denotes premium content | Feb 8 2012 
Feature
posted 31 Aug 2004 in Volume 1 Issue 3
A fine balance
Intranets can be a powerful tool, but they will never completely replace people creating and sharing business knowledge. If you think they can, your system will become a crutch that doesn’t support you. Warwick University Business School’s Harry Scarbrough looks at how intranets can fail or succeed from a knowledge-management perspective.
One of the events that led to Britain entering World War I was the assassination of Archduke Franz Ferdinand of Austria in Sarajevo on 28 June, 1914. And although it eventually would have been figured out, a Reuters sub-editor in London initially thought he was looking at the result of the 2:30 race at Longchamps race track, just outside Paris. The message read: Sarajevo (1), Ferdinand (2) Assassiné (3).
It would be comforting to think that with the benefit of the latest technology such mistakes no longer happen. The reality is that there have been numerous examples in recent years of the blatant misuse and misinterpretation of information. These examples range from the most advanced areas of rocket science (Mars missions that failed when imperial units were used to calculate propulsion, but were unwittingly entered as metric units by NASA), to the public services that we rely on every day. Failures in policing, social services – even decisions to go to war – have all been blamed on the poor use of information.
These failures have much to tell us about the scope for the application of new information technologies in business. The massive increase in the use of intranets, portals and many other IT-based systems has undoubtedly had many benefits for industry. But as the above examples show, there are limitations to these systems. Like other terms such as ‘military intelligence’ and ‘adult male’, the words ‘IT solution’ have become a classic oxymoron.
One way forward in dealing with the limitations of IT systems is to make them more sophisticated and wide-ranging. However, such attempts usually end up demonstrating that more information does not substitute for better knowledge. Take the example of the aerospace firm that developed an expensive information system to coordinate the development of a big aircraft project. The system assembled detailed information on all the thousands of parts and components that made up the plane, giving designers what they thought was a complete picture of its specifications. Unfortunately, while the system captured information on part size, shape and function, it did not cover the actual weight of components. Being massively overweight was, as its designers quickly realised, a liability for an airplane.
If more systems aren’t always the answer, what is? One important lesson from these failures is to recognise that in itself, information cannot create the knowledge we need to interpret and use it effectively. So we need to understand from the outset that intranets and other systems have in-built limitations; they cannot substitute for the creation and sharing of knowledge within the business. What’s particularly worrying about this point is that in the past few years many businesses have adopted intranets as a way of developing better knowledge management. In many cases, results were little short of disastrous with vast amounts of money being invested in systems that added little or no value and that certainly did not manage knowledge. The example of the anonymous ‘Ebank’ organisation came out of my research in this area. This bank’s experience, though extreme, seems to have been only too typical of similar projects in many other large organisations elsewhere (see inset).
There is still much to learn about the factors that enable some firms to make much better use of their employees’ knowledge than others. A new research programme, ‘Evolution of Business Knowledge’ (www.ebkresearch.org), addresses many of the outstanding puzzles confronting managers in this area. It is already clear, though, not least from the failure cases described above, that many of the answers lie in the intangible ways businesses work. Five factors, in particular, influence how much and how far people will share knowledge:
Shared practices
People share knowledge with others who do the same kind of work and solve the same kinds of problems. For example, doctors in the NHS will share experiences with other doctors, but not with nurses and certainly not with managers.
Shared identity
This involves feeling like that we are all part of the same group. The sense of belonging to a group is one of the reasons why some of the biggest barriers to sharing knowledge are national characteristics. Various research studies highlight the role of national identity in knowledge sharing. For example, British people are often happier to exchange ideas among each other as a group, rather than with other groups of a different nationality, notably the French. This issue of identity is one of the reasons why the debate about Britain’s role in Europe is not just a political issue; it has real impact on how well and with whom we collaborate in business.
Trust and reciprocity
Attempts to get people to share knowledge often fail because there is a lack of trust. This deficiency may be between employees and managers, but it can also occur between functions (how often do marketing and operations management people talk to each other?) and different locations.
One big pharmaceutical company was dismayed to discover staff in its different research and design labs were more likely to share knowledge with scientists outside the company than with people in other labs within the business. This is less surprising when you know that several of these labs were under threat of closure and the scientists in them felt they were competing with other sites. Trust is difficult to create, but even more so where people feel they are competing against each other.
Social networks
One of the most obvious things about any organisation is that the most important and valuable knowledge is channelled not through the intranet or corporate communications, but through informal social networks. This applies just as much to knowledge of how to solve day-to-day problems (‘Why can’t I get this machine to work?’) as to grapevine knowledge about the firm’s future strategy or who to contact to get things done. In short, knowledge flows best when people are connected socially, and not just through the organisation.
Dialogue as the basis of knowledge
The communication of information depends on codification, or shared standards and grammar. Sharing knowledge, however, is a much more complex process and usually works best when it’s interactive. This method allows people to develop a dialogue so they can explore the thinking and assumptions behind the messages they are receiving (the World War I example shows the value of this).
