)
Articles Archive
 |
July 2004
King Of Convenience
The need for Sarbanes-Oxley certification has boosted sales of internet service risk systems, says web pioneer Martin Chavez of Kiodex.
By JAMES OCKENDEN
Martin Chavez has banished the acronym 'ASP' within his New York-based company Kiodex. The concept of Application Service Provider - application access through the internet - has come a long way, he says, since the early days when ASP was nothing more than a copy of an application in a different location, and he refuses to tar his latest services with that brush.
"The first round of ASPs didn't do anything interesting," he says. "Companies bought a bunch of hardware, a bunch of connections to the internet and licensed a bunch of copies of, say, Excel. In essence, you had Excel installed on some computer at the ASP's network center, and the end user was simply getting a window onto that machine over the internet."
The business models failed, says Chavez, because no value was added. "The companies didn't own the intellectual property, and so it was a flawed business model from that perspective. But even if you did own the intellectual property, you then had a severe architectural problem. You were taking an application built for a client server -and simply extending the monitor cable over the web - that's all the early ASP models were doing."
Chavez recalls his early development work at Goldman Sachs. "My focus was on the energy capabilities of our risk management system, the Goldman Sachs Financial Workbench. We took this system that my colleagues and I had built and then I worked with another group of colleagues to make that available over the web.
"The rumor is, no one is using it anymore. And that's for any number of reasons," he laughs. "But what I found there was you cannot take a client-server risk management system and just do it as an ASP. You can fake it pretty good. But it's like putting lipstick on a pig. It's still a pig."
Chavez insists it is impossible to take an existing product and render it web-enabled. "I firmly believe you need to bite the bullet and rewrite from scratch. No one has successfully taken their client-server application and retrofitted it for the web."
This view will be interesting to other companies seeking to capitalize on the growing interest in web-based risk services. For example, OpenLink - traditionally a company that has avoided the ASP model, even through the web development boom of the late 1990s - now says it is planning a web service. "The interest is starting to get to a point where we can justify an ASP model - so don't be surprised to see an ASP service coming from OpenLink," says its chief executive Coleman Fung.
Fung was unable to divulge further details of the planned service, nor whether the service would be based around existing models and code or entirely fresh code.
"I think [some vendors] are going to learn the hard way. They will be entering a different business," says Chavez. "If they just create deployments for each one of their customers and serve them over the web - well, you're back to the old ASP model: it does not scale. If you haven't built your application so that you can scale it, you will become each customer's IT shop, which has no value."
Customer base
The very nature and appeal of a web-based service is in the outsourcing, says Chavez - it is to client-server applications what McDonald's is to catching, slaughtering and cooking a cow yourself. And it is this convenience which, according to Chavez, is why it has always appealed to smaller shops.
OpenLink's Fung acknowledges the demand for web services has rocketed since the industry retraction of 2001/2002."You had a lot of big energy companies - Enron, Dynegy, Mirant - who were doing a lot of risk outsourcing for the smaller utilities and energy companies. Through profit-sharing agreements, these big guys would trade excess capacity for the small guys. And when the big guys went, the small guys had to step up and do the job themselves. And they don't have the infrastructure and resources to support robust risk management software like ours - so they're looking for an ASP model."
Chavez agrees, although in his experience it is not only smaller companies that seek the simplicity of the 'McDonald's' solution. One of the first clients for Kiodex's WorkBench was Sumitomo, in Japan. "Sumitomo has its own subsidiary, SC Infotech, which supplies all its systems. But it had an unpleasant experience with its copper trading risk management in the mid 1990s. And so it looked for someone else - someone who was an expert in the risk management business, to be accountable for every single number that got calculated. It didn't even want to take two numbers and multiply them together -it wanted it all done by an outside expert, someone who is 'on the hook'," says Chavez.
Of course, it generally will be smaller companies - airlines towards the smaller end of the global 200, for example - that will seek the services of a company such as Kiodex or, when it is developed, OpenLink's ASP service. And one feature they will seek is the provision of outside data, such as forward curves.
For this reason, at the heart of Chavez' latest offering is what he calls "the messy work". "The front-month settlement price of West Texas Intermediate is the same for everybody," he says. "Why make each company go and figure it out, and feed it into their systems separately? There is so much background information - holidays, Platts prices, models - that is the same for everybody. There is no value in having a highly customized options model."
The benefit for the customers is this background work - and the cost - can be spread across a wide base of users.
Cost savings
It is the total cost of ownership that is the biggest draw, according to Chavez. And now firms are scrambling to meet the requirements of the Sarbanes-Oxley act, a further upsurge in business has stimulated. "I'm looking at, off the top of my head, 10 companies who have signed up who said "no" a year ago - customers who previously said: 'Hey, I can do it in Excel or OpenLink'. Now they're coming to us and saying: "Our auditors told us to turn off that system we're using." These companies are now looking at the total cost of ownership again. And they're looking at our system. Instead of going out and buying a risk management system, paying consultants to install it, paying people to deploy it and manage it, and then hiring some auditors to give them a certificate, they can just turn on the tap and pay us some money."
Copyright © 2004 Incisive Financial Publishing.
All rights reserved.
Used by permission.
First published in Energy Risk's Technology Special Report - July 2004
)