Twist & Route
Order-Routing ISVs Evolve with the Market
by Bennett Voyles
You’re a full service futures firm. You have to provide an order routing system for retail investors. One for proprietary traders. One for institutional traders. One for fund managers. You have to provide access to multiple exchanges around the world. In the U.S., you have to provide access to the floor as well as to various electronic order routing systems designed to support the floor. This access has to be high speed, reliable and fully redundant.
Build or buy? Building puts an astronomical strain on resources. You not only have to build the infrastructure, but the commitment to maintain all the connections with the exchanges, update functionality and provide customer support pulls resources away from your core business. On the other hand, if you buy an off-the-shelf package from one of the independent service providersƒis your customer yours or the ISVs? You want to staple this customer to your firm so you start to add a few bells and whistles that are strictly your own. You don’t want your service provider to develop them because he will want to share them with his other clients and your competitive advantage goes away. You’re also worried about the health and well being of your service provider. It hasn’t turned a profit in years. It has huge development costs because all of the exchanges have different interfaces and pump out new and complicated products at the speed of light.
“Order routing has become much more complex,” says Alex Lamb, director of development for Goldenberg Hehmeyer. “It has become widely recognized that electronic order routing is not just a peripheral thing that someone might choose to have or not to have—it is an essential part of being an FCM today.”
With all the attention given to order routing, the number of customers that take advantage of these systems is surprisingly low. Some firms say that less than half of their clients are connected to the markets electronically. “The reality is that there are still only thousands of screens, not tens of thousands of screens,” says Lamb. “The audience has not grown as much as everyone thinks.”
Tim O’Shea, executive vice president and director for institutional sales and marketing for Fimat USA says they are getting the most interest from hedge funds and professional trading groups. “We are really just beginning to see the demand for electronics from the institutional side. A lot of our clients still want to place a phone call to a Fimat broker on the floor or upstairs in the office. They rely on our broker’s expertise to provide market commentary, trading strategies and information flows from the exchange floor. Our brokers then decide to execute the order either in the pit or on the screen, depending on where the client will get the best price.”
How it All Began
Exchanges initially controlled order routing. They provided terminals to customers to access their electronic trading systems. Trading desks could be cluttered with up to 15 terminals. Traders learned how to operate key systems for active markets, but couldn’t remember how to work systems they didn’t use everyday.
In 1998, when Liffe introduced its Connect platform, it revolutionized the order routing business. Unlike other exchanges, Liffe elected to provide APIs (application program interfaces) to which ISVs could connect their front-end systems. This introduced a new level of participants into the order entry process. “We found ourselves in the position of working with some thinly capitalized startups to send orders to exchanges,” says Bill Sexton, chief operating officer for Refco. “It created an increased level of functionality for customers, but it initially caused concern and required us to closely manage the process.”
The initial systems were by most accounts rough and ready tools, typically either a difficult-to-use system developed by a major futures commission merchant looking to encourage orders over a new medium or the ungainly brainchild of a well-heeled floor trader worried about his future and a young techie hoping to get rich quick. “Immediacy wasn’t the primary objective,” remembers Lamb who co-founded one of the first ISVs, Trading Technologies in 1994. “It was volume of transactions and accuracy of transmitting and receiving the data back.”
Despite the difficulties, there was some excitement involved. “It was just a novelty to be able to extend out that front end,” recalls Christian Hauff, an e-commerce salesman for Barclays Capital in New York. “Even though that front end was a block—it was a brick, you couldn’t really customize it or anything—at least it allowed you to see a market and trade.”
Over time, traders and vendors say, the technology throughout the system has become much more sophisticated. Out went time-wasting nuisances such as pop-up windows. In came one-click orders. Today, traders seem pretty happy with their systems.
Profitability Remains Elusive
Many of the original ISVs have either merged or gone out of business in the past five years. Too many vendors, not enough resources to build and support the complex infrastructure, and lack of standardization among exchanges all contributed to their demise. Even for the market leaders, life is far from fat: Trading Technologies, for example, proudly announced its first profitable quarter this past January—after nearly nine years of operations. But TT, as it’s known to traders, is private, and isn’t saying how much black ink it used in the fourth quarter 2002.
At the same time, a number of the FCMs who had initially thought of front-end trading screens and order routing as a source of competitive advantage began to see it simply as a source of unrequited cost. Some of the largest brokers have stepped away from that business. Fimat was one of the first FCMs to offer a proprietary order routing system, Fimatrade, to customers. According to O’Shea, they decided to abandon the system last year because the ongoing costs of development were prohibitive and took focus and resources away from their core business. “When we decided to build, there was not a better alternative,” he said. “Once we felt the ISV solution could reduce our resources and was equal to or better than our in house platform, we chose to go the ISV route.”
Barclays Capital is one of the few FCMs that still supports its own complete system, according to Hauff. But even Hauff is quick to add that many of Barclay’s larger clients prefer to use their own screens which Barclay is happy to mix-and-match with its own infrastructure.
In spite of the exit of a number of FCMs and the drastic reduction of startups, the typical ISV is still in a fairly tough spot. Like contestants on a TV reality show, the good news is that they’ve survived; the bad news is that everyone’s having fried snake for dinner again.
