First of America has taken a business rather than a technically-oriented approach to determine the appropriate platforms for our existing and emerging applications. The major benefit of PC technology, we feel, will be in our ability to provide business applications that could be time constrained or functionally impractical to deliver using mainframe processes.

First of America uses PC technology for new applications as well as for right-sizing applications that currently run on our mainframe.

An example of a new application is our platform automation system, which provides current, comprehensive information about customers and products to our branch personnel.

An example of an application that we recently right-sized to a PC environment is our Farm Management System. This is a financial management system based on present and future agricultural assets of our farming customers.

We're probably taking a different approach from other banks. We're moving all of our processing off the mainframe and onto personal computers in a client-server environment.

Many banks still say big number crunching must be done on a mainframe, but we disagree. It may be because of our size. We're $8 to $9 billion in assets, with about 2,000 employees.

Right now we do have two big Unisys machines that we'll be replacing with PCs over the next few years. Because we're a savings bank, we are already very PC-oriented. All of our tellers and platform personnel already use PCs. We also use PCs in our asset resolution area, where we do foreclosure management. And we just recently signed an agreement to do loan originations in a client-server environment. General ledger and batch overnight processing are the last things that we'll move to the PC, over the next two years.

We're looking at moving any task that looks to be more efficient and cost-effective in a PC environment, rather than on the host.

A lot of our teller operations have already been moved on to PC LANs. We have 1,200 LANs at about 1,000 sites throughout California. Our tellers and branch employees use what we call COIN, for customer on-line information network.

In the loan area, we use a number of purchased packages that run on the mainframe. While we will continue to have these packages running on the host, we want to give the personnel in our loan offices portability and flexibility. So we will be adding functionality at the PC end.

In our portfolio management area, we are setting up worldwide access to a client-server data base, which will help us compete, since it will allow us to update our portfolio on a daily basis and improve accessibility of information.

A lot of conventional banking applications can be moved to client-server technology. But not every application is ready for distributed computing because of the reengineering needed to make it work.

Also, a lot of high-volume applications aren't quite ready yet.

The challenge is to pick and choose applications carefully and to understand the difference in the computing environments and make sure the move to PCs makes business sense.

We've moved a lot of our customer-information and customer-service applications to PCs to take advantage of the capabilities inherent in the PC environment, which give our customer-service representatives the flexibility to address customer queries quickly and efficiently.

Technically, you could move everything to personal computers in a client-server environment. But that would be dumb. The exciting part is deciding which applications are best moved to PCs.

Many of the customer service applications that were written in the 1960s and 1970s for retail service can now very successfully move to PCs. As financial service companies get more integrated products and customers get more sophisticated about what they want, we can move more and more customer-initiated transactions to PCs.

I think we can now realize the dream because distributed processing can be successful in the client-server world. I think it's a very exciting time.

The area where technology is migrating off host computers to personal computers is the branches. Branch employees need customer information at their fingertips. They need additional information in order to give better service. That's true of many, many banks.

We also offer customers the option of banking from their home, through their own PCs.

Our audit group has a very sophisticated PC network. They also have lap-top computers with dial-up capability so they can access the bank even when they're on the road.

We also use PCs for general office needs. We've got 1,200 PCs and our managers are using them more and more.

We made the decision to shift some applications from mainframes to PC networks in 1989. Mainframes continue to be our primary resource for high volume transactions like check processing, but even these systems are being front-ended by PCs.

Since the late 1980s, human resources, budgeting, real estate owned, and profit and loss forecasting have been moved from the mainframe to PCs. The first conversion, human resources, pinpointed the advantages of PC usage. A large backlog of software change requests existed and it was difficult to find programmers with the right expertise. Today this system is managed by non-technical people.

Our mainframe houses the major informational data base for the bank.

In the savings area, we've installed local computers at branches that do marketing and cross-selling as well as new-accounts opening. These terminals are linked to the mainframe.

In the loan origination departments, we have installed local systems to handle a loan from the point of application to the loan closing. Upon closing, it is uploaded to our mainframe for the loan servicing process.

If the loan goes into arrears, the information is downloaded to local PCs in our collections area where the collections personnel can call up a screen and get the information they need.

Subscribe Now

Access to authoritative analysis and perspective and our data-driven report series.

14-Day Free Trial

No credit card required. Complete access to articles, breaking news and industry data.