There are many ways of putting these ideas into practice within firms. Some organisations have focussed on communities of practice as a way of addressing these issues. Here, they are recognising that people doing the same things and sharing a similar identity are a powerful channel for knowledge sharing. An extreme example of this outside the business environment would be skateboarding. Obtaining knowledge about how to skateboard is not something you can easily do by reading books or even watching skateboarders in action. You need to be a skateboarder, actively involved in a community with its own rituals and culture, to really learn from your peers how to skateboard.
Another way of addressing these factors is to bring them into the arena of IT systems design. Intranets and other systems can support and be supported by knowledge sharing if they are properly aligned with the factors discussed above. One example of the successful use of IT to enable knowledge sharing is provided by Buckman Laboratories, a multinational chemical analysis company based in the US (see inset).
This means recognising that IT networks are not the same as social networks. It also means socialising IT networks so that they leverage and enhance existing patterns of social interaction. This involves designing systems that connect people and allow them to express their personal and group identity – not designing systems that codify knowledge or present an impersonal face to the user. Crucially, it also means enabling them to have a dialogue with each other. This is one of the reasons why e-mail, for all its problems, is so widely used. But other systems can benefit from some of the usability of e-mail while bringing in a greater discipline, such as the filtering of content and a collective memory for what has been said. Systems that adopt a conversational idiom, such as online forums and even web-logs, are capable of doing this.
Conclusion
Companies can certainly gain great benefits from the use of intranets and other systems. However, the technology on its own may yield comparatively little benefit if it simply expands the information overload in the business without supporting the knowledge and expertise that is required to make sense of that information. The latter is a much more difficult challenge. Ultimately, managers must have the courage to address the factors that enable knowledge sharing – trust, networks, and dialogue – with the same intensity and discipline that is applied to the roll-out of systems. As in our World War I example, applying intelligence to these problems is much more effective than going over the top time after time with systems that can never succeed.
Inset: Ebank
In the mid-1990s, Ebank, a large European bank, decided major business customers needed an integrated service that would allow them to access the bank’s expertise around the world, regardless of location. This was done because one of the bank’s largest corporate customers had recently terminated their account, complaining the quality of service and advice they received from different international Ebank divisions was patchy and inconsistent. The vision was to unify the bank around global knowledge sharing-processes and communications. Like many companies in the same situation, Ebank started an intranet pilot project to turn the strategy into reality.
If enthusiasm were the only measure of success, the initiative would have been a hit. Senior management coordinated the intranet centrally and allocated ample funding. A workshop was held to introduce information systems (IS) staff from the different divisions around the world to the new intranet technology. Their response was ecstatic, with huge enthusiasm being expressed for the technical possibilities opened up by intranets.
What happened next, however, defied top management’s plans. As the IS teams returned to their respective divisions and countries, they applied themselves assiduously to the task of developing intranet technology. Their focus, however, was on meeting the needs of their own internal clients, and not the need to share knowledge across the bank. Many of these internal functions and groups were desperate to have their own intranet as a badge of identity; a way of waving the flag more than as a way of improving services.
The result was that, by 2001, Ebank management were confronted by more than 150 separate intranets (they weren’t even certain about the precise number). Each of these was unique to a particular unit, with no connection to divisions in the rest of the bank, and little inclination to share knowledge with them. So despite a massive IT investment, Ebank’s efforts had only reinforced the boundaries between different groups, not reduced them.
Inset: Buckman Labs
Buckman Laboratories competes in a variety of businesses, from pulp and paper processing and water treatment, which makes up 60 per cent of its sales, to leather, agriculture and personal care. In 1989, Bob Buckman succeeded his father into senior management and made a personal commitment that knowledge would become the foundation of his company's competitive edge. Three years later, the implementation of an online- knowledge network marked the realisation of Buckman Labs' vision.
A fairly typical example of how that system operates concerns the need for specialist knowledge on pitch control. Pitch control involves working on removing or minimising the effect of pitch in the paper making process. Pitch is made up of sticky materials left over in the pulp fibres used in the papermaking process or contributed from adhesives or plastics in recycled fibres. Given the range of activities Buckman is involved in, there are frequent demands for knowledge of new or esoteric domains. In this instance, knowledge of pitch control was required for a work programme in an Indonesian pulp mill. When Dennis Dalton, Buckman's managing director for Asia, was proposing this programme he sent out a message through the company's corporate intranet asking for help on how he could go about preparing it. "I would appreciate an update on successful recent pitch control strategies in your parts of the world," he wrote.
The first response came three hours later, from Phil Hoekstra in Memphis. It included a suggestion for the specific Buckman chemical to use and a reference to a master's thesis on the pitch control of tropical hardwoods, written by an Indonesian student attending North Carolina State University. The second response came 50 minutes after the first, from Michael Sund in Canada, offering his experience in solving the pitch problem in British Columbia. Then Nils Hallberg logged in with examples from Sweden; Wendy Biijiker offered details from a New Zealand paper mill; Jose Vallcorba cited examples from Spain and France; Chip Hill contributed scientific advice from the research and design department in Memphis; Javier Del Rosal included a detailed chemical formula and specific application directions from Mexico, and Lionel Hughes wrote about his experience in South Africa. In all, Dalton's request for suggestions generated 11 replies from six different countries, stimulated new discussions, generated new knowledge and catapulted him into a position to secure a $6m order from the Indonesian pulp mill.
