The percolator effect

Since companies first began experimenting with KM in the 1990s, the discipline has entered a more mature age. Using the experiences of Electricité de France as an example, Jean-François Ballay describes how knowledge management has evolved and reveals how the company’s ‘percolation’ strategy has helped to instil a culture conducive to knowledge sharing throughout the organisation as a whole.

In the knowledge-based economy, companies and institutions need to empower and leverage their knowledge by every means possible to create value and increase profits. It remains a difficult task, however, as knowledge is not dealt with explicitly by company accounts, financial reports, nor by management boards. For this reason, most managers do not incorporate knowledge into the reasoning by which they make strategic decisions, representing a major risk for people, firms and the environment in which they operate.

The question now is whether firms and institutions are going to be able to foster a greater understanding of the role of knowledge and spread that understanding across all aspects of organisational procedure. This is fundamentally a cultural problem. Can we expect to see knowledge management enter a more mature age, and what form will this take?

Taking the French utility Electricité de France (EDF) as a case study, we can see how this transformation is taking place. The knowledge-management culture in EDF is leveraged by a number of fairly original means: first, its KM community – a network of about 700 knowledge managers – is a powerful lever for sharing lessons learnt from KM projects that have taken place all over the firm; second, through KM seminars, expertise in knowledge management is distributed to the firm’s managers, helping them to embed a knowledge culture in every business process. This is a significant example of what a ‘percolation strategy’ can be. Using this approach, solitary islands of KM pioneers are able to expand, and the ocean gradually turns into a continent.

The early age of pioneers: solitary islands in the ocean

At the beginning of the 1990s, the first knowledge-management projects appeared in a number of diverse industrial organisations, much like a series of islands. Certain firms around the world took part in a movement that was to mark the first age of knowledge management.

What brought about this shift in attitudes, and what were the main features of this phenomenon, so widely propagated in the business press? Was it really new? A retrospective glance shows that, in many respects, firms have always worked to capture and share their knowledge through various structures, processes, products and documentation. However, for the most part, existing business procedures didn’t focus on knowledge but rather on more tangible notions to do with products and services. Knowledge was active as a process but on a more intangible, that is to say subconscious, level.

This is the context in which the first knowledge-management projects took place. At EDF for example, those in charge of the Electrical Equipment department noticed a number of problems liable to affect production and innovation processes. These problems were linked to intangible resources: information, knowledge and competence.

At the outset of the 1990s, computer technologies and standards formed a very different landscape to the one we see now, ten years on: the web, collaborative work tools, portals and other communication networks didn’t yet exist. Documents attached to everyday activities were typically transferred in print from the sender to only a limited number of people, and as soon as they became out of date they were stored in archives or centralised documentation systems that were often difficult to access.

And in terms of skills management, the situation was hardly any better. Human resources remained a centralised affair, far from the current ideal of on-hand know-how, because models used to characterise skill competence were usually based on very abstract and general models. As for on-going training, the process for acquiring new skills tended to be rather slow and functional, although undeniably useful. Yet these procedures proved insufficient for dealing with the human consequences of global economic pressures: the destructive re-organisation of professional skills, staff turn-over, experts leaving the company, new recruits taking too long to acquire core skills, weak collective learning and capacity for innovation, and so on.

With the Diadème knowledge-management project at EDF, the Electrical Equipment department set up a system whereby each engineer became a ‘knowledge contributor’, gradually enhancing the widely accessible business knowledge base over the course of time. This took the form of a portal which, to use a very topical term, aggregated collections of documents, internet or intranet sites and active users’ files permitting one to find out who knew what, in addition to all kinds of potentially useful information.

For example, an engineer is questioned by a customer about the phenomena of over-voltage in transformers when jolted during transmission or stuck by lightning. By browsing into the knowledge base, the engineer is immediately able to follow up leads and discover they can question colleague X, an expert on the subject, before they answer the customer’s query. They may also discover that specific physical phenomena can occur in similar conditions and are able to download any relevant studies within a few seconds.

