
-
|
|
DR
|
2409
|
-
|
-
|
-
|
Business Benefits
Tavant helped the customer realize the following business benefits:
> Improved quality of the applications, with lesser, and at times, zero defects reported from production
> Faster turnaround time
> Enhanced operational efficiencies
> Improved loss mitigation process for the bank's delinquent customers
> Helped the customer identify critical system bottlenecks and scale issues prior to production
Tavant Differentiators:
Tavant helped the customer realize the following business benefits:
> End-to-end testing capabilities - right from Requirements Analysis to UAT Management
> Tavant Certified QA Engineers, with expertise in Mortgage Domain, CSTE, CSQA, HP QTP
> Reusable automation frameworks for Integration Testing
|
|
Conducts end-to-end testing of the application to help the customer achieve operational & financial excellence |
| |
| Overview |
The customer is one of the leading wholesale lenders in the United States. One of its group companies operates as a wholesale sub-prime lender in the US, offering direct loans to customers. The customer had been processing all loans and valuations manually, which was a tedious task. To automate the process, the customer started using an application called ReSolve. It needed robust testing of the application to achieve operational excellence. Tavant offered end-to-end testing support for ReSolve by utilizing its world-class resources in an on-site/offshore model. |
| Business Context |
The customer had been processing all loans and valuations manually. All information of applicants was entered manually from a physical form and copies of the loan origination document.
To automate the process, the customer started using ReSolve -- an integrated solution for processing and management of all delinquent loans for the customer. ReSolve became a part of the loan servicing group, and dealt with all the loans that went past the 91-day delinquency and became default candidates. ReSolve used to manage the processing of all delinquent loans from day one -- collections actions through final disposition.
ReSolve comprised numerous modules, represented in the image below:
The customer needed robust testing of ReSolve to achieve operational and financial excellence.
ReSolve greatly reduced the number of errors in applicants' data, arising out of manual entry, and also prevented redundant data entry in multiple databases. It soon became a critical solution for the customer because it: |
> Ensured savings of approximately $300 Mn in Default Management costs
> Reduced payment processing time from 5-6 days to 3 days
> Reduced REO timeline from 120 days to 90 days
> Processed over 50,000 loans worth US$ 1.2 Bn
| Solution |
Tavant offered end-to-end testing support for ReSolve by utilizing its world-class resources in an on-site/offshore model. Tavant's rich domain expertise, rigorous process framework and proven project execution capabilities played a pivotal role in ensuring a robust, reliable and efficient solution.
The testing and delivery organization at Tavant took complete responsibility of all aspects of the following testing activities for ReSolve:
> Prepare test approach and strategy
> Quantitative test management
> Establish test management processes and test reporting standards
> Establish and maintain effective communication and escalation hierarchy for quick resolution and minimum bottlenecks
> Conduct Software Quality Assurance (SQA) and Software Engineering Process Group (SEPG) audits to ensure compliance with CMMi processes
> Optimize testing methods, framework, techniques, tools and resources
> Ensure macro as well as micro reporting at regular intervals to minimize last minute issues
> Review requirements with Business Analysts
> Prepare and review test ware (including test cases, test data, automation scripts)
> Follow statistical approach (SQC) to define Acceptance Criteria viz. Defect Prediction, Defect Trend, etc.
> Establish acceptance criteria for delivery to the client
> Define entry/exit criteria for each phase of testing
> Execute framework-based automation using automation open source and industry standard tools like Watir, Selenium, Winrunner, QTP, LoadRunner
> Estimate test effort
Tavant provided the following testing solutions to the customer:
> System Testing
> Test Ware Review
> XML Testing
> Workflow Testing
> Web Application Testing
> Security Testing
> Browser Compatibility Testing
> Automation of Regression Testing
> Performance Testing
> Cross-Module Testing
> Integration Testing
Some of these activities are detailed below:
Preparation of Test Strategy
Tavant designed and successfully implemented a test methodology to be used across functional and performance testing of ReSolve.
System Testing
Before test cases were designed, requirement reviews and walkthroughs were conducted with Business Analyst for a clear understanding of the business and end-user requirements. It included:
> Functional Testing
> Field Validation
> Boundary Testing
> Screen Flow / Navigation
> Security Testing
Test Ware Review
3-tier (on-site testing/offshore testing, Dev and BA) reviews were executed for test cases.
