levers In previous generations, knowledge, ‘know-how’, was taught by the foreman, the professional, or the artisan, to the apprentice or articling student, on the job. Most people learned by following the example of the master or expert. In those days, performance evaluation was easy, because the boss always knew more than his or her charges.

Today that has all changed. We live in such specialized times, with such high staff turnover, that most employee ‘know-how’ has been accumulated from a variety of sources in different organizations, and the boss rarely knows as much about his or her employees’ specialized roles and functions as the employees themselves do. In today’s world, the most common and effective way for an organization to ‘create knowledge’ is to hire it.

The modern organization therefore uses three new, looser methods to standardize how things get done, shown on the chart above:

  1. Tools & Technologies: Forms and software can compel employees to do certain things in a prescribed order. To the extent ‘intelligence’ is embedded in these tools and technologies, some modest, mostly-generic learning can also be achieved from their use. Of course, some technologies automate processes completely, obviating the need for employees. Some robots can even ‘learn’ from their mistakes.
  2. Methodologies and Processes: Although most organizations have scrapped SOP manuals, since there are so few ‘standard procedures’ anymore, many organizations teach new recruits standard methodologies. As employees advance to more senior positions and move from company to company, these methodologies get diffused and tweaked, and simply become, for each individual, ‘the way I do this’.
  3. Culture Change Programs: Communications, training, measurement systems, reward & recognition programs and ‘internal marketing’ are the ‘soft persuasion’ techniques that organizations use to try to teach people how to do things, and get them to do them in a particular way. In today’s workplace most in-house training programs are soft-skills training. Except during periods when rewards are especially high, these programs are usually ineffective at getting compliance, and their effectiveness at transferring knowledge depends utterly on the need and motivation of the employee learner, which is generally not very high.

So what you have is organizations where everyone is doing different things, differently. It is therefore not surprising that internal knowledge transfer in organizations is abysmally low, and most ‘knowledge management systems’ designed to facilitate such transfers are spectacular failures. Even ‘best practices’ databases, once sold by consultants for thousands of dollars, have fallen from favour, as companies realize that business practices are now so specialized and individualized that someone else’s ‘best practice’, which evolved in the unique context of its developer’s work history and style, is unlikely to be recognized as valuable, let alone usefully adopted, by another company, division or employee.

And how about benchmarks? Many organizations say they want performance and process benchmarks to see whether they are doing things the best possible way, and the objective is a good one. But since every part of every business, indeed every individual’s job, is increasingly unique, benchmarks have largely become meaningless even as they have become harder to calculate. In fact trying to get department X of company B to achieve a benchmark attained by the similar department of company A is likely to lead to dysfunctional behaviour, to cause more harm than good.

So how does management function at all in such chaos? Increasingly, the answer is by defining roles clearly, setting goals and objectives, and offering help if wanted on how to achieve these set goals and objectives, and then getting out of the way. Pragmatic managers realize that things happen in most organizations for good reasons that are probably not obvious to them, and if a goal isn’t achieved it probably wasn’t achievable. One CEO I know calls it ‘management by loose frameworks and constant communication’. If the shareholders aren’t happy with performance, the manager may try to use one of the three techniques above to bring about behaviour change: New technologies to automate something being done poorly or inefficiently (or the simpler “if form XJ1 has not been properly completed after each job, the employee’s pay cannot be computed”); new methodologies to increase adherence to a preferred process; remedial training etc. But for the most part, the answer in business when things go awry is the same as it is in sports: change the players or fire the manager.

