Dave Snowden has been having a hoity toity about stupid categorisations over at Cognitive Edge, and he’s included the “People, Process, Technology” triad (some people add Content) into the Domain of the Damned. He asks: “All three have to co-evolve; why or why do people fall into the pit of reductionism and categorisation with such ease?”

Well I can think of one good reason why it’s good to have this trio (or quad) – not so much as reductive categories (Dave may be building a straw man here) but as useful reminders to people who need to span all three. Technology people do need to be reminded (forcefully and painfully sometimes) that process and people (and content) matter. People have to be reminded that magic only exists on TV and by sleight of hand, and the affordances and constraints of technology and tools have to be grappled with to be understood at the operational level. They don’t always understand that IT people can’t read their minds.

I see this trio (or quad) as a salutory mnemonic which has emerged from too many occasions where one or two of them have been privileged over the others – meaning that they have not co-evolved, and thereby the initiative has flopped disastrously.

That’s why I especially liked one of the Intranet Innovation Award winning cases this year, from a company called Transfield Services in Australia. The KM team there looked at a proposed rollout of Sharepoint Teamsites across the (global) company and evidently thought “aha! we need some process and people support here!”

So they put together their Gold award-winning “Teamsites in a Box” solution which helps new business units or communities set up their teamsites in a governed, user-and-collaboration friendly way. The solution helps a group define their requirements, deciding which functionalities get switched on or off, it helps them to define their objectives for the site and build a change management plan, there’s an intranet site about how to get the most out of a teamsite with good examples from within the company to look at, there are checklists and guidelines to help you through the setup and launch process, there’s training, both face to face and online, there is a governance structure with supporting policies looking after roles, ownership and responsibilities, deletion and archiving. ie They put people-friendly practices and structured processes in place to help the business units assimilate and get the best out the technology in a way that supports cross-company collaboration.

Though much of the rollout and management is what you might call “supported DIY” – ie locally owned – the initial set up is centrally provisioned by the KM team, and this enables them to look for prior examples of similar sites and iron out possible overlaps, or suggest linkages to other existing sites. Changes to the purpose, use or structure of the teamsites are also notified back to the central team, so they have a managed inventory of what the network of sites contains and what’s happening within it. A really nice mix of strong governance and guidance and light, local-context-friendly management.

Now this all looks like wonderful common sense. The fact it won a Gold Intranet Innovation Award suggests this kind of sense is not that common, and from what I’ve seen of normal teamsite rollouts (and Lotus teamrooms before them) is that structured, governed, consistent, supportive rollouts are more the exception than the norm. This is conscious co-evolution of people. process and technology.

Every now and then we struggle with clients who say something can’t be done within their organisation. They would never get permission or support from “them”. It’s even dangerous to ask “them”. Let’s just try to do this quietly without disturbing anyone, they suggest.

Unfortunately, if you’re trying to get any kind of large scale KM going, you can’t do it without “disturbing” people. At the end of the day, KM is not about what the KM team does, it’s about what “they” do out there.

So I was pleased to find this little gem of a post from Don Cohen from a couple of years back, about how a successful KM initiative started with an act of disobedience.

We have another client with a manager whom I would call flexibly stubborn. When one major initiative was rejected by senior management, where the more faint hearted might have just abandoned it, he took note of the sensitivities, went back and reworked it into a form that had fewer visible allergenic components, and gave it to them as a planned activity “for information”. Some important elements were dropped, but he made the judgment that they are currently unsolvable. At least something is getting done.

Now… how do we get knowledge managers to be more disobedient?

“Word Herder” takes issue with the Library of Congress’ allocation of the DDC classification 658.4038 to my book Organising Knowledge (ie Information Management, deeply buried behind Technology (Applied Sciences). It doesn’t seem like a very happy place to be, but then again, I’m not sure where in Dewey – or any single-tree hierarchical scheme – my book, or any general book on taxonomy work, would fit. We need facets.

