U.S. consumers and businesses control some $27 trillion in financial assets. The electronic commerce (EC) industry is counting on this reservoir of wealth to provide the raw materials necessary to buy, sell, invest and lend on the Internet. There is just one problem: Most of these assets are locked away, deep within financial institutions. Without an easy way to electronically access, analyze and

manipulate these assets, the EC industry will be deprived of its lifeblood, unable to fuel the engines of commerce.

The financial software sector is dedicated to solving this access problem by building bridges between the EC industry and the financial services industry. Currently, this sector is hard at work building the basic connections necessary to provide consumers and businesses with access to their financial accounts over the Internet. These efforts are not only opening up an important avenue for the flow of funds into the EC industry, but also increasing the general public's overall exposure to and confidence in the EC industry.

But the financial software sector's work will not stop at this level. Beyond just providing access to financial accounts, the sector is beginning to focus on how to apply new technologies to not only reengineer the process of providing financial services but also to enable entirely new services that were not even possible before the rise of EC.

The reality, (however), is that most businesses and consumers cannot access their financial accounts electronically and that even when they can access their accounts, subtle but significant problems prevent the financial software sector from fulfilling its mandate. To understand why this is the case, it is necessary to take a look at just how the sector is currently "solving" the problem of providing customers with access to their financial accounts. In order to provide on-line banking services, it is necessary to string together four basic components: front-end software, a middleware system, a core processing environment, and an electronic payment processing operation.

Front-end software enables consumers and businesses to access their financial accounts. There are different types of front-end software: voice response, screen phone and PC software (proprietary, personal financial management (PFM), basic access, on-line services and Internet software). The fear of betting on the "losing" software has led banks to offer as many different front-end options as possible. This "shotgun strategy" provides banks with piece of mind, but it also adds a great deal of middleware complexity.

The middleware component serves as a bridge from the new world technologies of PCs, on-line services, and the Internet to banks' old world legacy processing systems, which are usually large mainframe computers. Banks must build not just one bridge, but multiple bridges to translate messages between front-end software and legacy systems.

Core processing systems are the engines of the modern banking system in that they store, analyze, and process almost all account transactions and information.

In general, most banks do not have a single core processing system but, rather, a collection of separate processing systems, each having its own databases and applications. This means that each separate processing system requires its own customized bridge to the middleware component.

As with both front-end and core processing components, a bridge must be built between middleware components and electronic payment operations to electronically transmit bill payment instructions.

Stringing together these four components, along with a lot of hard work and a little bit of luck, will create an on-line banking operation that appears to work. However, poke a bit deeper and it quickly becomes apparent that the current on-line banking system is an unsustainable collection of technical compromises and quick fixes that at some point will have to be restructured. The current structure suffers from three major problems: lack of messaging standards, batch updating and core processing fragmentation.

As far as the prospects for a single standard go, while all of the parties acknowledge that there should be a single standard, the issue is getting caught up in a much larger strategic battle between banks and nonbanks. For now, suffice it to say that in the near term the industry looks as though it will have to deal with two standards, Open Financial Exchange (OFX) and Integrion Gold.

Prospects for solving the core processing problem look bleak. Banks are incredibly loath to risk even a slight disruption of their core processing operations. One middle-of-the-road solution that appears to be gaining momentum is for banks to install a separate core processing system that only services their on-line banking operations. This database is in turn linked to the bank's old core processing environment by a series of custom bridges. This solution allows banks to avoid disrupting their core processing environments, but enables them to enjoy the benefits of a unified core processing environment.

For those banks that do offer on-line banking and promote it heavily, on-line banking customers represent, on average, five to seven percent of their customer base. What's surprising is that some smaller banks have even higher penetration rates. Indeed, the highest penetration rates of all can be found in small credit unions targeted at prime demographic groups, such as the First Technology Credit Union, which currently has 26 percent or 15,000 of its 57,000 members using its on-line banking services.

This higher level of penetration in banks that offer and promote their services suggests that while only two million households used on-line banking in 1996, there were probably another three to five million households that would have used on-line banking if their banks offered and/or promoted it.

There has been a similar and perhaps even more intense explosion of interest in on-line trading. The growth of the on-line trading market has been so strong and so pronounced that even full-service brokerage firms such as Merrill Lynch and Smith Barney have announced plans to offer at least account access over the Internet.

While growth to date in both of these markets has been pronounced, what has attracted everyone's attention is projected growth going forward. Survey after survey indicates that there is strong consumer interest in using on-line banking.