For this type of knowledge base to be pertinent and effective, it is essential that knowledge workers are motivated enough to want to share information with others. They also have to learn to be selective (otherwise the knowledge base will fast become a collective dustbin). Finally, it’s essential to cultivate curiosity and open mindedness if the process of managing knowledge is to become a real business commodity. This is not simply a question of developing a collective archive but rather a source of knowledge that endeavours to be alive, continually evolving, transversal and open to the outside world. All these conditions are pre-requisites to success, and knowledge managers know how difficult they are to achieve. It is important to be patient, methodical and to secure the support of as many people as possible within the organisation and, of course, constant top-down backing.

In common with many other knowledge-management projects underway at around the same time, though, knowledge was implicitly regarded as an object. This is often the dominant perspective when people try to ‘extract’ knowledge from individual experts, and has led to the creation of many corporate knowledge bases that offer little real practical value. As such, the first age of knowledge management was dominated by a mechanistic, instrumentalist or, at best, behaviourist epistemology. People sought to capture knowledge in a database on the assumption that it would then be ready to use, much like a material product. This fundamental error was also found in numerous aspects of human communication. The behaviourist approach describes communication as a top-down process in which a message is distributed in a one-way flow, not allowing the receiver to react or to learn. We know this model is flawed, yet it nevertheless remains in many communication systems in use today.

The second age: lots of islands emerge

The communication model inspired by Palo Alto gives a better view of what a viable KM system should look like. In this model, the knowledge user is able to react and to learn, as indeed is the expert or teacher. This model relies on the value of feedback, and is based on interactions between people. It remains relevant for IT systems like intranets and more generally for knowledge-management systems.

Four main processes need to be considered: capture, transfer, renewal and a meta-process: socialisation.

Capture – a series of processes by which information and knowledge are identified, codified, indexed, classified, aggregated and discussed so as to enhance the organisation’s knowledge base. A key point is that the knowledge base is far more than just a database; rather it is the explicit form of the firm’s intellectual capital;
Transfer – refers to the processes by which information and knowledge are distributed, used, and then transposed and integrated through a learning process by the user. The user is not passive but, on the contrary, is able to incorporate, transform and apply the accessible knowledge, allowing information to be transformed into a value-creating process;
Renewal and creation – knowledge and information has to be adapted, transformed, destroyed and renewed all within a suitable time process. For example, business-intelligence and customer-relationship-management processes depend on this ability as clients and operational environments are constantly changing.
Socialisation – social interactions are fundamental in the process of knowledge creation and knowledge sharing. Discussion, groupware and communities of practice are examples of socialisation processes. They are immensely valuable in the validation, improvement, re-use and transformation of knowledge. For this reason, the socialisation process is in fact a meta-process for knowledge management.

This capture/transfer/renewal and socialisation (CTR-S) model (see figure 1) is an important target for second-age knowledge management. But achieving this goal is dependent on developing a systemic approach to KM, in addition to a more thorough understanding of organisations and human culture. In the CTR-S model, it is also easy to understand the two dimensions encompassed by the idea of the ‘knowledge worker’. The knowledge worker is both a knowledge contributor a knowledge user.

Figure 1 – the CTR-S model for knowledge management

Putting people first

The concept of intellectual capital has the advantage of encompassing the polymorphism of knowledge, being at the same time information, knowledge, expertise, skills and experience. It also fits well with the key challenge for knowledge management, which is to simultaneously further understanding in a number of distinct areas: information, organisation, management and HR management. Intellectual capital as a notion highlights the human dimension of knowledge and is helpful to remind us that human beings are key to the success of any enterprise.

The most visible and concrete aspect of any KM system, though, is typically the surrounding technology. But should knowledge management be considered a supplementary function to IT systems? Is knowledge management a specific IT system or is it something completely different? It is easy to get bogged down in this kind of debate, but it is useful to consider knowledge management from a multi-disciplinary point of view. Socio-psychology, anthropology, learning and teaching methods, philosophy and cognitive sciences all show that knowledge and memory are more than information systems.