A while back Mary Abraham tagged me on the meme of why when and where I blog, and over the past few days I’ve been coming to the conclusion that blogging for me is a way of storing my memory in the external environment. This is not just recall memory, but sustaining the continuity of thinking over time… blog posts help me to think things through discontinuously. As I come across topics and issues of interest to me, related to things I’m interested in thinking through, I try to relate and assimilate them. That’s also why categories are helpful memory mechanisms for me to recall where my thinking has been in the past.

So I appreciate it when I come across people who are also thinking things through as they blog. Lee Romero has started a series of blog posts (here and here) on how he’s developing some relatively simple metrics for the communities in his company. What I also like is the smart, thoughtful way he’s approaching the matter of metrics, not from the routine “how are we doing?” angle, but for how the metrics can give insight into the way that the communities are working, or raise questions worth asking.

Back to the meme: When do I blog? – Guilt (at not having blogged for a while), procrastination (from “real” work) or being grabbed by an idea. Where do I blog? Mostly at my desk, occasionally in airports. I tend not to blog so much when I travel.

Shawn has blogged about how we use or arrange our external environment to give us memory cues, partly based on a wonderfully serendipitous conversation we had yesterday in a bookstore, leaping from title to title and using them as cues for discussion and memory. Externalising our memories is also something a function that taxonomies fulfil, and is my official reason why I never allow anyone else to tidy my desk. It may look a mess, but whenever I have to sift through things to find that bloody document I know is in there somewhere I’m also being re-cued on all those interesting things I set aside to look at later.. That’s my story and I’m sticking to it.

Brain science tells us that we forget things so as to be able to focus our continuing or repeated attention on things that are important to us. It’s a filtering and discarding mechanism. We remember things if we are reminded of them. There’s a wonderful language learning website based on this principle, focused just on the memory aspects of language learning. It tests your memory of the target language by measuring your pauses and hesitations on receptive and productive tasks of the target language presented in context and then recalculates the optimal interval before it presents the language to you again, to lodge it in your long term memory.

The particular genius of this company, Cerego, is that they have also recognised the power of personalised context, so they have made the site a social networking site, where you can scrape content off webpages, videos, photographs, the system will identify the target language within it, and create learning content that you selected. And then you share it. There’s great potential in this approach for memory-intensive knowledge management and corporate learning needs, such as compliance-knowledge and technical knowledge.

But more broadly than this, it suggests that we need to design our environments to be continually reminding us of the things that matter, and not just assuming that if it’s in the database our task is done. We need cycles of reminding and remembering, and environmental cues to support this.

James Robertson has reviewed our KM Approaches, Methods and Tools – A Guidebook and said some very kind things about it:

“The value of this guidebook cannot be overstated. Too often, the potential benefits of knowledge management are lost under a sea of theory and jargon, making it hard for business people to put the techniques into practice. In their typical style, the Straits Knowledge team cut through all this, providing a no-nonsense description of KM techniques that is perhaps the clearest ever published. If you have an interest in knowledge management, this is a great place to start, and a wonderful jumping off point for further learning. Highly recommended.”

Two weeks ago I completed two weeks of reserve military exercise, at the end of which I participated in - observed, really - an After Action Review. It wasn’t the first military AAR that I had been a part of, but it was the first that I was impressed by. It wasn’t done the classic way, you know, going through the four standard questions. Instead, each key appointment holder asked to share what they would sustain (or do again) and what they would improve (or do differently). 

What really struck me was how frank those people were, in talking about what they would improve. It was tentamount to confessing that they hadn’t quite done such a good job. This is remarkable because as career soldiers surely their performance bonus hinges on, well, their performance. Why were they so willing to broadcast their own failings, I wondered. As far as i could tell, it had to do with the person leading the AAR. As the head honcho, the commander set the tone for the session. Because this was a staging exercise for a larger one to be held the following year, he was emphatic about how important it was to surface mistakes so that we wouldn’t repeat them the following year. He wasn’t commanding, but more pleading. I don’t think I was the only one who felt his earnestness. Judging by the way people were still so engaged 4 hrs into the session and closing to midnight, I’d say that he pulled it off. There’s something about listening to people’s failures that keeps us interested

