Intranet use will explode over the next five years, because it delivers simply and cheaply many of the functions that previous system architectures have failed to provide. But as businesses load more onto them, those virtues may be buried.
So says a new report from London-based research company Ovum Ltd. Ovum says by 2002, 78% of client systems connected to a corporate network in the U.S. will have intranet browsers installed on them. In Europe the figure will be 64 %, and in Asia-Pacific it will be 41%.
Intranets (which the report defines as private networks of Web servers) allow the creation of corporate information networks that are easy to use, seamless and global in coverage.
Furthermore, intranets are easier to manage and offer a simple, universal cross-platform client, using smaller applications, the report says. It added that spending on intranet services will grow to more than US$1 billion a year by 2002 as companies find that intranets are capable of being used for a much broader range of applications.
The intranet is appealing for several reasons, the report's authors say. It can be installed by piggy-backing on existing and more expensive client/server systems, and the client software is either cheap or free. It also cuts the cost of software installation, maintenance and training. And it is ideal for use by new-style virtual organizations or companies with mobile workforces.
But, the report warns, the cost advantages could be short-lived as companies try to do more with their intranets. "As the applications become more complicated and move beyond information publishing, their cost will increase," it says. "It may be inexpensive today, but it will not be tomorrow."
Most intranets today are dedicated to the publishing of information - corporate telephone directories, company procedures, news items - but Ovum believes this picture will change rapidly.
The authors see information publishing as just the first of four distinct but overlapping waves of development. The second wave will be the use of intranets for informal collaboration, the kind of job currently handled by the likes of Lotus Notes, Microsoft Exchange and Novell Groupwise.
Those products will continue to offer superior functions, but their market will be eroded away at the low end, the report says.
This means that browsers will offer basic groupware functions, while the separate groupware products will be increasingly confined to high-end applications that augment the functions of the embedded groupware.
The third wave will come with Web-enabled business applications, although the authors warn: "Much of the complexity that intranets were claimed to avoid will come back when organizations integrate Web technology with business applications."
The fourth wave will be the integration of workflow, or formal collaboration, into the intranet. But the authors said they fear that today's leading workflow packages do not scale well enough to be used on an intranet.
By 2002, the authors predict that information publishing will account for just 38% of intranet server usage. Informal collaborative working will rise to 32%; business applications to 26%; and formal collaborative applications to 3%.
The report contains case studies from several large organizations, including Buffalo General Hospital in Buffalo, New York, pharmaceutical manufacturer Glaxo Wellcome PLC and the New South Wales State Rail Authority in Australia.
From these, the authors draw 10 main conclusions on how to implement an intranet:
1. Intranets are typically piggy-backed on another IS investment - you do not need to blow your IS budget to implement one.
2. Start with information publishing applications. Follow with Web-enabled collaboration and transaction-oriented applications.
3. Establish policies about who is responsible for what concerning infrastructure, content, the development environment and what the intranet will contain - this will avoid trouble later.
4. Roll-out your first applications quickly - intranets do not require a 12-month requirements analysis to start.
5. Do not get hung up on the technology evaluation if you want to do Web-based publishing - the available products offer virtually the same information publishing capabilities and can be swapped out quickly - providing you avoid proprietary features.
6. Most intranet benefits are not directly measured in financial terms - keep an eye on anecdotal benefits to help justify your intranet.
7. Some users will be skeptical about intranets - implementers have identified the need to pay more attention to user resistance and cultural issues as key elements to ensuring user acceptance. When asked what they could have done better in implementing intranets, organisations said that they would have invested more in user education about what intranets were and what they could do.
8. IS drives intranet implementations, but user ownership of the content makes them work. IS should focus on policy and infrastructure management issues.
9. Implementers are keeping a watching brief on Java and ActiveX, but are implementing using CGI and other standard Web technologies.
10. Intranets will not reduce IS's management burden, but they will increase what you can do with the same resources.
The report is entitled "Intranets for Business Applications, User and Supplier Opportunities." Ovum is at http://www.ovum.com/.