Memory is not simply a store of information, waiting to be pieced together on demand like a video sequence. It is fundamentally a creative process using certain recorded material, re-constructed, re-organised and adapted to the requirements of each new context. “The past is unceasingly reconstructed by the present.”[1] In addition, it is important to realise that the impetus for action is not simply cognitive but also emotional. Numerous recent neurobiological studies have shown that knowledge is the outcome of various, multi-level interactions between impulses, emotions, feelings and consciousness.[2]

The continual interaction between knowing and doing has major effects on any knowledge-management project. Knowledge is not an object; rather it is the interaction between thinking and doing. Information alone does not necessarily indicate awareness of this interaction, which is why IT is no more than a tool for leveraging knowledge-management processes. The efficiency of information technology, as with all technical systems, can therefore not be gauged in isolation. What counts in a firm’s performance is the interaction between IT and the living organisation: human beings. If information technology is assessed independently from the cultural and organisational context, it will never achieve its full potential.

So what does a knowledge-management system have over an IT system? One way of looking at it is that it involves an approach where the sum of tools/organisation/human beings is considered and improved, rather than only one of these three dimensions.

This also explains why managing human resources is not enough to ensure efficient knowledge management. For example, a training policy will be more effective if it is linked to an employee’s everyday responsibilities. Today, many corporate university projects and e-learning initiatives are attempting to use intranet technologies to allow learning while doing – bringing training to the users in a specific work context. This approach is not automatically successful, as there is a risk that ‘tele-training’ becomes too mechanistic. Potentially, however, this approach could fundamentally transform the learning methods used by firms right across the globe.

The age of expansion

Intellectual capital works at numerous levels in a corporate context, each of which interacts with the other: people, processes, organisations, norms, technologies. The real problem for second-age knowledge management is to create a systemic approach, visible as interactions between different functions (HR, IT, management, communications) and core businesses (client relationship management, production, design, marketing and so on).

Observation of the management initiatives underway in an organisation is a good way of illustrating how knowledge management can spring up sporadically. For instance, the statements below were made by EDF to the business press, and in each the themes encompassed by knowledge management appear clearly, if implicitly:

“Human capital is key to the company’s future”:
- Improve personal commitment to jobs;
- Acquire the necessary new skills;
- Enhance the value of existing team competencies;
- Develop an innovative attitude.
“The intranet is a lever for change and business transformation”:
- Corporate portals become sources of knowledge leverage;
- Ensure the security of information and know-how;
- Protecting the company’s heritage should become part of company culture.
“Performance and quality are dependent on knowledge being efficiently managed”:
- Display, share and transfer results and lessons learnt;
- Improve team management.

KM at EDF and Gaz de France

Top-down objectives are not enough, however. A gap remains between simply announcing the firm’s strategy and the ability to make knowledge management a mature and professional process in everyday operations. This becomes the responsibility of information system project managers, HR management and managers in charge of individual teams and departments.

To illustrate the diversity of contexts in which KM is relevant, allow me to offer just a few examples of knowledge-management projects underway at EDF.

One of the International and European Division’s missions is to support the group’s subsidiaries, with the goal of contributing to their development and economic performance. It is not only a question of improving the profitability of each subsidiary but of setting up a truly global approach: sharing best practices, managing communities of practice, benchmarking, giving access to portals into common resources via an extranet and so on.

A second example is a learning organisation for managing the relationships between EDF and independent energy producers. Co-generation power plants and renewable energy technologies make it easy for anybody to produce electricity. Thus EDF has to deal with specific problems such as how to connect these independent power plants to the national power network. This in turn requires technical, legal and commercial expertise and information. A collaborative intranet system was designed in 1997 to help communities of practice to deposit and share their knowledge, access technical and legal documents, collaborate through dedicated forums, and so on. It is similar to the Diadème approach described above.