XML Testing
This testing methodology involved modifications in XMLs, as UI was not available, and validating them with XML schema and uploading them to database.
Workflow Testing
Tavant ensured whether certain specific workflow activities were getting completed in a loan workout cycle - right from initialization to completion.
Web Application Testing
Tavant executed web testing of ReSolve to determine whether it was working properly and good to go live.
Browser Compatibility Testing
Tavant executed browser compatibility testing of ReSolve to ensure that it was supported equally well by all major browsers.
Automation of Regression Testing
Tavant implemented a keyword-driven automation framework for regression test suite, using HP WinRunner, to reduce ReSolve's time-to-market, while increasing test coverage. Over
60% of regression tests were successfully automated by Tavant.
Performance Testing
Tavant carried out performance testing, using HP LoadRunner, to detect critical performance bottlenecks in the application.
Cross Module Testing
Tavant implemented smooth testing of complex inter-dependent modules of the application by examining cross-module dependencies tracker and integration checklists.
|
| Tools & Technologies: |
> HP WinRunner
> HP LoadRunner
|
|
Business Benefits:
|
> Zero critical defects reported by the customer
> 60% savings in regression testing costs owing to automation testing
> 30% cut in overall resource requirements
> 100% upturn in user capacity of the application
> Minimized loss severity ratios
> Maximized savings over Real Estate Owned (REO)
> 60% shorter processing cycle time and response time
> Tighter integration with vendors and among various functions within default management
|
|
Segregated from Unifi, the pricing engine significantly reduces loan processing time |
| |
| Overview |
The customer is a Fortune 500 and diversified financial services company headquartered in Cincinnati, Ohio. It has been providing quality financial products and services to its customers for the last 150 years. One of the largest banks in the United States in terms of asset size and market capitalization, the customer operates five main businesses: Commercial Banking, Branch Banking, Consumer Lending, Investment Advisors and Processing Solutions. |
| Business Context |
The customer had customized Unifi, an application developed in Progress 4th generation language (4GL) by Fiserv, for processing loans. For each loan request, xml data from application forms was picked and stored in Unifi. The data stored in Unifi was used to calculate loan prices through a pricing engine. The results were then fished and stored back in Unifi to process loans.
The customer felt that Unifi's code was difficult to maintain. It also realized that it required a deeper understanding of the base product code created by Fiserv and that of Unifi's extensive customization. The customer did not have any knowledge repository to understand the functioning of the system. Hence, it needed more research and a thorough analysis of the same.
Another challenge was that Unifi and the pricing engine were integrated. Due to this, loan requests for other systems of the bank also needed to pass through Unifi. This led to very high turnaround times for calculation of loan prices. |
| Solution |
Tavant segregated the pricing engine from Unifi, thereby enabling
the customer to process loans for its other systems easily. Now, th
loan requests for other systems did not have to pass through Unifi.
The pricing engine also filtered data for products like Home-Connect (HC), Mortgage-Loan-Originator (MLO) and Mortgage Business to Customer (MB2C) on the basis of input parameters.
For each potentially eligible product, the pricing engine:
> Ran calculations required to calculate adjustments (for example, loan to value, combined loan to value, etc.)
> Applied point adjustments
> Applied rate adjustments and calculated final rate
> Ran calculations to compute attributes like principal interest/interest only payments, fees, cash required to close, "Debt. to Income," etc.
> Applied guideline rules
> Applied product matrices
However, if any guideline rule or product matrix rule failed, the product should not be disqualified immediately; rather, failure conditions should be identified and analyzed by following the steps given below:
> Check if pricing is valid/invalid, based on failed guidelines
> Return pricing results
At any stage, if the eligible product or pricing cannot be listed, then a corresponding error message is displayed. Pricing results are returned immaterial of whether guideline rules are passed or failed. A flag is set to indicate that pricing is invalid and client application must use this flag not to let users proceed with and use the priced results. |
| Implementation Highlights |
The pricing engine functionality was exposed as a service, accessible via:
> Simple Object Access Protocol (SOAP)
> Message queues: Accessibility via message queues enabled even Mainframe Applications to use pricing engine's services.
The pricing engine notified the client application when any pricing rules, rate sheets or any other metadata was modified. Another advantage was that despite disintegration of pricing engine from it, Unifi system did not require any change. |
Tools & Technologies |
> OS: Windows XP or above
> Database: Progress, Oracle and SQL Server
|
Business Benefits |
> Processing time for loans was considerably reduced.
