• Business Intelligence
• Collaboration
• Competitive Intelligence
• Communities of Practice
• CRM
• Culture
• E-learning
• Enterprise Content Management
• Enterprise Search
• Intranets & portals
• Taxonomies

• News
• Events

Feature

posted 1 Apr 2000 in Volume 3 Issue 7

Organic knowledge management, part one - The ASHEN Model: An enabler of action

In a three-part series of articles originally published in this magazine in 1998, David Snowden laid the foundations for an approach to understanding the intellectual assets of an organisation using techniques derived from anthropology and based on the principle that 'we only know what we know when we need to know it'. It was a methodology that eventually became known as Organic Knowledge Management. This article marks the first in a three part series that updates and augments that material with the benefit of two years of additional research and practice. The first article in this series looks at the language of knowledge and suggests a model of description that leads to constructive action. The second article will provide a practical set of guidelines to enable the identification of knowledge, updating and augmenting the 1998 material. The final article will complete the picture, and consider the critical importance of heuristics to managing in the face of uncertainty. 

'A little knowledge that acts is worth more than much knowledge that is idle'
Kahil Gibran, The Prophet

The way we choose (and it is always a choice) to describe something determines to a large extent how we act in consequence. At a trivial level the pessimist whose glass is half empty will hoard what is left against the possibility of future shortage; the optimist whose glass is half full proceeds with greater confidence. More seriously in the 17th Century the same act or belief could be reformation or heresy, and in the modern era Derry or Londonderry spoken in innocence can identify an individual as belonging to a particular socio-cultural background with assumed beliefs and attitudes. This particular use of language to mandate response and action by its nature is equally present in all organisations.

Have we outgrown the 'tacit' and 'explicit' definitions?

The more intangible an asset appears, the more important the language with which we describe the problem. In knowledge management the words tacit and explicit dominate most conversation. Although the use of tacit is normally attributed to Polanyi's 1962 Terry Lectures at Yale University (Polanyi, 1983), its de facto use is to a large extent determined by two authors: Nonaka globally and Probst in central Europe. The common reading of both these authors, whatever their intent, too easily leads to implicit assumptions about the way in which knowledge should be managed that are inappropriate and in some cases down right dangerous.

The STET model (Nonaka and Takeuchi, 1995) has four transitions between the tacit and explicit states: Socialisation, externalisation, internalisation and combination. The examples are drawn from the manufacturing industry in which all four transitions are necessary to move from research to production. The model in many ways gave rise to the current levels of interest in knowledge management and provided many vendors of both hardware and software with a classification matrix for new (and too often old) tools. The difficulty with the model in use is four fold:

• It is often used in complete isolation from its supporting material, including highly valuable but underused concepts such as Middle-Up-Down, to define the totality of knowledge actions in companies. While the model is useful, it is not universally applicable in manufacturing let alone the service sector.
  
• It has an implicit assumption that knowledge is some form of thing or entity that retains a coherent identity through the four transformations. Knowledge is seen as an asset that can be created and managed; replacing products and raw materials as the primary focus of strategic thinking. One of many problems with this is that the transformation of knowledge between the tacit and explicit states fundamentally changes its nature.
  
• It leads to confusion between the container and the thing contained. Tacit knowledge exists in the heads of individuals or communities, explicit knowledge in documents and other artefacts. In practice, most useful knowledge has both tacit and explicit aspects and needs to be managed holistically. The STET model leads the average manager to manage the containers.
  
• There is an implicit assumption that tacit knowledge can and should be made explicit. The two are separate questions (Snowden, 1999), the fact that we can does not mean that we should, and more often we cannot without loosing something essential. Although many people use the Polanyi quote 'we can know more than we can tell' (op cit, p. 4) few read on to understand that we can always know more than we can tell, and we can always tell more than we can write. They also fail to recognise that the initial emphasis on the word 'can'.

Probst and his co-authors (Probst, Raub & Romhardt, 1998) offer a more seductive and simpler view of knowledge. Tacit gets little mention with the focus on a useful set of tools and techniques for managing knowledge, which can or should be codified. Knowledge is separated into two classes - that which can be codified and that which cannot, which is held to be genius and beyond the bounds of structured management. His book, recently translated into English, dominates thinking on knowledge management in Central Europe. It is attractive because it uses the principles and practice with which most western educated managers are familiar and comfortable. It reinforces the de facto prejudice that anything useful should be written down or embedded in a process. The mechanical metaphor of Business Process Re-engineering, Quality Management and the like predominates.