In most other cases in EDF, the approach for implementing KM relies on a so-called ‘knowledge compass’, the four principle directions of which are provided by the CTR-S model (capture, transfer, renew and socialise), as represented in figure 2. Each context is unique and every business needs its own specific strategy, but each characteristic can be mapped according to these four directions.

For instance, in an engineering division of Gaz de France, management realised that critical knowledge was not being transferred to the junior members of staff who had recently been employed. A specific programme was therefore put in place that allowed juniors to be sponsored. In another business, it became apparent that it was quite difficult to identify individual communities of practice. A project was subsequently launched to identify and leverage the communities that had emerged. Each of these projects can be mapped on knowledge compass, which is a very simple way to design a KM project and to make the project’s goals clear to everybody, which in turn is a good way of motivating people.

Figure 2 – the four principle dimensions for implementing KM

Knowledge culture: the percolator strategy

However, on a firm-wide scale only the innovators and knowledge leaders themselves are fully aware of the extent to which knowledge plays a role in their individual projects. For the majority, the paradigm remains hazy. Discourse and reality are frequently disconnected and new KM projects often remain localised and highly contextual; people’s experience with KM is not easily made readily available to others.

With this in mind, an additional source of leverage was set up at EDF and Gaz de France: the KM community. Sponsored by the CIO, the community was created during spring 2000, and has since attracted the participation of around 700 individual knowledge managers. All these people share their KM experiences with the rest of the community. Participants are thus able to discover instances of best practice and other valuable information about KM, which is in turn useful for leveraging their own KM business projects. And, of course, there are thousands of knowledge workers beyond the KM community members.

The main challenge facing the KM community is, as ever, a cultural one. It is a question of inventing pragmatic means of acquiring expertise in knowledge management where it is needed, which is in turn complementary to the hierarchical structure of the firm and its most important projects. The KM community has shown itself to be a tremendous source of leverage in this respect.

And in addition to the KM community, managerial staff at EDF and Gaz de France are also offered additional services, such as participation in seminars on key KM themes. Again, such initiatives have proved extremely helpful in supporting new KM projects right across the business, allowing specific experiences with KM to be encapsulated in distinct methods and concepts, and passed on to those who need them. Thanks to this type of initiative, the KM culture within the company is growing rapidly.

A useful model that shows just how this is happening is the percolation phenomenon. Consider, for example, a seascape scattered with innumerable small islands. In normal climatic conditions these islands would not be connected. If the sea level were to fall, however, small groups of islands would gradually start to link. Globally the sea would continue to dominate until a certain point when the majority of the islands find themselves suddenly connected. Beyond this point, called the percolation threshold, the islands become a continent.

Up until now, knowledge management in many firms has remained below the percolation threshold. KM innovations have been effective in certain areas but what is lacking is an overall culture, the capacity to make KM concepts and methods one’s own where necessary, and to implement them in ways appropriate to local contexts.

A complete transformation will only come about through cultural change at every level of the organisational hierarchy. A top-down, uniform programme of change is not necessarily the answer, but rather a multi-faceted, day-to-day approach that takes on board different contexts. The ultimate aim for a knowledge manager, as mentioned above, is to be a catalyst for the percolation of knowledge that would constitute the second age of knowledge management. Already in EDF, and in addition to the 700-plus knowledge managers, thousands of knowledge workers are benefiting from improved knowledge agility, the benefits of which to the company as a whole are immense.

References

1. Piaget, J., Problèmes de Psychologie Génétique (Problems of Genetic Psychology) (Denoël/Gontier, Médiations library, 1972)
2. See for example Damasio, A., Le Sentiment Même de Soi (Conscience and Emotions) (Odile Jacob, 1999)

Jean-François Ballay is knowledge manager and consultant at EDF and Gaz de France. He can be contacted at jean-francois.ballay@edfgdf.fr