What does this mean for the struggling, once-hyped discipline of ‘knowledge management’ ? Here’s a 7-point strategy for knowledge managers to cope with what Peter Drucker calls the greatest challenge of the 21st century: Improving the productivity of ‘knowledge workers’ (i.e. everybody):

  1. Focus knowledge and learning systems on ‘know-who’, not ‘know-how’. People get appropriate context, get what they need more easily, and learn faster, by talking to experts rather than looking for knowledge in databases or course curricula. If the front-end of both your intranet and your learning programs is a well-organized list of the company’s experts, and their contact information, your people will be well served.
  2. Introduce new social network enablement software and weblogs to capture the ‘know-who’. Corporate directories are often obsolete and incomplete. Job titles and job descriptions rarely tell the knowledge-seeker what their person really does. Organization charts rarely reflect either de facto expertise or de facto networks. The key is to get every expert in the company to capture as much of what they do and know in a personal weblog or similar repository. If it’s personal, it’s likely to be current, accurate, and complete, and to convey the context of the user’s expertise — centralized repositories rarely offer any of these advantages. Then use some of the leading-edge social software now being developed to analyze the weblogs and construct expertise finders and other tools to effectively connect knowledge-seekers to experts.
  3. Keep only selected, highly-filtered knowledge in your central repositories. The number one complaint of employees in leading knowledge-driven organizations is “I can’t find anything I’m looking for”. In central repositories, less is more. Get rid of anything that isn’t very current. Provide abstracts for all centrally-stored knowledge. Summarize, distill, analyze, synthesize to reduce documents to their essence. When in doubt, leave it out. Recognize the 80/20 rule and keep only the best 20% of what is available. Keep it only if the re-use value proposition is obvious. Index and organize it by potential (re)-use, not by subject taxonomies.
  4. Don’t overlook the value of plain-old ‘data’. The mining of raw high-quality data about customers and employees can produce extraordinary insights and opportunities. Data is not knowledge, but much valuable knowledge can be extracted from it.
  5. The bibliography may be more valuable than the document itself. The bibliography tells the reader where the expert went to get needed information. It can be a roadmap for others doing similar research, and can have a shelf life well beyond that of the document it was used to prepare.
  6. Don’t wait for people to look for it, send it out, using ‘killer’ channels. Just because people complain of e-mail and information overload doesn’t mean they don’t want and need timely, valuable information. If you’re a database manager, before you store anything, think who might find it valuable now and send it out. And send it out using the one or two ‘killer’ channels that users prefer: maybe e-mail, IM, voice-mail, whatever people complain least about in your company. And never broadcast — send individual messages to one person and start the message with their name. Personalized stuff gets read. 
  7. Create an internal market for your offerings by giving valuable stuff away. Analyses, research studies, trend studies, news synopses, surveys can be hugely valuable to both internal and external customers, but they’re hard to sell in the abstract. Give one away free to a customer and they’ll immediately see its value. It seems counter-intuitive, but it works.
This entry was posted in Working Smarter. Bookmark the permalink.


  1. Dave,First, the is an excellent post. I find myself relating to (and/or empathizing) with most of it.Do these seven points come from your head or Peter Drucker’s. No love lost either way you understand but I’m curious.I’d like to add one point. The ideas and theories presented here deal mainly with profit-oriented business/organizations. It may seem like a small distinction but there are other ways of doing “business” out there. Case in point – the military, or, for that matter, any GS organization. For my part, I regard myself as an expert in a rather narrow field and that’s what the military’s original motivation was in hiring me in the first place. The trouble is, in my organization, both middle and upper management is cycled every three to five years. New management comes in together with their own agendas and knowledge base. The efficiency of such an organization (or the lack therein) boggles the mind. I’m about to be “down-sized” myself. Updating my resume I find that my references are all “in the wind” and I can’t show off my current work because it’s all classified.I can’t remember what my original motivation was for replying but thanks for letting me vent.regards – rich

  2. Dave Pollard says:

    Thanks, Rich. The seven points are my own lessons, learned the hard way by making lots of mistakes, and studying the mistakes of others.

  3. Dave, Brilliant. Had to start thinking about how to stretch it. The case for blogging is building and it is not limited to just ideaflow. We should think big!

  4. edward says:

    I’m not sure that you are telling us all you know about this. I´d read your post in my rss feed and think that you can go further on this subject. Am I wrong ?edward

  5. Dave Pollard says:

    Stuart: I agree, and I enjoy reading your blog as well.Edward: Having worked in this area for 7 years I could write books about it. Let me know what your particular area of interest in KM is, and I’ll either post something or e-mail you something.

Comments are closed.