Both of these models play to the contradictory dualism that is the day-to-day practice of most managers. In calm and rational moments, they want things written down: Project reports, competitive intelligence and so on. They plan and build systems in which this 'Intellectual Capital' is widely distributed, and considerable investment is made in its maintenance, all of which is justified on an idealised goal of dynamic decision making supported by the availability any time any where of all and any knowledge that they or their subordinates might need. These overblown expectations are both created and reinforced by unscrupulous vendors of knowledge management systems and consultancy. In contrast, when the chips are down Dr Jeckyl becomes Mr Hyde and, either through direct management or a 'go fix it' instruction to a trusted subordinate, moves the problem from structured, explicit and pseudo-rational decision making to tacit empowerment based on trust and experience. Under these circumstances, simple rules and values predominate; it is always interesting to see the rationally constructed use of balanced score cards and the like being abandoned under pressure to focus on sales, profit and cash.

Dualism is inevitable if we see the organisation as a machine interacting with human agents. The very language of tacit and explicit militates against a more holistic view of the interaction between human beings and their artifacts. If we separate human decision making from the support of artifacts 'the touchy-feely, intuitive manager' then we rely on genius, or more often luck, at the expense of scalability; if we deify the artifacts building unrealistic expectations about the use of technology then we gain scalability and loose massively in our ability to respond quickly to uncertainty. A holistic approach requires us to describe things in terms that naturally lead to holistic thinking, while providing some form of categorisation that leads to effective action. Human beings need to categorise things in order to exercise their sense-making capability within organisations, but the categorisation should lead immediately to effective action. A hunter in the field needs to categorise prey from predator instantly to trigger a kill or flight response; no lesser degree of responsiveness is required in the modern organisation.

Knowledge is contextual; how and when the question is asked are vital

In A Hitchhikers Guide to the Galaxy, Douglas Adams writes of a society that constructs a computer to answer the ultimate answer to the ultimate question of Life the Universe and Everything. After centuries of thought, it 'in some trepidation' produces the answer: 42. At this point, an even more massive computer has to be constructed to identify what the question was in the first place! Interestingly, the new computer is organic, involving the planet Earth and white mice, who provoke the subjects of their study 'humans' by occasionally running the wrong way around a maze.

Knowledge is only known when it is needed to be known (Snowden, 1998a). It is triggered by events and by need. Normal consultancy methods in which a structured interview is created, possibly with a questionnaire, are premature in the early stages of knowledge discovery. Asking people what they know is a cruel question. A group of managers in a workshop asked to write down what they know will scribble industriously for ten minutes or so and will then start to look puzzled before they reach a halt. The reality is that if they wrote down everything they know, they would be there for the rest of their lives. A database can be listed; a human mind has to be stimulated. One of the most common phrases in all languages is 'I'll sleep on it'; in solving a problem an individual will stimulate themselves through conversation and reading, and then assimilate the results into something coherent. Knowledge Disclosure Points (KDPs) (Snowden, 1998a) comprise decisions, judgments, problem resolution and learning. They are the points at which we use knowledge. Any individual will find it easier to recollect the use of knowledge, even if they cannot meaningfully answer the question: 'What do you know?' Locating, categorising and summarising the KDPs in the community are the means by which we provide context: 'When you made that decision, what knowledge did you use?' is an adequate question in context and is more likely to reveal meaningful results. How we identify those KDPs will be the subject of the second article in this series.

However, while asking people what they know in the context of KDPs has proved more successful than conventional approaches, it still suffered in that knowledge so identified suffers from the problems of dualism, as mentioned above. There seems be an insatiable drive to codify. Two years of arguing with varying degrees of success to retain some knowledge in its tacit state gave rise in the field under fire to the ASHEN model, the components of which are described in the next section.

The ASHEN Model

The ASHEN model was created as a means of providing a linguistic framework both to help organisations identify what they know and to move directly to action as a result of the meaning provided by the language. It is designed to prevent the need for argument about the management of its outcome. The mnemonic form (HANSE in German) facilitates consistent use in the field. The five ASHEN components are:

Artefacts

art'efact, art-, n. A product of human art and workmanship; (archaeol.) a product of prehistoric art as dist. from a similar object naturally produced. [f. L arte (abl. of ars) + factum (neut. p.p. of facere make)]

The term 'artefacts' encompasses all the existing explicit knowledge and/or codified information within an organisation. The processes, documents, filing cabinets, databases and other constructed 'things' that encompass the codifiable to varying degrees of success. The management issue here is the removal of duplication and the general optimisation and ready distribution of such artefacts to communities that need them. The artefacts will always need to be in the right place at the right time - even though most people may be unaware of their existence for most of the time - this is a non-trivial management challenge for which technology can only support, but not provide, answers. Many artefacts exist but are not recognised. They may be notebooks of past exceptions events in the drawer of a staff room of a supermarket; a diary in a café frequented on a regular basis by field engineers or a web site using the free space in Hotmail used by individuals in competitive companies who shared a common interest. All three of these examples come from the author's own experience, and in each case were probably one of the most valuable assets identified in a knowledge disclosure exercise.

