Wired
From: Global Custodian

The custody business is not about what is happening in a given bank's systems and technology, says Open Information Systems' Tom McMackin. It is about what is being put on the customer's desk. by Charles Ruffel

Tom McMackin is principal of Open Information Systems (OIS), a Sandy Hook, Connecticut-based firm that specializes in technology applications for the securities services arena. McMackin spent 13 years at Citibank prior to founding OIS in 1994, much of that time in a Citibank division that sold technology to other financial institutions.

Today, OIS provides Information Manager, an electronic delivery product, to some 25 institutions, including Chase Manhattan, Fidelity, and SEI. McMackin talked to Global Custodian about the technology challenges now facing custody banks.

GC: You were almost a Citibank lifer. What gave you the confidence to start your own company?

McMackin: Because of what we were doing at Citibank, it became clear to both myself and Tom Groshans, who was working with me, that the real action was taking place at the customer's desk, not on the mainframes deep within banks that processed the data. We felt there was an opportunity to help reshape the way information was delivered to the customer's desk.

GC: How did you fund the start-up?

McMackin: We started with a 90-day contract to build an electronic delivery system for Chemical Bank, and then were able to take that prototype and further refine it for the likes of SEI and Fidelity. We were in the fortunate position of being funded by our customers.

GC: What do you do at OIS?

McMackin: We do two things. First, all financial institutions have this mess of data, and we attempt to normalize it so information can pass between the various points between back-office mainframes and the customer. Right now, that information flow is, by and large, staggeringly inefficient. SWIFT, for example, is in reality a wasteful way to communicate information. Because you can't presume what information the party at the other end is in possession of, you have to assume the worst case and convey all the information. Instead of an ISIN, you have to send complete security master details. Since we control both the sending server and the receiving client, we streamline this process so you only send the information the other party needs, is missing, or has changed.The second thing we do is design how all this data is presented on the customer's desk. That's where the real revolution is taking place. The custody business is not about what's happening in a given bank's systems and technology, it's about what is being put on the customer's desk. It's turning data into information that can be used to support decisions.

GC: How so?

McMackin: Perhaps the most visible development is that banks are discovering that putting traditional workstations on customers' desks is too expensive, and, with the developments of the Internet, there is a real alternative. The future is browser access, and the challenge is to find your way around the volume constraints and limits to flexibility that browsers impose. Browser-based systems are nothing more than dressed-up versions of the dumb terminals of the 1970s. The trick will be to keep the flexibility and added value that a local database provides, while lowering the support cost through thin clients. But the revolution is more fundamental than that. For all the talk, the problem is most custody banks still retain their back-office mentality. Look at what all the banks are doing-they are spending fortunes on the Y2K issue, on the euro, on recreating their back offices. While this investment is an absolute requirement to be in the business, it does nothing to differentiate their services. Back-office processing has become a commodity to the customer-the differentiator is the desktop.

GC: Expand on your thinking as to the revolution taking place on customers' desks.

McMackin: Right now, the world is made up of what we call 'fat' clients, but every bank is trying to convert them into 'thin' clients-with a thin client, the data and the application programs reside at the bank, whereas with a fat client, the data and software reside on the client's desk. The industry is moving away from fat clients because the cost of supporting such clients, most notably the data management and software distribution costs, are sky-high. But where banks are running into trouble-admittedly some more than others-is that, as far as customers are concerned, one size definitely does not fit all. The large and sophisticated plan sponsors want real flexibility. They are not happy with the sort of data delivery product that the banks are now designing. They need so much data that existing Internet technology simply won't cut it.

GC: So, if banks can't afford fat clients, but large customers aren't happy with the thin-client approach, what's the solution?

McMackin: Our approach has been to provide a single system that can deliver either a fat or thin client. Then, you can graduate clients from the one client to the other as communications technology improves and as messaging technology improves. One of the things we do is massage the data so that what is communicated to clients is only the data that has changed, not the data they already have. What clients want is real-time data-the end-of-day feeds that now dominate the scene will soon be irrelevant. Clients want to make decisions based on what is happening right now, and custodians using last night's pricing are not in a position to give that to them.

GC: Presumably, one of the reasons banks are focused internally is that their systems are close to imploding. How is it that these banks have gotten themselves in this position?

McMackin: All these banks are plagued by over-customizing over the last two decades-they have all these different systems that can't communicate with each other. There's nothing wrong with the basic hardware and software that these banks have. Instead, the problem is that, over the years, they have attempted to customize a solution for every division in the bank and for every other customer. Moreover, they prefer to find solutions internally wherever possible. At the money center banks, in particular, the systems group has to fail first before they will even think about going to an outside vendor. Obviously, that's an expensive and potentially calamitous way to go about evolving your technology.

GC: Why is custody littered with so many technology failures?

McMackin: In large part, because most of these big institutions make a lousy environment for technology wizards. The way you advance at banks is to work yourself into a management position, and creative technology people don't necessarily want that. They may not mind working 100 hours in a week, but they look for a certain type of environment. Big institutions, by and large, don't provide anything close to the environment they want. When I was at Citibank, they shut off the air conditioning in summer and heat in the winter at 6 pm. We get most of our work done in the hours they discourage you from working. And when it comes to this sort of work, it is the quality, not quantity, of people that matter. Big institutions think they can solve these problems by throwing a lot of people at them, but that's just dead wrong. Do you know there are only five people who run Microsoft Word on the development side? Numbers can get in the way in technology, but don't try and tell that to a money center bank.

GC: Will the present round of bank mergers help the survivors in their ongoing struggle to be more technologically adept?

McMackin: Quite the contrary, in fact. There are more and more redundant systems out there, and more and more customization is necessary to make everything work. Mergers focus banks on their internal workings. As I've said before, what goes on in the inner workings of a bank is not what matters. What matters is what you put on your customer's desk. Banks that are unable to grasp that central fact are in trouble.

Return to News Section