On-Line Explosion to Continue

On the supply side, large cost disparities between on-line and physical distribution channels are driving financial institutions to provide on-line offerings. In the banking industry, both PC and Internet banking channels are orders of magnitude cheaper than traditional branch transactions.

The cost advantages provided by on-line trading are evident in its dramatically lower commission rates. Before on-line trading, the "discount" trading industry typically charged $50 to $100 a trade. Today, several on- line trading companies are now charging under $10 a trade.

Projections of the on-line banking market over the next few years range from 10 million to 17 million households. While this is a fairly wide range of estimates, it is still at least five times as big as the current market. As for on-line trading, it is projected by Forrester Research to grow to over 10 million accounts during the same time period.

Amidst all of this growth, there are a number of trends currently under way which have the potential to dramatically change the financial software sector. These trends are setting off a series of battles that threaten to cause a good amount of friction in the sector over the next few years as major players battle over much larger strategic issues.

The most pronounced trend over the past year or so has been a distinct, accelerating movement away from proprietary "dial-up" networks and toward the Internet. This is not altogether surprising given that the Internet enjoys many practical advantages over "proprietary dial-up" services such as lower cost, easier administration and greater flexibility.

There is much more to this trend toward the Internet than just simply switching the way that customers access their financial accounts. At the root of this migration is a realization that the Internet may well indeed enable the financial services industry to play a new and potentially powerful role in the world of EC: the "Integrator."

The role of the integrator is fairly straightforward. The integrator's job is to bring together all of a customer's financial and transaction information into a single place and then to organize and analyze that information on behalf of the customer. On a practical level, this means that the integrator uses the Internet to automatically collect account and transaction information from all of a customer's financial institutions (brokerages, insurance companies, banks, mutual funds, etc.). By bringing together all of this information, the integrator would take much of the drudgery and hassle out of financial management and planning and would then be in position to offer a number of value-added services.

Seizing Information, Customer Control

The attraction of the integrator role extends well beyond just improving customer service. The role of the integrator affords an institution two major advantages: information control and customer control.

The integrator has access to all of a customer's financial information. Theoretically, this "information advantage" should enable bank personnel to identify cross-selling opportunities and to more accurately assess both credit risk and potential profitability. This advantage should allow the integrator to proactively respond to customer needs and price more aggressively than others can.

Not only will it take time to set up a relationship with an integrator, but the integrator will also, much like a good assistant or travel agent, "learn" a customer's needs and behaviors over time. These two features will make it difficult for customers to switch Integrators as they would have to train their new integrator over a period of time. These increased "switching costs" should bolster customer retention and control.

Ultimately, the goal of the integrator is to generate an increasing stream of revenues from its customers by not only charging for financial management services, but also by generating fees from customers' purchases of other companies' products.

One group of firms that are clearly sold on the power of integration is the PFM firms. Growth in the PFM marketplace has already slowed considerably and the PFM firms realize that unless they make their programs much easier to use, they will be permanently stuck with a total potential market size of just 10 to 15 percent of the population. The Integrator role, with its ability to facilitate the automatic collection, input, and analysis of customer data, represents the panacea that the PFM firms have been looking for.

An Integrator is a Customer's Best Friend

Not only will an integrator see every single transaction that a consumer or business conducts, but it will become a customer's trusted advisor and counselor.

This role should put integrators in position to collect a small advisory or brokerage fee on almost every single transaction their customers undertake. Such a position has for some years represented the "holy grail" of the EC industry.

However, these firms face one critical problem in their drive to position PFMs as integrators. How do they automatically collect information and assemble it into their programs? There are currently no standards in place for passing messages between the different components of the financial software sector. This lack of standards not only creates a number of technical difficulties, but it also makes it very difficult to combine data from two different institutions.

So the two major PFM firms, Microsoft and Intuit, along with Checkfree, proposed the OFX message standard. With the OFX message standard in place, a consumer's financial statements could automatically flow into the PFM programs, and the PFM companies would finally be able to fulfill their vision of becoming integrators.

The only problem with this vision is that PFM companies are not the only ones interested in being the integrator. Financial institutions understand the power and attractiveness of the integrator role just as well as the PFM firms do. What's more, financial institutions can clearly see that the integrator will have an important, highly valuable role in the world of EC. Naturally then, financial institutions also want to serve as integrators.

Of the financial institutions, banks have taken the most aggressive tack to date-but also the most predictable. Their solution to the problem of who will be the integrator relies on two well worn tactics: shared ownership organizations and regulatory protections.

The banks' drive to establish shared ownership organizations started in 1995 when Bank of America, NationsBank and Royal Bank of Canada got together to buy Meca Software. In buying Meca, the banks hoped to create a credible alternative to Intuit's Quicken and Microsoft Money.