Their value is in their natural occurrence; they developed based on the real needs of individuals. Attempting to change their nature would be dangerous. To take the example of the field engineers: The book in question was used daily to communicate valuable information about health and safety procedures, 'work a-rounds' on technical parts, gossip about managers, information about customers. The mechanism of its maintenance was that each engineer would casually read it over a cup of tea and then write their own observations before leaving. One of the solutions proposed when it was discovered was to enhance existing hand held computers to capture the same information in the field. This missed the point; the artefact was a part of a social setting and involved social obligations. The solution was to endorse the use of the café in return for managers being allowed to photocopy the book on a weekly basis; and to sell the idea to the engineers by telling them to keep two books.

The key is to respect naturally occurring artefacts and to separate the creation and capture of knowledge from its analysis and distribution. It may not be neat and tidy to do so and may appear to be anti-rational and sub-optimal; nevertheless, it works.

Skills

skill, n. Expertness, practised ability, facility in doing something, dexterity, tact. [ME, f. ON skil distinction, cf. SKILLS]

In this context skills are those things for which we can identify tangible measures of their successful acquisition. If I employ a plasterer then I can measure the deviation from a vertical plane of his work, and the time taken to complete the job. Customer relationship is a different thing to measure, and although it has aspects of 'skill' , the term is not enough in its own right. The time element is an important aspect of the skill measurement. The author is a reasonably accomplished carpenter, but a skilled chippie can accomplish in one hour a task that is a weekend's work for the amateur.

Skills are something that organisations know how to manage. They are an organisation's most readily codifiable knowledge assets. Training needs and skills analysis are well known techniques. Training courses, moderated work experience - the gambit of techniques available is wide and well proven. However, there is always the danger of the codification heresy; the belief that once something is written down, then it is shared. Most of the published 'success' stories of Intellectual Capital Management often suffer from this heresy. To illustrate this let us return to the plasterer example. Anyone who has tried to plaster a wall based on the codified knowledge of a book - say, The Ten Easy Steps to Perfect Plastering - will know the issue. Following the instructions does not mean that the plaster will stay on the wall, or that you will not have to burn out several sanding machines to achieve any smoothness. Too many organisations in building their Intellectual Capital Management systems are actually creating legions of amateur plasterers. While skills can be codified, time has to be taken to internalise them. The management task is to catalogue the skills, understand the time horizon and resource requirements for their acquisition and plan accordingly.

Heuristics

heuris'tic (hur-), a. & n. serving to discover; ~method, system of education under which a pupil is trained to out things for himself, so ~s n. pl. [irreg. F. Gk heurisko find, see -IC]

Heuristics, or rules of thumb, are one of the most valuable of assets and may be articulated without the need to render them fully explicit. They are the effective way by which we make decisions when the full facts are not known - or are not knowable in the time available. A good example is the CEO looking at a range of investment proposals without sufficient time - or the inclination - to go through the detailed case. The decision criteria often take the form of a simple rule set: Has someone I trust checked this out? Will it impact on my targets for this year? Will it distract key staff from other more important targets? These may or may not be articulated, but they are often known to the CEO s inner circle. They are also the means by which experts and/or professionals make decisions in conditions of uncertainty. An example would be: 'If the gauge goes above that level, in these circumstances, then I'll look at the problem again.' The essence of heuristics is that they have fuzzy edges and therein lies their power. They allow greater consistency in conditions of uncertainty but follow the pareto principle that 80% is good enough. Over time they may become fully explicit and become artefacts, or they may remain tacit - only available to an expert community. Recent work with a group of engineers revealed some interesting heuristics, some of which could be codified and distributed - but the general comments about their use were summed up by one engineer who said: 'It's a good rule and I use it all the time - but I wouldn't let anyone with less than ten years experience anywhere near it. Until then they can do it by the book!'

For management, identifying and codifying heuristics is a fast track and generally cheap way to spread valuable knowledge quickly. The act of making the heuristics explicit can also clear away false assumptions and out of date working practices, where the context in which the original and mostly un-stated heuristics were developed no longer appertains. The third in this series of articles will look in more detail at the use of heuristics in knowledge management, drawing on ideas from complexity theory.

Experience

exper'ience1, n. Actual observation of or practical acquaintance with facts or events; knowledge resulting from this [ME, f. OF experience f. L experiential f. EX1periri pert- try : see -ENCE exper'ience2, v.t. Meet with, feel, undergo, pleasure, trement, fate etc.); learn, find; (that, how, etc. [f. prec

Experience is the most valuable and also the most difficult of the tacit assets of an organisation. It is difficult for two reasons: