Have you noticed that the term "knowledge management" is making ever more frequent appearances in conference agendas, on book lists, in job titles and even as a university discipline. I guess it's no wonder when all the leading thinkers and analysts are telling us that competing on the basis of intellectual effort is critical to sustained business performance in today's new knowledge economy.

Value in business is created more by intangible assets – what and who you know – than by the physical assets, plant and equipment you own and deploy to produce goods and services.

Consequently, it is now making sense for business leaders to examine the knowledge underlying their businesses and how that knowledge is used for enhanced business results. This is enormously aided by the sophistication and ease that modern information technology brings to the codifying, storage and dissemination of knowledge.

But managing knowledge remains a tricky task. It is a bit like herding cats.

We are confronted by dilemmas and divergent approaches to knowledge management – technology vs people solutions; codification or personalisation; how to capture knowledge without destroying creativity.

I have a keen personal interest in solving these knowledge management dilemmas. My own organisation, the Australian Business Foundation, is a research think-tank whose core purpose is the generation and communication of break-through ideas that boost Australia's competitiveness and prosperity. In other words, a knowledge-based organisation.

Therefore, several excellent analyses of knowledge management reported in the Harvard Business Review have caught my attention.

Firstly, in the May/June 2000 edition, John Seely Brown and Paul Duguid contrasted process re-engineering with knowledge management as radically different approaches for managing in the new economy.

Re-engineering focuses on top-down restructuring of business processes, people and information to gain sustainable competitive advantage. It assumes that value creation can be described and documented and readily repeated in a relatively predictable competitive environment.

Knowledge management, on the other hand, is a more bottom-up, practice-based approach. It assumes managers can best foster knowledge to enhance the effectiveness of their enterprise by responding to the inventive, improvisational ways people actually get things done. It recognises that value-creating activities are not always easy to pin down, and that the business environment is constantly changing and essentially unpredictable.

Seely Brown and Duguid see the challenge as a balancing act to capture knowledge without killing it. Unharnessed, ill-disciplined ideas are unlikely to turn a profit, but too much structure constrains the freedom of thought that drives the continuous innovation on which businesses in the new economy depend.

They cite the experience of Xerox and the difficulty of knowing what you know about your company's best practice. Customer service reps who fix Xerox machines are supposed to go by the book – the manual or expert system in the machine that identifies error codes and consequent instructions for repair. This is a fact, but Seely Brown and Duguid identify that their real success comes from how the reps learned from one another by sharing stories about how they had fixed the machines and adapted to the idiosyncrasies of individual units in different circumstances. This sharing of tacit knowledge occurred at informal breakfast gatherings before work over coffee, questions, laughter at mistakes and gossip.

Hansen, Nohria and Tierney in the March/April 1999 Harvard Business Review draw distinctions between a codification and a personalisation strategy for knowledge management. Codification involves extracting knowledge from people, codifying and documenting it and storing it in databases, where it can be accessed and used easily by anyone in the company. Large consulting firms like Ernst & Young and Accenture are examples, achieving scale in knowledge re-use and allowing cost savings and volume business without diminishing quality.

Personalisation strategies closely tie knowledge to the person who developed it; sharing is mainly through direct person to person contacts and IT systems are used more to communicate knowledge, not store it. Strategy consulting firms like McKinseys or the Boston Consulting Group, which focus on highly customised services to clients at a premium price, are examples of this personalisation strategy.

These authors make the point that an enterprise's knowledge management strategy should reflect its competitive strategy for creating value for customers and returns for itself. In particular, its choice depends on its answers to the following questions:

• do you offer standardised or customised products?
• do you have a mature or an innovative product?
• do your people rely on explicit or tacit knowledge to solve problems?

Codification approaches make more sense when operating with mature, standardised products and where explicit knowledge, like software codes or market data, are used to solve business problems. Personalisation approaches are more appropriate in enterprises with highly customised service offerings or innovative and novel products and where tacit knowledge (like business judgement, operational know-how or scientific expertise) is more crucial to business outcomes.

Finally, another angle is introduced by Hansen and Oetinger in the March 2001 Harvard Business Review, coining terms like "t-shaped managers" and "human portals".

They define the t-shaped manager as one who freely shares ideas and expertise horizontally across the company, while simultaneously remaining fiercely committed to vertical business unit performance. Using an in-depth study of BP Amoco, Hansen & Oetinger explore how best to promote and discipline formal peer group assistance and knowledge-sharing arrangements so they demonstrably contribute to bottom line business results.

Knowledge sharing in BP Amoco is institutionalised in job descriptions, promotion criteria, key performance measures and work organisation. As a result, Hansen & Oetinger clearly come down in favour of knowledge management strategies which are about brainstorming between people, rather than electronically moving documents around, provided that these efforts are aligned to specific business performance targets.

Undoubtedly, we will hear a lot more about knowledge management in all its guises. Let's not forget that the end game is how we become genuine learning organisations and thrive by competing on the basis of cleverness and know-how.