Cory Banks has been talking on actKM about a project where he wants to share lessons learned through anecdotes about examples and incidents – I suppose to communicate the contexts where learning took place. One of the challenges in such cases is how to get the stories about failures (perceived or real). In the Ning social networking group The Mistake Bank, Cynthia Kurtz has just started a discussion thread about “mistake haikus” – “I’m wondering what would happen if people shrunk stories of mistakes down to a sentence or two. The point would be mainly to disguise the details of an embarassing story down to something so vague, yet still instructional, that it would be okay to tell.”

An interesting idea. Here’s my contribution, which still needs a little work!

Eager for challenge,
an impossible project
fails when winter bites.

A very generous concession speech from John McCain just now, asking his supporters to reach out and work with President-Elect Barack Obama to reach common ground and fix America’s problems. Hope indeed.

The learning ripples from the actKM conference are still spreading, on the actKM forum as well as in the blogosphere.

Matt Moore has published a very nice, pragmatic white paper based on his actKM conference session, giving guidance on how to justify your KM efforts to your various stakeholders. It forms a very nice procedural companion to the white paper I wrote some time back on “How to use KPIs in Knowledge Management”.

The assumption in Matt’s white paper is that you’ve been doing KM for some time and have some activity and output measures to show. Matt’s principal advice is to be aware of the political dimensions of decision making and support in your organisation – that your different stakeholders have different agendas, different things matter to different stakeholders.

It’s not just a mechanical numbers game. In fact, he says, numbers such as activity and downloads metrics “are vital to you but there’s a problem with them: No one else cares. And your task here is to make people care. So these numbers by themselves will not be enough.” So you need examples of impact to give context, understanding, and force to your numbers.

In a follow-up post on the actKM listserve, Brad Hinton makes the excellent point that numbers can mean different things in different contexts, which is why qualitative input is so important alongside the numbers to help interpret them correctly. He puts it so well that I’m quoting him in full (he kindly agreed):
—-

“When I had to supply metrics for the CoP’s I managed at a former workplace, I reported on activity (akin to hit rates) broken down in sub-categories like questions/answers/news/market opinion, etc. Because I often followed up on exchanges to hear from the person that received the greatest benefit, I often used their quotes in my report. And I would usually feature a “case study” of an exchange where the end result was positive (generated a new deal, saved time, helped my client, found a colleague with the info I needed, etc.).

“There were a couple of CoP’s that had relatively little action – low “hit rates”. But the value was often particularly high – for example, our CoP on the grains industry had low hit rates for a number of rational organisational and industry reasons but a monthly review of soil moisture, rainfall and crop conditions during a season were highly valued, particularly by lurkers in our credit department! By identifying the value (in this case risk mitigation on credits), low hit rates weren’t seen to be as “disappointing” to management as first thought.

“At inductions, I always used a couple of really successful “case studies” to showcase CoP’s as a way of learning on the job and making personal networks within the organisation.

“My conclusion is that “value” has to be regularly communicated and made visible using (where applicable) quantitative AND qualitative metrics.”
—-

And in a sign of a really good conference, one of the participants, Kerrie Anne Christian has blogged about how insights from the conference helped her address a sudden extinction threat to a number of CoPs in her company, and reminded her how important it is to keep the “elevator pitch” messages flowing to key stakeholders, so that they are aware of the value being created.

For me these two messages – helping our stakeholders see the value in terms that are meaningful to them, and looking beyond the blind numbers – are paramount to sustainability in KM. This is not about achieving salvation, where faith alone (depending on your tradition) can get us where we need to be. This is business in the mortal world. We have to justify what we are doing in terms of works, real impact on the business. And just playing the numbers game, as Matt hints, is effectively relying on blind faith. Nobody really knows what the numbers mean until we have concrete examples to show us what they mean.

I was sorry to have missed the actKM conference this year, but thanks to Matt Moore, we have some of the sounds from that venerable affair (they had a capuccino machine!).

