Future proofing

01 Oct 1996 View Comments
A Computing logo

A year ago, the Anite Group was in a sorry state. The UK network hardware, integration and software specialist was struggling to survive in a highly competitive market and saw # 19.4m wiped off its pre-tax profits for the year.

A boardroom shake-up and company restructuring followed. But instead of slowing down its spending, Anite's response was to pour money into research and development (R&D) and encourage a culture where innovation can flourish. Chairman Alec Daly says sales are on the up and the company is on the road to recovery.

It's a well-worn cliche that the UK has always been brilliant at coming up with new ideas, but useless at turning them into commercial products.

But Anite, which used to trade under the name Cray Electronics, is a good example of how UK IT companies are beginning to break down that belief in the 1990s.

Unfortunately, the DTI's UK R&D Scoreboard report shows that the R&D expenditure of other countries puts the UK to shame (see box).

At 7.7% of its sales, Anite's R&D spend is nearly double the UK average.

In the DTI's report, it ranks 10th in the league table of spending in the UK electronics sector, behind foreign-owned high-turnover companies such as ICL, IBM and Hewlett-Packard, which carry out research in the UK.

Mike Pilbeam, research director of Anite's network hardware division Case Technology, says that although the company spends a sizeable chunk on R&D, it is not concentrating solely on research. He believes modern companies have to be very focused on R&D and that spending has to be closely linked to the needs of the business.

'The days of inventing a black box which does something interesting and then making a market for it are long gone,' he says. 'In those days, a gap in the market would be identified before letting the engineers go away for up to a couple of years to develop a solution. The solution would then be passed to the marketeers who would try to sell it.'

Today, he says, the process is one of carefully listening to customers' needs, then creating a product at the right price and in the right timescale.

The R&D boffins, production people and marketeers would all work together in what he calls 'virtual teams' during the whole cycle. 'This means that right from the start people from all parts of the business are there to make sure there is sanity added all the way through the cycle.' It also prevents a product from being out of date by the time it goes on sale.

But tying R&D closer to short-term business needs is not the right environment to create the technology breakthroughs of the future, which require research over a longer period of time.

One organisation which has witnessed this change in corporate R&D is CRL, a 60-year-old division of EMI. CRL, originally the corporate R&D department of Thorn EMI, is famous for inventing the computerised brain scanner. This was before Thorn EMI decided to concentrate on its music business and sold off many of its divisions. As part of a smaller and more focused company, CRL 'became irrelevant', according to Nicholas Walker, its deputy group manager of systems and applied science. So the company started to carry out R&D for 'anyone who would pay for it'.

'(As part of Thorn EMI) CRL was like a central research lab that did research partly dictated by the company and partly by its members' own research interests. But like the rest of the world it has changed very dramatically,' Walker says.

Because CRL is dependent on performing research for outside companies, most of its work is related directly to products rather than future technologies.

Few European IT companies can afford to carry out the type of long-term research that IBM, Intel and Microsoft can. Even European-based heavyweights ICL and Siemens had to join forces to set up the Germany-based European Computer Industry Research Centre.

CRL still does some long-term research, though this tends to be funded by the EU or government agencies. One such body is the European Space Agency (ESA), which commissioned CRL to conduct research into document retrieval.

'This was a pure research programme. ESA just wanted a demo and a report,' Walker explains, adding that this project had, like many pure research projects, actually resulted in a commercial application. CRL approached document retrieval system specialists Fulcrum for assistance. 'We felt we had developed a commercial product and so we developed a relationship with Fulcrum. We hoped the product would be commercially exploited, but we would never have had the money to do the research on our own.'

Walker believes that increasing commercial pressures do not necessarily threaten pure research. 'Being charged with a commercial approach is not wholly bad. It encourages many of the good ideas that we come up with to actually become commercially successful, instead of languishing on a shelf,' he says. Commercial pressure helps IT research to be targeted, he adds: 'There are an infinite number of things you can do with computing, it is more of an art form than a science.'

Component software company Ilog knows that to survive it has to be clever with its R&D. It regards itself as a reseller of R&D, saving its customers from having to carry out extensive research themselves.

According to Mari Georges, Ilog's associate director of research and development: 'We do a lot of specialised R&D, which saves our customers from having to do the same thing. Instead of building whole applications from scratch they can economise by using our components and put more money into their own specific development.'

Ilog was set up in 1987 as a developer of artificial intelligence systems and in 1990 began developing re-useable software components based on C++.

'The company started with a technology push, as is often the case with small start-ups, but we soon realised we couldn't carry on like that.

You have to be market-aware and as a result strike a balance with your research,' says Georges. 'We keep an eye on our customers and try to be attentive. Therefore our research is quite focused.'

Although Ilog has an emphasis on R&D, all its research ends up as a product or an enhancement to an existing product, says Georges. 'There is a small amount of long-term research and then there is the R&D which is focused on the next six months.'