The same three lead banks who purchased Meca also started, in partnership with IBM and fourteen other banks, a company called Integrion. Integrion's first major act was to announce that it was proposing its own "single message standard" called Gold. Like OFX, Gold is designed both to solve the technical problems facing banks and to enable integration of information.

Like the banks, the technology firms have also resorted to their own tried and true methods. Microsoft and Intuit have put the full weight of their development efforts behind the OFX standard in an attempt to establish OFX as the de facto standard simply by the sheer weight of its support within the financial software sector.

Microsoft has been methodically using its considerable influence and standards-setting prowess to line up developers across the industry to support the OFX standard.

No Cutting Out the Middle Man

In response to this situation, many banks have begun to think about cutting the PFMs off at the pass, by integrating customer information before the PFMs receive it. By performing this integration "server side" on their own computers, banks would eliminate the need for consumers to even use a PFM.

But alas, this strategy is almost impossible for most banks to execute because they lack the technical infrastructure necessary to integrate account information. With neither the time nor resources to adequately restructure their internal systems, many banks may be forced to accept the OFX standard and the triumph of distributed, multi-tier computing that it represents.

One wild card in this entire situation is that consumers may choose to do their integration elsewhere; not at their bank or on their PC, but at a third-party Internet site set up expressly for the purpose. These "Integrator" sites would be dedicated to providing integration services and, unlike banks, consumers could be sure that these third-party sites did not have a vested interest in selling them a particular financial product. And guess who owns the two biggest Integrator sites? Microsoft and Intuit, the two biggest PFM firms.

Although banks don't realize it, the emergence of the integrator site has the potential to turn the tide of battle. Unlike the software world, where the PFM firms dominate the distribution channels and can leverage their installed base, the Internet presents banks with a comparatively wide-open situation. Banks can open their own integrator sites and substitute them for the centralized databases and unified processing environments that they currently lack.

At first glance, Microsoft's involvement in this battle seems non- sensible. Why should Microsoft risk alienating all of those big banks? After all, these banks buy tens of thousands of copies of Microsoft's system software and applications. Surely Microsoft is more interested in that business than it is in being an Integrator.

In fact, Microsoft is indeed interested in the banks' internal business and that interest extends well beyond selling copies of Windows and Word. What Microsoft is most interested in is the core processing business, otherwise known as the back-end or the glass house. According to the Mentis Corporation, while banks spent only $100 to $200 million on on- line banking software in 1996, they spent about $2.7 billion on back-end processing, data centers, and applications. Microsoft can read a pie chart as well as the next company, and there is no doubt that Microsoft recognizes that the glass house is where the money really is.

What company now dominates the glass house in most financial institutions? Why, it's IBM. Who is Microsoft really after? That's right, IBM. No doubt Bill Gates has dreams of clusters of NT workstations humming away where once sat monolithic IBM 3090s; no doubt Lou Gerstner has nightmares that Bill Gates' dreams will come true.

When put in this light, Microsoft's support of OFX begins to make sense. In supporting OFX, Microsoft is actually supporting a standard that will help it in its drive toward the glass house. OFX requires "OFX servers" which are essentially Windows NT PCs that process OFX messages as quickly as possible. Theoretically, a group of OFX servers linked or "clustered" together is capable of processing millions of OFX messages a second. After awhile, OFX servers start to sound a lot like, well, IBM mainframes.

IBM is well aware of Microsoft's intentions and has done a masterful job convincing banks that Microsoft is trying to take their customers, when Microsoft is, in all likelihood, more interested in taking IBM's customers.

So how is Microsoft going to finesse the issue of OFX and its Integrator potential? Doesn't the integration issue have the potential to derail Microsoft's entire glass house initiative by antagonizing the banks? Yes, it does.

Who'll Win the Battle of the Elephants?

So rather than risk that possibility, Microsoft is going to great lengths to assure banks that it has no designs on their customers. They are even offering pieces of their PFM (Microsoft Money), called Active X controls, to the banks for free and allowing the banks to incorporate these controls into their Internet sites as they see fit.

Will this strategy work? So far Microsoft appears to have the upper hand, but it's not clear that IBM will be outmaneuvered. This "battle of the elephants" is far from over, and the outcome is still uncertain. However, one thing that is certain is that any firm which gets in the middle of this battle risks being trampled to death.

Piper Jaffray's Bill Burnham authored "The Electronic Commerce Report."Preceding excerpts are part of an exhaustive undertaking to be published by McGraw-Hill in January 1998. For more information, contact 612/342-6000.

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.