Matt asked Mark, David, Arthur, Graham, Keith and Cory for their words of advice to new knowledge managers and kindly created a podcast. Here are the highlights:

• Don’t try to boil the ocean, focus on 2-3 things that help the business (Mark Schenk)
• Try to find small things that make a big difference (Mark Schenk)
• Don’t do things that other people can do (Mark Schenk)
• Focus on addressing business problems, don’t get distracted by the KM theory (David Gurteen)
• Find someone in management who cares, and then give them successes that make them look a hero (Arthur Shelley)
• Define your KM activities into a project and break it down into bite sized pieces (Graham Durant-Law)
• Be prepared to evolve, start small, and grow (Graham Durant-Law)
• Start in the middle (ie in an area of the operations) and grow it outwards (Keith De La Rue)
• Discover the KM-friendly activities that are already happening and work with them (Cory Banks)

Matt Moore and Dave Snowden have been having a scrap (read the comments) about politeness. It just seems a little incongruous. Neither eminence is well known for being troubled by such petty interferences with direct, startling and insightful expression.

So assuming it isn’t some elaborately staged (or improvised) joke here you have two people who have a very high regard for each other rubbing each other up in entirely the wrong way, and in consequence the original question for discussion posed by Matt (some sweeping advice by Dave Snowden at the actKM conference about what knowledge managers should focus on) gets entirely lost in the sound and the fury that ensues. Clearly, if it isn’t a joke, recourse to “politeness” among people who are naturally provocative is a proxy for “I’m feeling prickly today”.

I happen to think Matt asked a great question, one that many knowledge managers probably ask in their own heads but don’t dare speak out. It’s asking for some detailed examples that would help knowledge managers follow the advice. But it was posed provocatively and the hackles have been raised. Politeness is in the air. The intent of the question is lost.

Oh well. Sometimes it is just a bad hair day, and despite everyone’s best efforts, you’re not going to fix a disagreement or get to explore what you wanted to. I have found that forgetfulness is a great friend in such circumstances.

I wondered about blogging this. And I thought actually this happens all the time in KM and we never acknowledge it. Sometimes people just rub each other up the wrong way, or understand things in ways we never intended, and things don’t go rationally. Sometimes you’ve just got to go home, watch a mindless movie, reset your brain and start again the next day as if nothing ever happened.

Politeness, though… that’s a really good one.

For those of you who plan to travel by air with our KM Method Cards (who wouldn’t?) you might want to think twice about packing them like this:

When I was in Columbus for the iCKM conference last week, I went to see a doctor. I had been traveling for almost a week with a bad sinus problem that had started relatively benignly in Singapore and then worsened rapidly as I travelled.

For those of you who’ve had it, sinusitis + air travel = exploding head every time you make a descent. And I had been making a lot of descents. The other interesting effect was that it appeared to send jetlag effects into hyperdrive. So for the curious the formula looks something like this:

Sinusitis + airtravel = (Exploding head x # descents) + (jetlag effects x # hours time difference)2

I wasn’t keen to repeat the experience on the journey back to Singapore. The hotel suggested the nearby public hospital, but the only way to get attention as a walk-in patient was to go to Emergency and I couldn’t in conscience warrant that. The hospital suggested an “urgent care” centre at a local health centre, which does take walk-ins. So I walked in.

I was taken in hand by a grandmotherly lady named Madonna who told me “We’re not set up for international visitors but don’t you worry honey, we’ll get you seen”. She was right. The patient registration system had us both crouched in front of it as we figured out together how to trick it past US zip codes and phone numbers as validated required fields, missing US social security numbers, and no local emergency contact person in case I keeled over in the midst of being treated there (we gave that doubtful privilege to a person called Hyatt Regency).

And Madonna, bless her, did get me seen, though it took an age (a whole afternoon out of the conference) to do so.

Later on, as I waited in the local supermarket pharmacy to get my prescription filled, I spotted a front page story in the local Columbus newspaper. It said that local police were complaining that their (Federal) crime reporting system had no category for murder-suicides.