To maximise the benefit of long-term research, Ilog has partnerships with other software developers, such as Iona, Object International and Borland. 'We exploit these alliances because this brings synergy to our research. When it's of mutual benefit, we strike an alliance,' Georges explains.

Working with academia is another way for Ilog to get the technology know-how it otherwise couldn't afford. It has close relationships with the University of Brussels and Esprit, the EU technology R&D programme. Each year Esprit pumps millions into R&D by stumping up 50% of the costs of selected research, normally undertaken by consortia of suppliers, universities and users.

'Through Esprit projects we work with outside specialists so we can bring in experience we wouldn't otherwise be able to pay for ourselves. Also, with Esprit programmes we tend to get a built-in set of beta sites. The end-user partners always make nice beta testers.'

Working with partners is not the only way for companies to develop future technologies. Financial reporting software company SAS Institute says it has struck a good balance between the short-term product development needs of the company and its wider, long-term technical aims.

Steve Morton, SAS' UK technical manager, claims SAS spends up to a third of its turnover on R&D, although much of this involves product development rather than research. However, SAS staff are given freedom to pursue their own personal research projects, but in a way that reflects the company's commercial priorities.

'We encourage personal research, but we are also a commercially minded company, so a lot of effort goes into developing existing products. At the same time we still do specialised research,' explains Morton.

Instead of having a central lab where staff are charged with coming up with specialised future technologies, SAS allows each engineer time to pursue their own research interests in addition to their normal duties.

'Our research people have commitments to work on our financial software products and depending on where they are in a product's cycle they will either have no time at all or a small amount of time to pursue their own ideas,' says Morton.

An example of this is the development of a neural network analysis tool for the company's reporting software. One of SAS' scientists had a personal interest in neural networks and was allowed to spend time on combining the technology with SAS' tools. He presented his work, which combines the artificial intelligence characteristics of neural network technology, with SAS' financial analysis application, at the company's annual user group meetings in 1992 and 1993.

'He came up with very novel ideas,' Morton says. 'A few users picked up on what he was doing and so did some people in SAS. All it lacked was a proper user interface, which we then went and built. Now we have a finished neural network application.'

Morton claims that if SAS staff come up with a research idea that could bear fruit, they are allocated resources to investigate it. This results in small teams of up to five working together on projects. Such flexibility, Morton claims, is a model for R&D that balances innovation with the need for products that sell.

It is easy to get bogged down with the everyday stresses of work and meeting deadlines. But if more companies encouraged and listened to their developers' ideas, the IT industry could see more innovations making it commercially.


'A warning signal is flashing,' declares Ian Taylor, minister of science and technology, in the preface to the latest DTI report on R&D investment.

UK companies are consistently failing to invest as much in research as foreign competitors, according to the report, UK R&D Scoreboard, published in June.

Although investment in research by UK companies increased in 1995, it was the lowest rise in five years, the report found. R&D spending by UK companies rose only 4.2% in 1995, compared to 6% and 7.5% in the previous two years.

These figures are all the more worrying when compared with the 9% increase in UK company sales - twice the rate of investment in R&D - and an 18% leap in profits. Until this year, UK companies had been closing the gap on foreign firms in terms of R&D investment. This was the first time since the scoreboard was launched in 1991 that the UK had slid back.

The biggest investors in R&D in the UK are in the pharmaceutical industry.

Three out of the top five R&D spenders on the UK scoreboard are pharmaceutical companies. GlaxoWellcome spent # 1.2bn - more than double the amount invested by the top electronics company (see graph).

Figures released at the beginning of this month by the Office for National Statistics revealed that the wider picture is just as depressing. Spending by the Government, private companies and universities on R&D increased in real terms by 3.7% from # 13.8bn in 1993 to # 14.6bn in 1994.

Richard Freeman, corporate chief economist at ICI, describes the 'weak performance' revealed by the R&D Scoreboard results as 'mostly disappointing'.

The world average spend on R&D for electronics companies, including computer hardware manufacturers, is around 11.5% of sales (see graph). UK companies have steadily fallen from just above, to just below, 4% since 1991.

International support service companies, including software and IT services firms, spent around 6% of sales on R&D, but their counterparts in the UK spent 4% in 1995.

Robin Bulow helped compile the scoreboard during a secondment to the DTI from Barclays Bank. He says there may be mitigating circumstances for why UK IT companies are not investing in R&D at the same rate as foreign competitors. 'The trend is of particular concern in the IT sector where there is an emphasis on new products. You have to question whether they are spending sufficiently,' he says.

The UK and the top 300 international R&D scoreboards were prepared from the annual reports and accounts of the companies, published up to 31 May 1996. To obtain a report free of charge, call (0171) 215 1217.

Reader comments
blog comments powered by Disqus
Windows 10 - will you upgrade?

Microsoft has made an early version of Windows 10 - its next operating system - available for download. The OS promises better integration and harmonisation across platforms, including mobile and desktop. Will your business be upgrading?

37 %
27 %
15 %
21 %