Associating computers with profits


26 Mar 2003

When the tech downturn first hit such was its severity that tech firms of all sizes scrambled for ways to reduce costs. Recruitment plans were frozen, marketing campaigns ditched and salary bonuses scrapped.

Another easy target were research and development (R&D) budgets, the thinking being it didn’t make sense to plough money into product development when demand had fallen away so sharply.

One of the biggest companies to buck this trend was the world’s fourth largest software company, Computer Associates (CA). While industry-wide R&D spending was taking a sharp turn downwards, CA was spending more than ever on its new product pipeline, says chief technology officer, Yogesh Gupta (pictured).

“Over the last three years we have spent over US$2bn on R&D, which is a 66pc increase over the previous three-year period,” he remarks.

CA spends a whopping 23pc of its revenues on R&D, which is high even by the standards of an industry that has historically spent heavily on research. The company has over a dozen R&D centres dotted around the globe employing some 4,000 people. The network used to include a small development centre in Dublin before the company decided to consolidate that research area back to the US.

Supporting such a weighty infrastructure might seem suicidal at a time when IT budgets are being slashed, but Gupta says it is underpinned by a sound business case. “When times are tough, if you don’t invest in the future, you don’t necessarily have the right products when times improve. So we very much believe that investment in R&D is critical and we’ve made a conscious decision to do so.”

Of course you have to be in a position where you can afford to ship this level of R&D spend. Fortunately for CA it is. The company has built up a US$4bn cash pile in the last few years despite the downturn.

Gupta is a rarity in the IT industry – a loyal company man. He’s been with CA since 1989 – “which in our industry is almost three lifetimes,” he quips. During his tenure, he has seen the company come through a frenetic acquisition phase where it cherry-picked companies and technologies around the world to enhance its market position. That strategy, which was driven by mercurial CEO Charles Wang, brought with it some problems such as integrating the technology, not to mention the employees of acquired companies. The accession to CEO of Sanjay Kumar in August 2000 represented a break with this avaricious takeover policy. Kumar preferred instead to take a path of organic growth and look within the company for new products and technology.

“When Sanjay took over one of the mandates he put out was that he wanted to focus on internal development and come up with new technologies and products,” Gupta recalls.

This renewed focus on R&D was very welcome but simply boosting the dollar investment does not guarantee better products or more of them. A more useful measure of achievement is patent output, Gupta feels. “Spending is really not success; spending is just the cost side of the equation. But when Sanjay asks what have you produced, we can say that over the last 12 months we have delivered over 200 new products. We have filed for more patents in the last two years than we did in the previous five. And we have been granted more patents in the last year than we were in the previous three.”

Gupta proudly adds that CA outstrips its competitors in the number of new patents filed. Almost half of its 200 patents are in the area of infrastructure management systems – software that manages networks and provides intelligent analysis of data. The company also holds many patents in the areas of storage management and e-security.

The patent output, no more than the amount of money invested in R&D, necessarily translates to the bottom line but Gupta feels it is highly significant for a range of reasons. “Firstly, it demonstrates to ourselves and to the world that we are doing innovative technology work. Secondly, it protects our intellectual property. Finally, it means that we can go out and license our technology, although this is a small part of our business.”

CA has been traditionally reluctant to license technology for fear that licensees would use it to develop competing products. However, it has begun to license technology that is not part of its core portfolio.

CA engages both in pure and applied research. Pure research – general research that may or may not be applicable to CA’s core areas – is done exclusively through sponsoring projects at a number of top universities in the US such as the Massachusetts Institute of Technology (MIT) and University College Los Angeles, plus several international ones. Occasionally, these projects will yield technology that can be taken and used by CA in its own business. For example, a predictive analysis server that is now used in CA’s network management systems came out of research done at Case Western Reserve University in Ohio and two soon-to-be-launched products are based on work done at MIT.

“If we do 10 projects at universities, there might be one of them which we can use. When we do internal research we expect that ratio to be higher,” Gupta remarks.

Internal research is business specific and very much shaped as a result of CA’s own core business and by wider trends and opportunities in the marketplace. Unsurprisingly, many new research projects relate to the burgeoning area of wireless technology. One important area of focus is the development of management solutions for Wi-Fi networks and the company already has products in the marketplace which address the security aspect, that prevent unauthorised users from accessing Wi-Fi networks for example.

Another potentially profitable area of exploration is web services – software that allows different applications from different sources to communicate with each other. Here, CA’s focus is on developing products that make the use of web services reliable, safe and secure. “In our industry, the hype always precedes reality, quite often by a long period of time, but in web services we are starting to see applications being built and readied for launch,” Gupta says.

Perhaps the strangest technology that CA has been involved with is a refrigerator that has a separate insulated oven built-in. The device stores food during the day and then automatically switches on in the evening to cook the meal as the tired commuter is on his or her way home from work. No, CA has not entered the consumer electronics market; it provides the Linux-based software that manages and secures the device.

It is through investments in research projects such as these, which, Gupta believes, will keep CA at the forefront of its field.