Third-party software costs are often cited as the No. 1 complaint of mainframe customers. What is CA doing about that?
[That] I won't get the issues I heard in 2001 with customers, that they feel CA is unfair in its pricing. We want to make customers feel like the price-to-performance of the platform is going to get better. We're doing things that are aggressive, taking advantage of the zIIP processors with our technology, which I don't think IBM envisioned any of the ISVs [independent software vendors] doing. We're aggressively pursuing that and our customers are starting aggressively to take advantage of them.
The mainframe is one big thing and the line items are one line item, so it's very visible. When you buy distributed servers, everything is in small pieces, but when you add it all up, it dwarfs the mainframe in terms of the cost to customers. With server sprawl and all this complexity, the mainframe doesn't look so bad any longer, and I think that customers are understanding that this mainframe, when you look at the total cost of ownership, it's a cheap, cheap platform.With this so-called mainframe revival, how often do you see customers that are new to the platform?
There are scenarios, like in China and Hong Kong, but these are banks that are buying an entire infrastructure. They didn't start with two branches, bring in a mainframe and grow up through it. In Brazil we're seeing some of it. But that's not the place you're going to see major growth in this market. You're going to see it in the high end. The guys who came up and developed companies in the last 10 years tend to be distributed. The people that are going to gobble them up are going to be mainframe guys, the big guys. That's what fuels this growth. From an IT perspective, what problems do big mainframe shops run into, when they start acquiring a bunch of smaller companies?
One of the big problems our customers are throwing at us is that they've acquired 10 companies over the last 10 years, and although they've gone through consolidation physically, they haven't done consolidation at an enterprise architecture level. Now they're at a point where they want to do that and want to have a standard mainframe environment. CA has a lot of mainframe software products. Are there customers that either don't use all the software they have installed or aren't getting enough value out of the products they use?
We do three-year transactions on average, so we've gone through two major recycles at least per customer since 2000. It's an intensive exercise, where customers look really hard at what in the portfolio they in fact use or don't use. So the concept of shelfware as it relates to the mainframe, I don't believe to be a reality any longer, at least in the case of CA.
On the features and functions side of it, the high-end customers are always pushing the envelope with us. But there are customers that have an environment that works, and they don't want to introduce change. Change can be a bad thing in IT. They're looking at it like, "If it ain't broke, don't fix it." Because of that, there's a lot of value that's not being gotten by the customers that probably could, and as time goes on, it increasingly becomes an issue.What does CA do to help solve that problem?
So if you've got 20 products, we're trying to find the four, five, six that really contain greater value. The user groups will say pretty consistently that support has gotten better, the mainframe is getting better. That's not what they beat me up on. They will beat me up on things like your install procedure, how you install products, that to learn one doesn't mean you know another. They'll slap me around on that; they'll say things like it takes them seven years to get through our entire portfolio in terms of upgrading releases given our schedule of bringing up new technology. So making sure that these things are alike [is] incredibly important.