What will be financial institutions' competitive strategies in a future heavily influenced by the Internet? An expert on competitive strategy, Harvard's Michael Porter, offers only two basic paths:

Cost leadership or focus. Through scale, regulations, or factor inputs, low costs yield higher margins or more share. Examples are State Street, in global custody, and credit unions.

Differentiation. Through innovation or appeal to narrow market niches, products are distinctive. Examples are Schwab's mutual fund supermarket OneSource and Capital One's data-driven card segmentation.

For financial institutions, the convenient delivery of services affects both cost and differentiation. Convenience has traditionally meant dense branch networks in good locations, call centers open 24 hours every day, personal bankers and brokers, corporate calling officers, and the ability to gain access through all channels.

By definition, only a few financial institutions can achieve an advantage. Measures of success include return on assets, return on equity, market multiple, or gain in share on a strategic business unit basis. Current leaders tend to be low-cost regionals (Fifth Third, for example) or companies with distinctive processing businesses that have a high share of market (for example, Synovus or Bank of New York) or that have a takeover premium in their stock price.

Mr. Porter says nothing specifically about size. But according to Bank One chairman Verne Istock, "Size is definitely a competitive advantage, and we must use it to create superior products and services for our customers."

On the Internet, these same basic issues arise. No financial institution will keep advantage simply by being there. The issue is: What cost and differentiation strategies will work on the Internet? What is the relationship between a Web strategy and an off-the-Web strategy?

Internet transaction costs look low, assuming steady-state operation. Yet to get there, a financial institution must invest heavily in infrastructure and content. One major banking company was horrified recently to discover that it had 240 Internet projects under way.

U.S. financial institutions' 1999 Internet spending will be about $3.1 billion, which is not just for technology. By 2002, every large bank's total Internet costs may exceed $200 million a year.

Internet expenses are growing ominously. These skyrocketing costs support:

New product features and functions, sometimes in fourth- or fifth- generation sites.

Greater reliability, such as mirror sites to assure fault tolerance.

Increased bandwidth for graphically intense applications, including larger, better Web site graphics; video; and interactive, multimedia applications such as shockwave.

Middleware and back-end connections; data base reengineering.

Capacity and scalability (witness recent problems at E-Trade).

Electronic mail back-office processing capability.

Employee training and workflow reengineering.

Internet scale is like any other-high fixed costs require more volume. So financial institutions are racing to attract eyeballs, hoping to get more than just their existing customer base. One thing that won't work for financial institutions is the strategy of becoming a portal. 100hot.com lists the Web's most popular sites, of which a mere two are financial, neither of them banks. The general-interest Web site shakeout is already intense.

Start-up Internet banks need volume desperately and don't have a current account dowry. But through the magic of outsourcing with players like Security First Technologies, Home Account Network, or Digital Insight, the costs are reasonable. Connected to a real-time core system, de novo banks may even enjoy lower unit costs despite their lower volume, simply because less integration is needed.

An unanticipated wild card is growth of Internet-stimulated back-office costs. The volume of transactions on core systems will accelerate. By 2001, for example, Bank of America's system transaction count could easily exceed one billion a day, squeezing the underlying batch architecture unmercifully and possibly requiring core system rebuilding.

The Web allows a revolutionary outsourcing of data entry to the customer. Now they fill out electronic forms, instead of bank clerks' filling out paper forms. But for every dollar saved, two will be spent on training customer service employees, helping confused customers, and researching Internet-generated exceptions.

Furthermore, financial institutions won't be able to shut old channels aggressively. This won't bother the start-ups but will inhibit existing players. They'll keep asking why they can't make a profit on the new channel. The net score on costs is a competitive advantage in the short run to the de novos and in the long run to the large players that actively invest but stay off the bleeding edge.

The Internet is a spawning ground for improved financial products and features. Its capabilities include:

Reducing old barriers to investment. For example, initial public offering access is expensive and limited to institutional investors. Wit Capital and others, however, have a strategy to sell scarce equity over the Internet to accredited investors for 10% of the usual cost. Other offerings in this vein include private equity, venture capital funds, and even angel funds, in which a selected group of investors gets access to initial public offerings. Thebasic concept is to create differentiated products with the new technology.

Corporate bill presentment. For example, large MCI/Worldcom customers used to get voluminous paper phone bills. Now, using EDS' interactive billing services, the bills are presented electronically.

Customer control of access to their own account statements through their PCs, instead of relying on the batch cycle.

Event triggers that could initiate push notifications, changed credit controls, or helpful advice. The ability of the Internet to "push" documents to users offers much in the way of added convenience, mimicking the benefit of "snail mail" statement stuffers.

Manipulation by customers of financial data bases like statement data, combined with analysis of payee data.

Presenting of check images on the Internet and eliminating paper statements.

Personal pages for investment ideas, stock portfolio tracking, etc.

Downloading of e-money to smart cards.

This multiplicity allows for experimental differentiation. Yet the reality is that only a few new entrants will do the new things. The vast majority will be satisfied with doing things new to them but not new to the evolving industry and therefore not an advantage. Too many will mistake this corporate change for gaining an advantage instead of seeing it for the just keeping up it really is.

Web interfaces are improving much faster than market share is shifting. Laggards will soon get the picture, and in three to five years, as the technology and ideas stabilize, Internet access and features will become much more of a commodity. Some of today's early movers will even turn out to have been on the bleeding edge and will be leapfrogged.

Automated teller machines offer a parallel. Thirty years ago, banks sought advantage through early (and expensive) investments in them. Citibank built its own machines and refused to join shared ATM networks until the 1990s. Likewise, BayBanks sought to saturate eastern Massachusetts, seeking the best locations.

Yet today, ATMs are essentially a shared utility with standard interfaces. A stand-alone business of capturing and selling cardholder transactions is appearing. New ATM interface features will evolve but not as competitive advantages for specific banks.

Reviewing 100hot.com's listings of (purported) high activity finance, funds, and loan Web sites, one is struck by immense incongruity. The sites listed have virtually no alignment with the industry's largest financial institutions. For example, only 10% of the 100hot loan sites are well- known, large lenders. It is as if previously unheralded entities had wrested large parts of consumers' destination traffic from the big players.

The moral is that previously established brand names and customer bases will not work automatically for Web surfers. Consumers want products they like, and they will buy from anybody within reason. We are witnessing the all too rapid establishment of a new Internet archipelago that won't respect anything but functionality, low price, and ease of access. Balancing that evolution with cost and differentiable product, without losing current customers, should be enough to keep every financial institution fully occupied in the next few years.

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.