> Unifi system did not need any change.
> Better performance and better maintainability than the earlier Unifi pricing engine.
> Intuitive and smart client UIs, reducing the complexity of core pricing engine and improving its manageability.
|
|
| |
Easy product implementation enables the customer to deploy products rapidly and enhance customer satisfaction |
| |
| Overview |
The customer is a well-established financial services leader in the U.S. Its core retail products consist of life insurance, long-term care insurance, annuities and mutual funds. The customer had been using various legacy systems and found it difficult to consolidate policies from them into a single system. The customer's legacy systems were migrated to an online application that enabled users to access information about their life insurance products at the click of a mouse.
|
| Business Context |
The customer had been using multiple legacy systems and found it difficult to consolidate policies from them into a single system. Therefore, it needed a business model that would allow for automated processed organization of policies within a foreseeable time period.
The customer needed a private-public sector collaboration model that would meet all its needs and value systems. Additionally, it also wanted to automate information exchange with its customers.
Additionally, none of the legacy systems being used by the customer could support all its existing or new and complex products. To be updated, the legacy systems, many of which were obsolete, would have required a significant investment in time and resources, which was not feasible. |
| Solution |
The customer was advised to consolidate its policy administration systems by adding a new system with better functionality and flexibility. Subsequently, the customer's legacy systems were successfully migrated to an online application that enabled users to access information about their life insurance products at the click of a mouse. The Variable Pay Admin System (VPAS) helped the customer eliminate manual processes involved in policy administration.
VPAS offered various levels of security to users. It included a report processor, which was used for generation of ad-hoc reports and running them. VPAS also had the capability to import/export data from/to other interfaces to create Management Information System (MIS) or custom reports.
|
Implementation Highlights |
> New product addition every year
> Re-priced module in Variable Universal Life Insurance in 7 weeks
|
Tools & Technologies |
> OS: Windows XP or above
> Database: Progress, Oracle and SQL Server
|
Business Benefits |
> Easy product setup and implementation enabled the customer to deploy products rapidly and enhance customer satisfaction
> Re-usable code enabled the customer to design multiple life insurance products with minimal programming and virtually no regression testing
> Simple security setup for various levels of users and groups, and dynamic menus that could be configured or removed to suit customer's requirements and regions
> Detection and recovery of errors in Daily Events Processor, Reports Processor and user interface
> Policy administration for such functions as new business setup, agent/agency definition, billing, remittance, surrenders, loans, investments, and so forth, compliance testing and undo/redo capability
> Generation of standard/custom reports and extracts from import/export interfaces
|
| |
| |
System also helps the customer achieve standardization of reporting of asset's status |
| |
| Overview |
The customer is a Sri Lanka-based fast moving consumer goods company with local manufacturing facilities, reporting to the regional business groups for innovation and business results. It has operations in Home Care, Personal Care and Foods domains, and owns over 20 strong brands that are leaders in all the categories that they operate in.
|
| Business Context |
The customer needed a system to integrate its asset's data and their details with that of its Head Office within a short span of three months. The system was required to map various reporting procedures as per the management hierarchy periodically. Additionally, the customer also wanted to bring about transparency of funds (available or spent) in the company.
|
| Solution |
An application module called Fixed Asset System (FAS) was developed for accounting of the customer's tangible assets (e.g., land, machines, building, furniture, etc.). These assets represented a substantial financial investment by the customer. FAS module took care of the cost of acquiring, maintaining, insuring and replacing these fixed assets, along with their depreciation, which could have a substantial impact on financial statements as well as tax liabilities of the customer.
FAS was developed to set up, maintain, transfer and retire the customer's fixed assets, and integrated the FAS module with General Ledger (Finance) module. Changes in the FAS module was made to build necessary controls and functionalities to meet the customer's business requirements.
