Section 17.1. Project Scope


17.1. Project Scope

Before delving deeper into the technical details of the Intelligent Finance project, we will look at the scope of the project from both a business and technology point of view.

17.1.1. BUSINESS IMPACT

In 1999, Halifax was one of the most respected banks in the UK but was also overshadowed by the "big four"Barclays, Royal Bank of Scotland, Lloyds TSB, and HSBC (Halifax ranked fifth at the time). These four banks together controlled over 80% of the UK banking market in 1999. In order to attack the market share of the "big four," Halifax needed a distinct offering, which would differentiate Halifax in the market, as shown in Figure 17-1.

Figure 17-1. Competition and customer demand were the main drivers for Halifax to move toward new innovative banking products and modern access channels.


At the same time, the UK banking market had seen something of an e-banking gold rush in 1999, with the Co-operative Bank's Smile, followed by Prudential's Egg, HSBC First Direct's 'Little Fella,' and Abbey National's Cahoot launched in quick succession.

As a result, Halifax was under huge pressure to deliver their new bank in an extremely tight timeframe. Halifax management at the time estimated that they would have about one year to execute their plan and create a product that was good enough to succeed in this highly competitive market.

17.1.1.1 Greenfield Project

In order to meet these challenging timelines, Halifax decided to invest GPB 120 million to build the new bank. In October 1999, Halifax hired Jim Spowart, the former chief executive of Standard Life Bank, to head the new company, which was initially called Greenfield. Three months later, the new bank identity was rolled out with the brand Intelligent Finance.

The benefit of being given a blank sheet of paper to create what was internally dubbed the bank of the future was that there were no legaciesthe IF management team was free to reinvent the ways the bank should look and how it should interact with its customers.

But there were also some obvious challenges and disadvantages in the Greenfield approach: There were literally no existing structures or processes; everything had to be invented from scratch.

17.1.1.2 Offsetting

In order to differentiate itself in the market, Intelligent Finance adopted a new concept, called offsetting. CEO Jim Spowart and his team developed the concept of inter-linked accounts, which they called jars. These jars would allow customers to see how the money in their debit and credit balances measured up. They envisaged an offsetting function across all products. As a result, customers are only charged interest on the money they actually owe the bank. For example, if a customer had borrowings of £150,000 and £50,000 in savings and/or a current account with Intelligent Finance, interest would only be charged on the £100,000 outstanding loan, in return for no interest being charged on the savings or current account. Because no interest is earned on credit balances, the customer is not required to pay tax. Over the term of the loan, this can save thousands in interest charges and enable the customer to pay off the loan early.

17.1.1.3 The IF.com Success Story

As we mentioned earlier, since it fully launched in November 2000, Intelligent Finance has been a huge success. In November 2001, Intelligent Finance announced that it had a total of £8.9 billion in balances in hand and forecast to complete. Savings and current account balances amounted to £2 billion.

About one year laterin February 2003Intelligent Finance announced that savings and current account balances increased by 50 percent to £3.3 billion, and customer accounts doubled to 600,000. In March 2004, Intelligent Finance announced that break-even was achieved the previous year. By the end of 2003, the bank had assets of £15.5 billions, with customer accounts reaching 820,000. Figure 17-2 provides an overview of Intelligent Finance's development from a business point of view.

Figure 17-2. Timeline of Halifax Intelligent Finance project.


17.1.2. TECHNOLOGY IMPACT

The decision to build the new bank on a green field also had a huge impact on the technical architecture. On one hand, it is often easier to develop new systems without worrying about existing systems that require an integration effort. On the other hand, starting from scratch involves a lot of decisions that do not have to be made when you have existing systems. Given the extreme time pressures of this project, many decisions were made very rapidly. This included hiring and training over 1,000 new staff members, finding office space, putting management infrastructure into place, and designing the actual software architecture.

17.1.2.1 IF's Service Architecture

At the heart of the architecture is the IF Banking Engine (initially referred to as OnePlan Engine). This engine comprises three major parts: Open Account, Fulfillment, and Service Request. In addition, it provides abstractions for business entities such as Customer, Financial Consolidation, and Underwriting Modeling and infrastructure services such as Process (Workflow), Alerts, and Messaging.

The IF architecture had to integrate a large number of heterogeneous sub-systems, including back-office and front-office systems. The back-office systems include customer account management systems, credit scoring, links to other banks and external credit-card providers, scanning and imaging, document management, printing, and workflow management. The user access channels include call center and IVR (Interactive Voice Recognition), Web channel, and email. Figure 17-3 provides an overview of the Service-Oriented Architecture as implemented by Intelligent Finance.