Nevertheless, ISVs are an important link in the food chain. “There are a few key players now that seem to be well positioned,” says Sexton. “They seem to have found their niche in serving different customer bases.”
One Size Does Not Fit All
There are as many different approaches to order routing as there are firms, but FCMs agree on one point: no single solution fits all of their customers’ needs. Professional traders gravitate toward screens with the most sophisticated trading tools, one-click order entry and high speed. Among the contestants, professional traders look for a particular type of tool that reflects their discipline, says Lamb.
Retail customers want a less complicated approach and more protection against errors. “Retail customers require a more intuitive platform for their style of trading,” says Sexton.
Fund managers want to plug their computer generated models into an order routing system so when their system generates a buy or sell signal, they can send the order directly to the exchange. CTAs also have allocation issues. The CTA applies an algorithm to distribute fills among his customers and he wants to be able to apply that algorithm electronically. Firms that handle a lot of fund managers are writing that functionality into their order routing offering.
Some customers, such as CTAs, want to be able to use their own front ends and want just an API from their firm. But offering an API creates new headaches for FCMs. “You have to be able to control who, how many users, and how they connect to your APIs,” says Lamb. “A customer could redistribute it to his customers. Now it is possible to have users out there that the FCM doesn’t know anything about.”
The ISV Shuffle
To garner market share from the various customer bases, ISVs are scrambling to differentiate themselves. “It’s connectivity and it’s functionality on the screen,” says Rob Moore, general partner of Marquette Partners in Chicago. “Those are the two deals.”
Under the broad heading of connectivity, ISVs compete on speed, reliability and number of exchanges they support. Connectivity used to be a matter of creating links between the few electronic exchanges and trading firms. Now, a number of ISVs have created linkages with so many different exchanges that one of their key offerings has become the ability to offer access to many markets around the world. GL Trade, which reportedly has the largest number of connections, links more than 70 markets and over 300 brokers via the GL Net.
While ISVs that emphasize routing and connectivity tend to dismiss front ends as a commodity, a few front-end-focused firms have succeeded in making it a point of differentiation. Today’s customers don’t want to see just one market on a screen, they want to see a screen full of markets, maybe several screens of markets. They want to be able to make calculations between those markets. They want stronger risk-analysis tools. They want quick execution. And they want it all done very, very fast. Slight differences, say some industry observers, such as Trading Technologies’ reversal of mouse functions so that execution takes place not on the downclick but on the release, can have a big impact.
ISVs can offer functionality across exchanges, standardizing some trading features and offering functions that the exchange itself does not support. For example, not all exchanges accept stop orders. Some of the order routing platforms, O’Shea explains, have the ability to have stop servers which means that these orders get held back at the brokerage house and when that price is reached at the exchange, the orders are released and filled at the stop or stop limit price.
Traders often become attached to particular screens, and don’t want to switch. “They’re a fairly superstitious crowd,” says Harris Brumfield, CEO of Trading Technologies, which markets some of the most popular screens in the U.S. market. “They think the screen makes a difference, so they don’t want to change it. If you’re making a living with somethingƒit gets pretty sensitive.”
Yet loyalty isn’t synonymous with monogamy—an unfortunate fact for front-end vendors. Hearing traders talk about the ISVs’ different capabilities makes using multiple providers sound like a perfectly understandable strategy, like a golfer talking about why he packs several different clubs. Moore says Trading Technologies has better functionality for futures, while RTS is better for options. And some firms don’t write to certain exchanges, he adds: “Trading Technologies for instance does not write to the Italian exchange, whereas RTS does.” However, Moore adds that most trading firms will keep a majority of their traders on a single system in order to simplify their risk management.
Beyond providing front-ends and connectivity to exchanges, ISVs are scrambling to differentiate themselves in a variety of ways. Most providers will do extensive and expensive custom programming for big clients, according to Joe Campagna, director of financial services consulting for RMS McGladrey in Chicago. Some develop white-label versions of their front-end for important brokerage firm clients.
Others see potential in bundling other offerings into the system. Rolfe & Nolan executives, for instance, believe that the company’s experience as a back office provider will serve it well in adding more services, such as stronger risk-management functionality.
What’s next? Sexton sees competition coming from the equity side of the business where system providers are looking for scale and opportunity because they are not seeing the equity volume they were seeing a few years ago. “The systems in the equity world are more mature than what we have been seeing in the futures world,” says Sexton. “It’s easy to route an order for a distinct futures contract to the exchange it trades on. However, if some of the recent competitive changes between exchanges come to fruition, we will need to respond to routing orders to competing pools of liquidity. This introduces a higher level of smart order routing than has previously been required in the futures world. If you need an order routing system to look for best bid or offer, and to route orders based on some complex set of parameters unique to each FCM, that introduces a level of complexity that futures ISVs have not had to deal with.”
Right now futures and equities operate in two very distinct worlds: different exchanges, different clearing organizations, even different IT departments within the firms. As the demand to offer cash, equities, and futures forex products on a single screen grows, the incentive to blend the two worlds will increase.
