The knowledge management system

Feature

posted 1 May 2000 in Volume 3 Issue 8

The knowledge management system

This month's Your Say section focuses on the KMS; what are the factors that make one knowledge management system more effective than another?

Accounting for knowledge

When you say the word 'system' to a group of people, a good proportion of them will interpret the word as meaning an IT system. But a knowledge management system is far more than an IT solution. It is about a way of working; a way of working systematically, to deliver the value tied up in the knowledge of your staff, your teams, your suppliers and your customers. It's also a way of being, it's the core of your belief system.

Let's take an analogy for working systematically to deliver value, by looking at the management of one of your other assets. Lets look at money. Now I know money is not a strong analogy for knowledge. It is a lot more tangible for a start, and a lot more measurable, and everyone relates to it. But let's ignore for a moment the differences between the two, and instead compare the management systems that businesses apply. One reason that I pick money for the analogy is that you don't mess about with it. My HR colleagues tell me money doesn't motivate, but people still don't take chances with it.

The systematic management of money requires a combination of people, processes, technology and culture. You will need people such as accountants, financial planners, commercial managers. You have a whole range of business processes like budgeting, financial tracking, auditing of accounts. No doubt you have software technology you use; from base-level spreadsheets to more elaborate enterprise-wide accounting solutions. You may be moving into e-commerce, and investing in technology to move money electronically. Finally you have a money-savvy culture. Your people understand that 'money counts', they 'watch the bottom line', they look for 'bang for the buck'. They know the value of a dollar, and if times are tight, they watch what they spend. Your money management system works, because it is a holistic integration of people, process, technology and culture. These things have been about for so long we sometimes don't recognise them as being components in the money system.

What about your knowledge management system? Do you have the people? Do you have knowledge accountants, knowledge planners, knowledge managers? Do you have processes for knowledge tracking, or knowledge auditing? Do you budget for knowledge at the start of a project, in the same way you budget for money? Do you have the technologies you need at a local and enterprise-wide scale? What about your culture? Do people keep as close an eye on knowledge as they do on money? They may know what they earn, but do they know what they learn?

Analogies can be pushed too far, but I think in this case a comparison of the two shows that a knowledge management system, if it is to succeed, needs to look at the holistic integration of people, process, technology and culture issues. Part of the success of the BP Amoco programme (a detailed case study appears on page 7 of this magazine) was the close attention we paid to this integration. We created knowledge roles, we introduced knowledge processes, and we brought in knowledge technology. The result has been a deep and lasting change in the culture, and the delivery of real tangible business benefit. It didn't happen by accident, but as a result of creating a system to manage our knowledge in the same systematic fashion that we manage our money.

Nick Milton is the managing director of Knowledge Transform(SM) International. He can be contacted at:nick_milton@ktransform.com

The knowledge audit

To all who are searching for the knowledge management system that will increase the bottom line, make productivity skyrocket, and clearly demonstrate the value of knowledge sharing to the organisation, I offer this advise: Don't focus on the end result. Set your sights higher, to the strategic planning level. Companies that have achieved the greatest success in their efforts to create knowledge enterprises are those that began with an inward focus. A good knowledge management system is not simply a matter of the right technology. Those who are successful today began with an assessment of the knowledge environment of their organisation. That is the function of the knowledge audit.

The knowledge audit provides a 5,000ft view across the entire company and enables the astute observer to home in on any anomalies for further analysis. Groups will stand out because they exhibit positive or negative variances in the factors you are measuring. Good audit data shows how staff accesses, uses, shares and stores the information and knowledge needed to do their jobs. It uncovers organisational and cultural inducements and barriers to knowledge use and transfer, and highlights the best practices and lessons learned that often becomes useful across the business. Most importantly, it provides the basis for system design and the data for establishing ROI targets.

While in and of itself, the knowledge audit won't immediately translate to the bottom line, it does guarantee that your organisation is moving in the right direction, so its knowledge efforts are expended wisely and in a targeted manner. This can be seen in the turnaround in the R&D department of a major petrochemical company I recently worked with. This department had embarked on several knowledge-based initiatives, each of which had full upper management support. Despite this, the company's reuse of acquired know-how and expertise had not been impacted. Our knowledge audit uncovered several obstacles that the efforts to date had ignored: Cultural differences across various geographic locations, discrepancies in 'management speak' versus 'management action', and process realities completely contrary to established knowledge sharing practices. We also uncovered a cultural approach to team building, which had been virtually untapped by the previous knowledge initiatives. Once we clearly saw these strengths and weaknesses we were able to develop an action plan that exploited them. Amazingly, minor modifications to existing systems brought about major changes and a floundering knowledge initiative was turned into the desired killer application.

The most effective knowledge audits are not narrowly focused to a specific department or group, but administered throughout the organisation. Once you involve the entire company, you will find points of view which are frequently overlooked, but which become critical success factors in managing and sharing knowledge. As part of the analysis, it is important to assess the internal factors that potentially inhibit or promote knowledge sharing in the organisation. Thus the causal effect between these critical elements and the reality of how the business is functioning from a knowledge-sharing standpoint can be empirically tested.

Assessing the knowledge potential

An organisation that has little in the way of formal KM technology or practices, may still demonstrate an ideal environment for leveraging KM practices and technologies. Similarly, an organisation that has KM technology and practices, may nevertheless demonstrate an organisational environment that undermines its KM efforts. Neither situation is ideal. Understanding where and how to overcome the inadequacies of each is the purpose of a knowledge audit.