They either had to capture it as one thing or the other, but cases where someone killed another person then turned the gun on himself couldn’t be captured. In consequence they were having difficulty figuring out how bad the problem was, or whether there were any distinctive patterns or predictors in such cases.

So what started out as a simple medical expedition turned out to generate several taxonomic insights:

•absent categories can impair your ability to manage – if your taxonomy fails to predict the categories you need (eg short term visitors who are non-US citizens, combination crimes) then it’s very difficult to put formal management processes in place around those categories – they are technically invisible

•attitude matters – if Madonna hadn’t had the customer focus she had, I might easily have been excluded from her treatment world because technically, in the registration system, I could not exist

•improvisation around “invisibles” often happens but it is always provisional and depends on the attitudes and capabilities of the staff confronted by the problem. Even where the improvisation becomes habitual it tends to remain a tacit practice not captured in formal processes, so it remains invisible, and the fact that staff make the problems go away means that the formal managerial process is never confronted with the problem and therefore does not recognise or accommodate it – ie people who are forced to use poor taxonomies develop improvisation strategies that very quickly conceal the shortcomings of the taxonomy and hinder its correction.

I guess this is why the first rollout phase of a taxonomy or any management system built around fixed category structures, needs to be extremely responsive to mismatches and failed categorizations, so that improvisations don’t get enough time to bed down and conceal the problem. Secondly, when new category needs become apparent (eg we need to examine the phenomenon of murder-suicides), all systems need a way of alerting their taxonomy manager with the requirement.

The murder-suicide phenomenon by the way is a very good example of why single hierarchical taxonomies, where there is a single place for every entity, simply do not work in human systems. The Ohio police need facets. But that’s another story.

Yesterday I keynoted at iCKM in Columbus Ohio on “Accountability, Professionalism and Performance in Knowledge Management”. I drew a fairly bleak conclusion on all three elements as they are inextricably linked:

• Accountability – there is none
• Professionalism – we don’t know what it means
• Performance – we don’t know what it would look like

Ed Dale has done a pretty good job of summarising my main points.

Here are the session slides. I referred in the speech to the podcast interview with Dave Snowden and Larry Prusak, which is here, with a good session transcript guide linked here. I also presented the North American data from our iKMS survey on how much organisations seem to be investing in KM capabilities, and the full data tables from that survey are given together with a video speech on a related topic given last year here.

I didn’t end on a completely bleak note, nor do I agree that KM is dead (I seem to have given some participants that impression). In the last couple of slides I do look at some of the constructive things we can do (and are doing in Singapore) to break out of the viciously short non-learning cycles that knowledge managers go through.

If its a tough job being a knowledge manager, it’s an even tougher job to trust them with your organisation’s culture and infrastructure. A survey we conducted last year for iKMS found that most knowledge managers are teleported into their jobs from somewhere else, have very little prior knowledge or expertise in KM, get precious little training and development, and move on to something else within a year or two.

Of course, they could be appointed into that role because of some skill or competency they bring from somewhere else, so this state of affairs is not necessarily a disaster (and we know lots of good teleported knowledge managers) but it does impose incredible strains on these people as they struggle to get to grips with their responsibilities, new frameworks and approaches, and the complex political demands of a KM initiative.

Absolutely the most dangerous knowledge managers are the confident ones in their second KM jobs with a “success” under their belt. They are dangerous because they have typically been appointed on the back of their prior success, and they very often believe that this success represents the correct formula for KM implementation and it bears repeating. Again and again. It doesn’t of course, except by sheer random chance.

KM needs are highly context-sensitive, and even apparently similar organisations respond differently and function differently under the surface similarity. Business needs almost always differ. So while you can repeat the use of guiding frameworks and techniques across different organisations, the content of your implementation will be repeated at your peril. Danger signals here are the use of the phrase “When I was at… we…”. If you hear that from your knowledge managers, sack them. Immediately.