Figure 17-3. Intelligent Finance's OnePlan Engine acts as the hub in the center of its banking application landscape.


17.1.2.2 Basic, Intermediary, and Process-Oriented Services

Referring to our service classification as defined in Chapter 5, "Services as Building Blocks," the Intelligent Finance service architecture can be divided into two service layers: a large number of basic services in the backend and one very large service in the middle, which is a mixture between a process-centric and an intermediary service.

There are different kinds of basic services in the backend. A good example is the Halifax-owned current account system, which provides XML-based services for accessing customer accounts that reside on the mainframe. Intelligent Finance decided to leverage this existing system from Halifax and to combine it with an off-the-shelf banking package that added mortgage, savings, and personal loan accounts to the existing account functionality on the Halifax systems.

The second service layer in the system is occupied by one very large service, which represents the Intelligent Finance banking engine, covering the functionality required to provide the seamless integration of the different accounts.

Technically, the banking engine represents a mixture between a process-centric service and an intermediary service. For example, the banking engine provides access to the different customer accounts that reside on different sub-systems. This part of the banking engine functionality does not really add much business functionality; it has the characteristics of a service access layer (see Chapters 5 and 6) designed to provide a unified interface to a set of basic services with heterogeneous service access technology. Other parts of the banking engine service provide process-centric logic. For example, all the service request features are provided through the banking engine, such as "replace lost credit card." All the service request features of the banking engine are based on a workflow engine. The banking engine provides the interface between the workflow engine and the user access channels through a set of process-oriented service interfaces. Approximately 250 different service request types are implemented this way.

Another set of banking engine interfaces is dedicated to the offsetting functionality of the bank. Again, these interfaces combine the characteristics of process-oriented and intermediary services. On one hand, the banking engine service provides the necessary process functionality that is required for customers to control the balances on their individual accounts. On the other hand, the banking engine acts as an intermediary service to the extremely complex calculations in the basic services, which take place, for example, if mortgage and credit-card interest is being set off against interest on savings.

The design of the centralized banking engine service provides many benefits to the frontends (user access channels) that access the service. For example, the current design enables all user access channels to share the same functionality and provide end customers with a consistent view throughout the different access channels. In addition, the design enabled very efficient integration between the different access channel technologies. For example, the call center application provides agents with co-browsing capability (see Chapter 10), effectively enabling them to get exactly the same view that the end user would get on his own data through the Internet. In addition, call center agents have a so-called super screen, which provides additional information, such as the customer's contact history.

On the other hand, the centralized design of the banking engine service also has some disadvantages, which we will discuss later. The most important problem related to the design of the service is the lack of modularity and the cross-dependencies between the different interfaces provided by the service, which in particular make development and maintenance of the service difficult. Given the tough schedule the Intelligent Finance team was under when designing the first version of the system, this issue simply was less important; the team is currently in the process of addressing it successfully.

17.1.2.3 Project Schedule

Fueled by the fierce competition in the UK banking market and the boom in Internet-based e-banks, the schedule of the Intelligent Finance project was extremely tight. Figure 17-4 provides an overview of the most important events that led to the successful launch of the bank at the end of the year 2000.

Figure 17-4. Development schedule of IF.COM.


After the decision was made in the middle of 1999 to go ahead with the project, Halifax started to recruit the core team to find key suppliers of technology and IT services and to undertake all the associated activities necessary to set up the bank.

By the end of 1999, the initial architecture of the system was finalized, including the service design for the core banking engine. The actual development of the system mainly happened in the first half of 2000, with ongoing integration tests, unit test, and scalability tests happening in the middle of 2000. During the peak time of the project, 500 consultants, project managers, and developers were involved.

Given the scale and timeframe of the undertaking, it must be credited to the skills and commitment of the project team that the actual launch of the system went smoothly and without any major glitches. While other organizations experienced serious problems during the launch of their Internet identities in 1999, the Intelligent Finance was a huge success from the very beginning.



    Enterprise SOA. Service-Oriented Architecture Best Practices
    Enterprise SOA: Service-Oriented Architecture Best Practices
    ISBN: 0131465759
    EAN: 2147483647
    Year: 2003
    Pages: 142

    flylib.com © 2008-2017.
    If you may any questions please contact us: flylib@qtcs.net