At most banks, spending on information technology is akin to an unguided missile.

Information technology spending is out of control because it often does not proceed logically from a plan or vision linked to an appropriate overall business strategy that maximizes shareholder value.

Lacking such a plan, banks become in effect demand feeders. They lavish technology funds on the wheels that are the squeakiest or on activities that create the most revenue (though not necessarily the most value). Alternatively, money is spent in a purely reactive way - i.e., "we've got to buy what the bank across the street has."

The short-term consequences of this unfocused spending are:

*Overinvestment in poor businesses and underinvestment in good ones.

*Persistent complaints about the inadequacy of systems from line executives in businesses shortchanged by a poorly conceived spending approach.

*Frustrated information systems managers who are whipsawed by inconsistent support from top management.

The longer-term consequences of a lack of focus are even graver - in many cases no less than institutional self-destruction. The reason is that banks whose technology expenditures flow from a rational plan or vision end up doing things a lot faster, better, and more cheaply than those without such a guide.

Banks that know best how to utilize their information technology dollars can bring new products and services to market six to nine months earlier than average to below-average allocators of information technology funds. They are also positioned to integrate the systems of future acquirees into their own anywhere from two to six months faster.

Perhaps most important, best-practice utilization of information technology dollars pays ample cost dividends. First Manhattan Consulting Group calculates that moving from an average to a top-quartile position in information technology performance can pare noninterest spending in, say, the corporate trust business by some 12%.

This raises the return on revenue in the business by enough to boost its return on equity 24 percentage points, or some 45%. Efficiency gains stemming from a move from average to excellent information technology performance are less dramatic in other activities but still noteworthy. In mortgage origination, the most efficient information technology utilizers will earn ROEs 8% greater than the average bank. In credit cards, the payoff is a 4% boost in ROE, while in middle-market and branch banking the increment comes to about 3%.

Since the penalty for being just an average information technology utilizer is so substantial - great enough, in many cases, to spell the difference between being an acquirer or a target - the less advantaged banks have great incentive to begin an internal dialogue that will culminate in a business-driven technology plan. Such a dialogue has the added merit of revealing where a bank has a competitive advantage and therefore should make more proprietary technology investments. Equally important, it will reveal where no such advantage exists, indicating that additional proprietary investment may prove counterproductive and that there is instead a need to improve one's position by developing ties with other entities - bank and nonbank.

The consequences of the failure to develop a technology plan informed by a shareholder value perspective show up most graphically in the retail area. That's because the drivers of value in the retail business are changing rapidly - more rapidly, it turns out, than the capacity of some to restructure their information technology spending.

Thus, one regional bank that invested over $15 million to upgrade its branch delivery platform had to write off a major portion of this investment as a result of a 30% downsizing of its branch system. This downsizing was a direct result of a strategic shift to alternative delivery systems, primarily telephone banking. Another regional bank invested in excess of $20 million in new branch hardware and local area network/wide area network infrastructure and then had to scrap the originally purchased PCs because they did not have the capacity to support the expanded origination and marketing applications resulting from the bank's new retail delivery strategy.

The need to introduce new retail delivery systems as well as to provide information support for radically different retail sales strategies will, in fact, place huge demands on the technology resources of most banks. It is becoming increasingly clear that the key to improving sagging retail returns is a much more extensive and integrated understanding of the customer than was required in the past. Achieving that understanding necessitates the creation of an entirely new facility, a customer- knowledge-based data file.

Such a facility far transcends the traditional customer information file. It includes detailed internal data not previously gathered or integrated - e.g., information on product usage and customer profitability derived from improved transfer-pricing calculations and records that reveal transaction intensities and channel preferences. It also includes external data on credit history, buying patterns, and demographic profiles.

Besides helping to realize the potential of retail, additional information technology spending is required to accelerate merger integrations and introduce new cost-saving and revenue-enhancing technologies. As noted, those banks that have good systems can assimilate merger partners much more readily than those without these systems. If a bank can get a two-month jump on systems integration, it can raise the net present value of a merger with a $15 billion acquiree by about $28 million. If it can telescope that merger integration by as much as six months, the bank's gain amounts to $85 million.

Entirely new technologies eventually will be needed to introduce new products. Future revenue producers include imaged bank statements for the consumer and imaged lockbox systems for the wholesale customer, stored- value smart cards, EDI products, and new derivatives products made possible by improved analysis and anatomization of existing-product cash flows. Consumers have already expressed their preference for copies of checks produced by imaging systems, and many will be attracted to the smart card, which can be used as a debit card, credit card, or portable computer.

It is not clear that the industry has the resources to support new products and delivery systems, assimilate new acquirees, and at the same time invest in new technologies. It is clear, however, that traditional systems are inadequate, beset with costly functional and product processing gaps. Additionally, the large number of fragmented systems creates a serious redundancy problem.

It is for these reasons that many banks are starting to focus on strategic alliances. One can envision revenue-enhancing and cost-cutting alliances penetrating several areas of banking life. For the moment, however, alliance activity is centered on bank transaction processing businesses - those technology- and operations-intensive activities that soak up about 20% of overall bank noninterest expense but would preempt much less if there were a higher degree of interbank cooperation.

Businesses susceptible to alliance activity include check and card processing, data center management, mortgage servicing, securities servicing, and trust record keeping. In the past few years, First Manhattan Consulting has assisted in the birth of a number of alliances in these businesses, including those of Bankers Trust and First Fidelity in check processing, Wells Fargo and Card Establishment Services in merchant processing, and Chemical and Mellon in the securities area. Technology plans that aim at reducing redundancy via strategic alliances could save the industry as much as $1 billion annually in operating expenses as well as untold billions in initial capital expenditures.

In order to make the correct tradeoff between proprietary investments and strategic alliances as well as to otherwise subordinate information technology spending to a rational plan or vision, every bank needs to find the answers to four basic questions. These are:

1. What technology foundation elements need to be put in place?

2. Which businesses should receive the most technology support?

3. How should the deployment of technology be managed within and across businesses?

4. What resources, in what time periods, will be required to implement the needed technology?

Developing a business-driven technology plan means creating a process that interactively provides business managers with the information needed to guide technology development. It also means creating a process that allows senior management to evaluate technology spending tradeoffs and understand all available options for meeting technology demands. Only then is it possible for a bank to claim that it has created the kind of technology plan which will harness future spending to the service of the equity owner.

Mr. Willis is an executive vice president at First Manhattan Consulting Group, New York.

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.