You can start to trust your knowledge managers when they are into their third or more role, when they have sat with a KM implementation for three or more years, and when they have several failures to talk about in a reflective way. Ie they are not burned out or depressed or too afraid to try anything but they are evidently observant, reflective, and ask lots of questions about what goes on in your organisation. And they experiment. When you talk to these knowledge managers about their past job roles, you will be struck by the diversity of their experience and the approaches they have taken, and the things they want to try. They are hungry to continue learning.

Then there are the “naturals”. These are the rare beasts who have a way of tuning into an organisation and reading it at the functional, political and cultural levels almost intuitively. These are the people who seem confident and inspire confidence but work hard at sustaining that confidence with results. So even if it’s only their first or second KM role, they have somehow already got those hard earned habits of their more battle-scarred seniors – they are watchful, reflective, questioning, they make a genuine effort to understand, and they manage risk while not being afraid of failure or discouraged by slow progress.

How do you know if someone is potentially a “natural”? One good sign is that the ratio of questions to answers is always greater than 1; another is they are pragmatic – they naturally zoom into the business goals, and don’t get overly distracted by the KM theory, fancy language or a neurotic preoccupation with “correct” process.

I’m going to be speaking on professionalism in KM at the iCKM Conference in Columbus Ohio next week. Give me a shout if you are going to be there and would like to meet, and especially if you want me to bring along postage-free copies of my Taxonomy Book, the KM Method Cards, the KM Approaches Methods and Tools book, or the iKMS KM Competency Framework!

Goodness, when did KM stop being about intangibles?

There is quite a strong body of KM practitioners in Israel, and they have been playing with the KM Method Cards to take a ground up approach to developing KM programmes, as Yigal Chamish reports:

“Each team (2-3 people) got several random cards of Approaches, Methods and Tools.
· Each team had to focus on a practical and real organisational need which will be supported by KM
· Each team had to develop a KM program to meet the need by choosing/using the Approaches, Methods and Tools given.
· Then, each team presented their program in front of colleagues for review and ideas

People were interested in participating and enthusiastic using the cards, and responses were positive and demonstrate satisfaction – the cards did come across as relevant to actual organisational life and need.”

Obviously you don’t want to force fit the approaches and methods to meet a need, so I guess this exercise (and the similar “Dominoes” version from Matt Moore the other day) is really helping people become aware of the variety of ways they can approach a KM need. What interests me so far in the examples of use we’re collecting is that the cards function as a focus for conversation, dialogue and learning as much as for brainstorming and planning. Ie they help people reach a shared understanding about what they should be doing in KM.

Here’s a picture of Yigal’s workshop in progress.

Something to think about from Randall Munroe, not unconnected to Peter Galison’s recent essay on the high cost of secrecy in the book Agnotology.

I’m a big fan of Nathan Wallace, the “let’s turn our intranet into a wiki” guy. He has just blogged about Juice, the initiative that won his team a Gold award in the Business Solutions category of StepTwo’s 2008 Intranet Innovation Awards. It seems simple enough: supplying new hires at his company with ICT equipment kits. But in true Wallace style, his team stepped beyond the procedural and aimed for transparency, user control balanced with accountability and outcomes that support the business and the internal client:

“The first impression on new starters for IT and the business as a whole has been transformed. On their first day we can now hand them a computer, mobile phone, mobile broadband, etc all configured and in working order. This truly shows that we are an efficient, organised company who value your arrival and expect you to be productive. In feedback to HR, basic logistics has gone from the number one frustration of new users to the item they are most likely to raise as unprompted positive feedback. For managers, the new starter process now requires no forms and works well regardless of the managers proactivity or experience in onboarding. This is a huge relief and reduced burden on their time.”

Now this isn’t really knowledge management, but knowledge managers could learn a lot from the Wallace approach, particularly with the key realisation early on that the IT Department needed to separate procurement (read, acquiring and implementing a KM system) from fulfilment (read, knowledge workers being happier, more productive and collaborative in their work) – and that they needed to focus much more on the fulfilment end to reach their objectives. Read the full case and examine your consciences.