Financial institutions flocked to account aggregation - and got more headaches than anything else. Will the technology ever fulfill its promise?
By Steven Brull
December 2001
Institutional Investor Magazine
The future of full-service private client management is meant to be taking shape at J.P. Morgan Chase & Co. The firm’s high-net-worth customers can log on to an online portal, enter a password and voilá: Up comes a consolidated real-time view of their assets, including marketable securities, partnerships, trust accounts and bank balances. Clients and their advisers can massage and analyze the data in myriad ways as they weigh investment options that dissect cash flow, taxation, inheritance and other complex planning issues.
This slick service comes courtesy of account aggregation, a state-of-the-art technology that burst onto the financial services scene in 1998 with all the hype and fanfare of the then-raging Internet. In the days when pundits and prognosticators still spoke of killer apps, aggregation appeared to many to be the ultimate killer - giving individuals the kind of data retrieval and analytic resources previously available only to big, rich institutions. It promised to revolutionize all of consumer finance.
But as a closer look at J.P. Morgan shows, the revolution still hasn’t booted up. Only a handful of the firm’s private banking customers are using the service, and most of the data they see has been pulled together manually by 100 or so employees of Kinexus, a start-up company under contract to J.P. Morgan. If these clients didn’t each have a minimum of $25 million in investable assets, the bank wouldn’t bother to offer the service - it couldn’t afford to.
“We’ve gone through a cycle where anything that was e-related needed to be worked on and was sexy. Now we have to go from the sizzle factor to the substance factor,” says Yawar Shah, an executive vice president of e-business strategies at J.P. Morgan. “There is no killer app.”
Aggregation still holds undeniable appeal as a financial tool, but like many Internet phenomena it was oversold by hyperenthusiastic technophiles. Fewer customers than predicted have embraced the technology. One reason: The mechanics of collecting accurate, detailed and complete data and delivering it to the customer remain difficult, largely because of inconsistent storage formats and conflicting communications standards. It has proved much costlier than expected to offer the service, which is why some big institutions like J.P. Morgan are providing full-blown aggregation services only to their wealthiest customers.
All this is quite a change from the heady early days, when the financial world buzzed with talk about the powerful new applications. Aggregation worried many in the industry who saw it as a threat to their control over customers. Firms feared that if their clients could get a complete reading of their total financial state and see all of their records from the institutions holding their various accounts, they would have less need for their main banker, broker or adviser. Aggregation threatened not just disintermediation but also a shift in the balance of power, away from companies that had historically profited from their hammerlocks on data and toward more informed and demanding online consumers.
Still, many financial institutions were quick to embrace the technology, at first out of concern that if they didn’t, they might lose customers to competitors who did. But these firms also concluded that creating comprehensive snapshots of their customers’ finances might help not only to solidify relationships but to make those relationships more profitable and long-lasting, particularly if the firms could use some of the information to cross-sell products.
But as the financial houses rolled out aggregation services, they found them to be more complex and time-consuming than expected. It’s easy enough to download an account balance via the Internet - millions of banking and brokerage customers do it on their own all the time, and they can authorize their banks or other aggregation suppliers to do it for them. But to present the information in a form that allows for sophisticated analysis in combination with other data sources and software tools is far more challenging and expensive. With profits under pressure and tech budgets constrained, most financial institutions have held back on developing systems that don’t offer immediate revenue or cost benefits.
One result has been that only a relative handful of consumers are using online account aggregation. Anil Arora, CEO of Yodlee, the leading provider of aggregated account data to financial institutions, reckons that more than 2 million individuals have signed up for the service. That figure, extrapolated from Arora’s observation of Yodlee client institutions, represents about one out of eight online financial consumers in the U.S.
Arora insists that aggregation gained mass popularity faster than any other financial technology. “It has grown faster than electronic bill payment and faster than the introduction of ATMs decades ago,” he says.