FAS module was customized such that it could perform functions, including:
> Creation and maintenance of assets
> Maintenance of asset account information, insurance data, book details and asset components
> Modification in depreciation books
> Retirement, idling of a group of assets
> Transfer of assets
> Various reports generation (for Audit, Head Office)
|
Tools & Technologies |
> Operating System: UNIX
> Database: Progress
> ERP: MFG/PRO
|
Business Benefits |
> Ability of the customer to best use the available funds
> Timely access to updated asset information for the purchase and financial teams
> Clarity on fund flow and asset value for each department
> Online tracking of the status, value and location of assets
|
| |
| |
Facilitates Online tracking of status of checks received from business partners |
| |
| Overview |
The customer is India's largest Fast Moving Consumer Goods (FMCG) company. It owns over 30 brands spanning 20 distinct categories, including personal care and hygiene, food and beverages, home appliances, etc.
|
| Business Context |
The customer was operating in a financial ecosystem that included its customers i.e. end-customers, banks, mail services and dealers/vendors/suppliers.
The customer had an agreement with banks that allowed it to encash post-dated checks immediately by either paying a minimum amount or a percentage of the check amount to the bank. The bank would then release the check amount to the customer. Due to this, tracking of payments had become a little complex.
Moreover, end-customers, banks, mail services and dealers/vendors/suppliers had been using disparate payment tracking systems, which made cash management difficult.
The customer had been using manual, time consuming and costly procedures for tracking payments, due to which it was not able to make the best use of flow of funds through post-dated checks. The customer also found it difficult to track prepaid checks from end-customers, banks and dealers/vendors/suppliers.
Due to all these challenges, the customer wanted to standardize the payment procedure(s) of all concerned parties. |
| Solution |
An application module called Cash Management was developed to maintain details of the payment instruments i.e. checks, demand drafts, etc. This module was basically designed to keep track of cash flow among the customer, banks, end-customers and dealers/vendors/suppliers. It also kept a track of checks from customers to the company, banks, mail services and all intermediate parties.
The module enabled exchange of data among all parties on a daily basis.
|
Tools & Technologies |
> Operating System: Unix
> Database: Progress
> ERP: MFG/PRO
|
Business Benefits |
> Online tracking of the status of checks
> Capability of the customer to best use the funds
> Timely access to up-to-date financial information for the customer's financial team and auditors
|
| |
| |
Enables the bank to run dynamic auto loan campaigns, be more responsive to customers, and react to changing market conditions |
| |
| Overview |
The customer is a Fortune 500 diversified financial services company based in the US. The customer had been using legacy methods to manage an auto loan pricing model and business data for offering auto loans. Tavant developed an auto loan pricing modeling application that considered 15 standard business parameters to determine the best-fit price of an auto loan for auto buyers, but could be extended to any number of parameters. The application enabled the customer to dynamically define/change the pricing model without any downtime. This helped the customer be dynamic and responsive to emerging market conditions, giving it a competitive edge in the field.
|
| Customer |
The customer is a Fortune 500 and diversified financial services company headquartered in Ohio. It has been providing quality financial products and services to its customers for the last 150 years. One of the largest banks in the United States in terms of asset size and market capitalization, the customer operates five main businesses: Commercial Banking, Branch Banking, Consumer Lending, Investment Advisors and Processing Solutions. |
| Business Context |
The customer had been using legacy methods to manage business data for offering automobile loans to its customers i.e. auto buyers and for driving auto campaigns. The customer’s pricing systems had to be updated with business data manually; also, the data stored in them was neither extensible nor easily manageable. For each small change or update in data, a new change request had to be raised and a new initiative launched -- which was a very long, time-consuming process.
Also, the customer needed to plan weeks in advance to run time- and region-specific business campaigns for auto loans, react to market conditions or respond to competitors’ business tactics. |
| Solution |
Tavant developed a dynamic Web-based auto loan pricing modeling application for the customer that offered a sophisticated mechanism to define and manage business data for auto loans. The application considered 15 standard business parameters to calculate the best-fit price of an auto loan for auto buyers. The Web solution simulated spreadsheets/decision tables to organize business data, offering extensive flexibility, and letting users integrate business data seamlessly.
Therefore, the application helped the bank organize business data on auto loans intuitively and dynamically, rendering it more competitive and responsive to the needs of auto buyers.
The application segregated data as per the customer’s requirements, allowing it to run time- and region-specific business campaigns for automobile loans.
|
Highlights |
>Zero-defect delivery of the project
|
Business Benefits |
> Enhanced productivity.
> Reduced response time to react to competitors’ tactics and customers’ needs.
> Ability of the customer to run time- and region-specific business campaigns for automobile loans.
|
| |
| |
|
|
 |
| Our Customers > |
|
|