The latest research from The Delphi Group, shows that within the next year, knowledge management applications will be in use in every area of the organisation. Without insight provided by a knowledge audit, you risk having opportunities go unnoticed and thus under-utilised, and real obstacles go undetected. A word to the wise: Forget about finding the 'killer' application. No application can reach critical mass without first being strategically positioned.

Stacie Capshaw is a senior consultant at The Delphi Group. She can be contacted at:sc@delphigroup.com

Three steps to success

Success stories abound and often contain a plethora of 'micro level' issues such as training, education, support, and so on, but at the macro level there are generally far fewer components. The following three seem to be ëcommoní to most (reliable) success stories and yet paradoxically missing in many modern, so-called solutions. I submit them to you:

It's a 'personal thing'

A successful KMS always contains the ability to enable one-to-one relationships, even where users are numbered in hundreds of thousands, even millions. In other words, it's a 'personal thing' regardless of the user's interests, location or role. Let me be clear about my definition of 'personalisation' and 'relationships': The retrieval of relevant and related information and data that is best suited to the user's needs, and presented in way that adds value to the user's relationship with the information, which in many cases involves other users. The most effective solutions offer a choice between covertly or overtly measuring and understanding the user's needs, a choice that must be controlled and understood by the user.

If this objective is realisable, and I believe that it is, then traditional methods of summarising data become much less important than many of the traditional KMS applications claim. Unstructured and structured data is of potentially equal importance and can be analysed simultaneously.

Timing

A second facet of a good KMS process seems always to be able to deliver data and information to the user proactively, at the right time, and 'in the face'. That is, without the user being asked to engage with a cumbersome front end or specialist set of tools. How many traditional KMS claim this feature? Not many, in my experience.

Right information

Thirdly, personalised and correctly timed information must be the right information, defined as: Relevant information or data, obtainable in source or text formats, and in context with related information where required. Where the information or data is pure ëpeople knowledge, (all right, I'll use the terminology 'tacit') and unrecorded in textual form, then it must contain sufficient contact personal details.

As I said in the introduction, there are many 'micro issues' involved with a successful KMS, but the three listed above appear in all the winning combinations I have reviewed or used. Knowledge being what it is, I'm sure some of you will disagree!

Ian Black is the head of communications and public affairs at BAe. He can be contacted at:ian@ianblack.org.uk

The future: Knowledge fractals

Most knowledge management systems are concerned with facilitating the effective storage and retrieval of corporate knowledge to give everyone within an organisation immediate access to relevant information, no matter where it resides. These systems perform an invaluable function in today's knowledge-rich environment, and in many cases have made significant contributions towards elevating organisational performance. But these are early days, and many significant improvements to knowledge management systems are still on the horizon, one of which is the development of knowledge compression techniques.

We are experiencing an exponential growth in the volume and therefore the flow of information - broadband networks are now being installed world-wide, and they promise to dramatically alter the way in which we communicate and do business. However, the rate at which this growth continues to increase is constrained by several factors, one of which is bandwidth. With the currently available technologies for storing and processing digital information, knowledge management systems are able to instantaneously retrieve knowledge from vast repositories. Again, we seem only to be bounded by the rate at which electronic storage and processing technologies continue to improve. Meanwhile, we will carry on building our 'mountains of knowledge' to ensure that the 'right knowledge can be made available to the right person at the right time'.

One way of tackling the bandwidth constraint is data compression. Instead of trying to transmit the same amount of data more quickly, the volume of data sent is very much reduced by structuring it in a more efficient way. For example, when we receive large email attachments, they are often 'zipped', which means that the original data has been compressed, often by a factor of more than ten. In a similar way, it is suggested that leading knowledge management systems should place a greater emphasis on knowledge compression techniques, not only to improve storage and retrieval efficiencies but more importantly to encourage a more profound understanding of the knowledge contained within the system. In other words, to combat the constant danger of knowledge overload, we should strive to do more with less.

Within the field of data compression, there is a relatively new method called fractal compression. In this approach, repetitive patterns of an image or a graphic file are identified and matched. Two patterns are considered the same, as long as one can be stretched, squeezed, or rotated to match the other. This means that two very different images may be created from one underlying set of patterns. And, because patterns rather than pixels are viewed as the fundamental blocks of image construction, the fractal algorithm can fill in missing detail during decompression by mathematically scaling the pattern instead of just inserting extra pixels. This results in a higher degree of accuracy for decompressed images, which may be very important in multimedia, engineering and design applications.

By analogy, leading knowledge management systems will be those which provide a capability for developing the underlying frameworks or algorithms for 'knowledge fractals', a term used to define a collection of different bodies of knowledge, all of which have broadly similar underlying patterns. The result is that, in addition to a much reduced volume of static knowledge, knowledge management systems will contain a number of frameworks, which would dynamically 'create' or 'recreate' knowledge on demand.

To achieve this noble aim, the main challenge lies in developing the frameworks that underpin the knowledge fractals. In comparison to data compression, knowledge compression requires much more intelligent and elaborate mechanisms for interpretation, a task that will undoubtedly require human as well as machine intelligence. This final point further highlights one of the fundamental truths with knowledge management systems - it is the way in which people use knowledge management systems that determines their success, and not the systems themselves.

John Cummins works for AtomicTangerine, a venture consulting company. He can be reached at:jcummins@atomictangerine.com

If you would like to write a Your Say article, please email the Editor at:slelic@ark-group.com