Whether it has lived up to its hype, however, is another matter. Earlier this year Celent Communications, a Boston research firm, projected that the U.S. user base would reach 9 million by year-end 2002 and 35 million by the end of 2004. Quadrupling the Yodlee estimate in a year doesn’t seem to be in the cards, especially as some analysts believe that Arora’s number may be overstated. Cambridge, Massachusetts-based Forrester Research, for example, counts fewer than 1 million users in the U.S. And it terms almost half of that group “pseudo-aggregators” who use personal financial management software such as Quicken rather than Yodlee or similar technology.
Providing aggregation services to this pool of customers, Forrester concludes, is a losing strategy. Catherine Graeber, the former Wells Fargo & Co. retail banking executive who is leading Forrester’s aggregation research, explains that financial institutions cannot generate enough revenue from cross-selling products to cover the $90 cost of providing aggregation to an account over three years. Indeed, by her estimate, they would lose $54 per account over that period.
“Financial firms are rushing into aggregation without a solid business model, and they won’t succeed,” warns Graeber. “Consumers just aren’t linking the competitive accounts institutions need to mine the data. And many institutions lack the technological infrastructure to analyze that data to create a customized offer.”
So is aggregation just another Internet aggravation? No way, say its many adherents. One possible solution may be to focus the service, as J.P. Morgan has, on the high end: the extremely wealthy and the group known as the emerging affluent, who have anywhere from a few hundred thousand to a few million dollars to invest. These clients have a greater need for the service - and a greater ability to pay for it. Another, related approach is to view aggregation as a tool for client service or, ultimately, as part of the expense of winning and keeping customers.
J.P. Morgan has made aggregation a top priority at its private bank, which has $300 billion under management, but doesn’t offer it to the hoi polloi in its branch system, the old Chase Manhattan Bank retail accounts. High-net-worth customers throw off enough cash to cover the cost of data capture and formatting, a laborious process that makes up roughly half the cost of client servicing, says Michael Holden, who is in charge of implementing new technologies at J.P. Morgan Private Bank.
“Account aggregation started as one of those buzzword concepts,” Holden explains. “The reality is that we want to do it to give a better picture to clients and plan better investment strategies. From the adviser’s view, it’s a way to give more effective advice. Good advice is built off good information.”
Few would argue with that proposition. Citigroup, E*Trade Group, Fidelity Investments, Merrill Lynch & Co., Morgan Stanley and others have all invested in aggregation precisely because they regard information as power; their customer relationship management strategies demand a broad and intelligent approach to client services. What could be better than a system that breaks through institutional barriers to provide a full accounting of customers’ wealth?
“We see tools like this as very, very key for customer retention,” says Jane Jamieson, a Fidelity executive vice president in charge of institutional clients. Fidelity has much to gain by knowing and serving its customers better. With the nation’s largest pool of 401(k) retirement assets, the firm wants to prevent attrition when participants roll over their accounts. More than 70 percent of all U.S. rollovers, which total $250 billion to $300 billion annually, don’t stay put, according to Darlene DeRemer, who heads e-business advisory services at NewRiver, an Andover, Massachusetts-based consulting firm.
But like J.P. Morgan, Fidelity is trying to reconcile promising theories with practical complications. Steven Elterich, president of the Boston-based mutual fund giant’s e-business unit, is clearly sold on the technology. “We look at it as a way to develop a relationship with customers,” says Elterich. Yet he notes that aggregation is technologically immature; instead of simplifying the gathering of and access to customer financial data, it puts a greater burden on firms like Fidelity to translate a jumble of often incomplete and incompatible data feeds from external sources.
“The hardest part is how to make it easier to make sense out of all this stuff,” says Elterich, who emphasizes the importance of meeting clients’ needs through a combination of compelling products and better technology. “For some of these things, it’s impossible to figure return on investment.”
FINANCIAL AGGREGATION HAS BEEN THROUGH a life cycle - and a hype cycle - even more condensed than those of the Internet itself. It began in 1998 with the formation of VerticalOne Corp., an Atlanta software company whose founder, Gregg Freishtat, claimed credit for the idea. VerticalOne saw banks as its primary customers, and the following year the company was acquired by Atlanta-based S1 Corp., a leading supplier of software to online banks.
By that time VerticalOne was slugging it out with Yodlee, a Silicon Valley start-up. Where VerticalOne and S1 stressed their friendliness and compatibility with banks and other financial institutions, Yodlee emphasized its hip brand name and invited consumers to use its yodlee.com Web site as a personal financial portal, which some banks and brokerages viewed as a competitive threat.
As financial institution executives woke up to the technology and its implications, many were wary of allowing their customers’ information to be downloaded by, or with the cooperation of, competitors. Some bankers joined consumer advocates in criticizing aggregation as a license to invade customers’ privacy and as a legal-liability can of worms; it was unclear, for example, whether the sender, the receiver or the transmitter of data would be responsible for inaccuracies. Few wanted to play ball with Yodlee at all because of its direct-to-consumer portal approach.
But by early 2000 the climate had turned more benign. Bits, the technology policy arm of the Financial Services Roundtable, a Washington trade group, convinced VerticalOne, Yodlee and other technology companies to accept standards and principles that underscored financial institutions’ central importance in the customer relationship and in safeguarding the privacy and integrity of data. Then Yodlee, which in February 2000 had hired former Gateway executive Arora as chief executive officer, rechanneled its marketing energies in a business-to-business, rather than business-to-consumer, direction, and financial companies responded positively.
By mid-2000 Yodlee had signed a contract with Citigroup to power its myciti.com aggregation platform. America Online, Chase Manhattan, Fidelity and Morgan Stanley soon jumped on the bandwagon and became customers. Meanwhile, VerticalOne was building a customer list that included CNBC.com, Wells Fargo and the Yahoo! Finance portal. Financial institutions were less concerned that aggregation would encourage customers to move their business elsewhere; they hoped instead to use the technology to upgrade service and discourage customers from defecting.
But at the same time, the Internet economy was cooling, and S1’s revenues and stock price declined precipitously. One of the company’s restructuring moves was to sell control of VerticalOne to Yodlee in January of this year. Suddenly, there was a dominant aggregation technology player - but in a softening market. With financial institutions’ online customers cutting back on their stock trading and clients’ net worths heading south, there was less demand for the constant account updates that aggregation services offered.
The demand for aggregation is also limited by the way in which data is collected, interpreted and analyzed. The basic technique, pioneered by VerticalOne and Yodlee, is called screen scraping. Once programmed with a consumer’s account numbers and passwords, a screen-scraping system can troll the Web for bank deposit and credit card balances, airline miles and other such data. In the U.S. and abroad, Yodlee can scrape from some 2,500 sites, each with multiple types of accounts that must be interpreted.
That’s where the process gets trickier. Assuming that the data is accurate and sufficiently detailed - with screen scraping, there is always the risk of hitting sites that have crashed or have not been updated - financial institutions want the information formatted so that it can be directly input into spreadsheets, then sliced and diced in asset allocation models, retirement planning calculators and the like. But the data feeds are nonstandard; there is no single technical protocol to ease translation and formatting at the receiving end. And without some form of standardization, financial institutions will remain hobbled in their attempts to deliver the timely data and analyses that a truly robust aggregation service requires.
Fidelity’s Elterich, for one, believes that standardization is inevitable because the value of aggregation will compel financial institutions to cooperate. They might, for example, adapt or improve a common protocol such as Open Financial Exchange, or OFX, which online financial institutions now use to process bill payments. “Down the road we’ll see more data exchange capability,” which will help financial firms to offer advice and enable clients to act on what they learn by moving assets around, says Elterich.
Notes Robert Hall, president of revenue enhancement at Carreker Corp., a Dallas-based financial software and consulting firm: “Aggregation is not just information; it’s really about decisions and actions. It creates an opportunity to provide more services.”
But Hall points out that financial institutions have long struggled to derive profits from customer information. Access to comprehensive, quality information - and the ability to interpet it wisely - will be key, and that’s where an industrywide standard can help.
Bits has organized an aggregation services working group to tackle the standards issue. The group consists of more than 300 people from 68 organizations - including banks, securities firms, insurance companies, aggregation vendors, portals and regulators - who have been meeting biweekly with the goal of drafting a policy statement early next year.
“There is a consensus that we have to go beyond screen scraping,” says Bits staff director Leslie Mitchell. She can’t say where the effort is likely to end up - whether at a standard data protocol or perhaps with a neutral information clearinghouse that would take responsibility for routing aggregation data among institutions in compatible formats. But whatever the solution, agreement among such a diverse constituency will come slowly and implementation even more so. Markets and technologies almost always move faster than deliberations on standards.
A further complication is the proliferation of technology suppliers. Yodlee, despite its acquisition of VerticalOne, is not so dominant that it can dictate a standard or resolve all conflicts. Other vendors are carving niches: New York-based CashEdge, which has made inroads into the Canadian banking market and as a supplier to Yahoo!, integrates aggregation with online payment services; Kinexus, also in New York and backed by J.P. Morgan’s venture capital subsidiary, J.P. Morgan Partners, specializes in data collection and analytical tools for high-net-worth clients. Woburn, Massachusetts-based ByAllAccounts targets family offices and other advisers to the wealthy.
Also selling technology for brokers and advisers is Advent Software. The San Francisco-based company uses Yodlee for bank account aggregation within its broader Advent Trusted Network program, which receives account data directly from about 80 financial institutions and feeds data to some 1,500 firms, according to Steve Lewczyk, the company’s vice president of corporate development.
In November Advent announced an agreement to integrate its data and financial tools with Microsoft Corp.'s .Net technology; by early next year the system will be available to more than 6,500 Advent clients under the label MSN Money Professional. This combination provides a glimpse of what an open data collection infrastructure might look like, but it’s still a proprietary initiative and not a truly standard “open architecture.”
“We don’t see a vendor coming out with all the data you’re going to need,” notes John Patterson, president and CEO of Business Logic Corp., a Chicago-based software integration company. Patterson predicts that financial firms will take matters into their own hands, starting with bilateral agreements to exchange data. “Somebody has to flinch first - say, Fidelity doing a deal with Citi,” he says. “Once that happens, people won’t want to be left out.”
That might take care of the supply, but what about demand - will customers care? Optimists say that the technology and services being tested and refined by banks like J.P. Morgan in the high-net-worth market will eventually filter down to millions of emerging affluent households.
Even Forrester’s Graeber, who is highly critical of aggregation strategies to date, sees potential there: “The emerging affluent already have complex financial lives. Yet they’re not locked in with loyalty to an adviser,” she says. Such openness could spark new rounds of competition with aggregation at the core as banks try to enhance their relationships with brokerage services and as brokers like Charles Schwab Corp. and E*Trade expand into banking.
“Mass affluent” customers with $500,000 or more to invest are Fidelity’s stated target, and Graeber sees the firm as a well-positioned “trusted brand for unbiased advice.” And one year ago Fidelity, which had been offering portfolio planning and modeling tools to some of its 9.5 million employer-sponsored retirement account participants since 1988, began to offer these tools also to its retail customers, who have 8.4 million accounts. More than 126,000 individuals had registered as of November.
Later, Fidelity plans to download stock trading data directly into Intuit’s TurboTax and other tax-filing programs. Scott Gilmour, a Fidelity senior vice president, says that such seamless data feeds could save an active investor hundreds of dollars, because accountants charge an average of $5 for each stock transaction they input.
But seamlessness is still a ways off. Fidelity has 30 to 40 engineers and software developers working full-time to improve the technology and its user-friendliness. And Fidelity’s income from mass affluent accounts is less likely to cover the near-term costs of aggregation than are the profits from J.P. Morgan Private Bank’s multimillionaires who, for now, are more typical of the technology’s active users.
Sanjeev Dheer, CEO and co-founder of CashEdge, sees a natural evolution toward a mass-market payoff. “It’s going from euphoria in the beginning to a much more serious group of users, who are smaller in number but more interested in analytics and transaction capabilities. Then it will grow steadily,” he says.
But it could take years of slow, incremental gains before millions of people are taking aggregation for granted. As Fidelity’s Elterich notes, the technical obstacles aren’t trivial, and they require an industrywide response. Says he, “We have to go to the next step.”