An Uneasy Partnership
ISVs and FCMs have become uneasy partners in order routing. According to Lamb, ISVs are hungry for design input from their users. FCMs frequently discover gaps in the tools, but are reluctant to share that information with their ISVs because they don’t want their competitors to benefit from their discovery. “ISVs need to understand what is proprietary and what they should build for you, possibly for you to then keep for a period of time to give you an edge,” says Lamb.
Because of its diverse customer base, Refco maintains proprietary order routing systems and links with the exchanges in addition to partnering with ISVs and offering multiple platforms. “While recognizing the importance of the ISVs in the chain and to remain responsive to customer needs, Refco developed an ISV strategic outsource relationship,” says Sexton. Refco joint ventured with EasyScreen to offer a customized solution to customers called EasySolutions. “We get the benefits of the traditional ISV without having to directly manage the development. By dictating the functionality requirements we also ensure that our offering to customers will be unique.”
O’Shea says Fimat made a decision a long time ago to have a multiple platform approach with a single connection. They provide clients with a direct connection through a financial extranet, Radianz, which has relationships with multiple brokers, exchanges and clearinghouses. A customer who wants to connect to Fimat’s electronic trading platform and the International Securities Exchange, for example, can use a single connection to the Radianz network instead of maintaining multiple connections. “Once a customer is on Radianz, we are capable of providing him any one of our ISVs’ systems through that one connection,” says O’Shea. “Right now we could have one trader that uses one of our ISV systems because of its exchange access and a trader right next to him could use a different ISV system because he likes the functionality better and only needs access to a few critical exchanges. This approach makes it very easy for us to change from one ISV we support to another.”
On O’Shea’s wish list is an application service provider (ASP) that would provide exchange and client connectivity as well as support and maintain the front-end GUI (graphical user interface) and FIX interfaces for our clients. “With commissions decreasing, it’s becoming more and more difficult for brokers to remain profitable. The concept of outsourcing the e-trading infrastructure would go a long way in helping us reduce our investment and free up resources to provide better service to our clients because we wouldn’t have these large projects and infrastructures to support and manage.”
Early attempts at establishing an ASP model have been too expensive. “We’re finding now that when we do a comparison of an ASP solution vs. an in-house solution, it’s still better for us to do it in house,” says O’Shea. “The marginal costs will go down as we add business. Right now, we haven’t found an ASP model that provides us that kind of scaling capacity or pricing model.”
Stay Tuned
But the general consensus seems to be that the number of players in the ISV game is likely to dwindle further sooner or later.
Some observers are skeptical about the long-term health of the exchange network business. Philippe Buhannic, CEO, TradingScreen, a New York-based trading service provider, cites the trend toward greater consolidation among both brokerage houses and exchanges as one reason to doubt the strength of this segment. A second threat, he says, is the growing adoption of the Financial Information eXchange (FIX) protocol, the electronic messaging language that a group of software providers, brokers, and financial markets are trying to establish as the common transactional vocabulary of electronic securities trading. The more standardization there is between exchanges, the less value connection-makers should have to offer.
David Hall, director of marketing for patsystems in London says strong exchange volumes have postponed a shakeout in the ISV space. “Whilst exchange volumes continue to grow so strongly, I guess a number of firms will grow with it,” he says. “But ultimately there will be a shakeout. It won’t quite be last man standing, butƒthe time will come,” he says.
Some observers fear that this ferocious competition will lead eventually to two or three strong firms that have tremendous hold over the market. But Campagna, who advises major FCMs and other large players on their technical systems and strategy, thinks otherwise. He says that the economics and the industry structure are such that the ISVs will never be terribly strong.
Campagna’s prediction: in five years, most of the ISVs will be gone or purchased by an FCM. “I think you’ll get a couple of firms who will probably buy the ISVs just to keep their systems up and running,” he says. Some will be absorbed. Others may remain stand-alone companies for awhile, he says.
Brumfield seems more optimistic that growing futures trading volumes may continue to help businesses in the space. “If Eurex and LIFFE can do [the volumes they now trade] I can’t imagine what the U.S. [exchanges] could do when they go totally electronic,” he says. “I think the possibilities are enormous.”
Leslie Sutphen, president of Financial Markets Consulting, Inc., and Mary Ann Burns, editor-in-chief of Futures Industry magazine, contributed to this story.
ISV
An Independent Software Vendor makes and sells software products that run on one or more computer hardware or operating system platforms.
Source: Whatis.com
API
Application Programing Interface is a language and message format used by an application program to communicate with the operating system or some other control program such as a database management system or communications protocol. APIs are implemented by writing function calls in the program, which provide the linkage to the required subroutine for execution.
Source: Techweb.com
ASP
Application Service Provider is an organization that hosts software applications on its own servers within its own facilities. Customers rent the use of the application and access it over the Internet or via a private line connection. Also called a “commercial service provider.” With the advent of the Web browser as the universal client interface, the ASP market is expected to grow rapidly.
Source: Techweb.com
Bennett Voyles is an independent business journalist based in New Jersey.
