In my last post I discussed the question of knowledge loss in the form of the impending retirement of baby boomers. I wrote, “Organizational memory, like human memory, can be a stubborn thing to change and often results in a this is how we’ve always done it syndrome,” and suggested that maybe we don’t want to capture and continue to use their knowledge. Obviously, this is not something that should (or even could) be done, but the point was to look at the problem from “outside,” to try to get a more objective view of the situation.

The beauty of the human mind in memory formation, and thus learning, is that it is inherently flexible. I believe the same to be true of organizational memory and learning. The stubborness I mention in learning and memory above is a learned behavior that can be overcome, if the desire is there. Unfortunately, my experience has shown me that most people – and organizations – have this stubborn type of memory, which leads to the inability (or very hard time) and sometimes unwillingness to learn new things. And by new things I don’t mean new information molded to fit the way you already think of something.

A better way to describe these two aspects of memory/learning comes from The Power of Impossible Thinking: Transform the Business of Your Life and the Life of Your Business. In her discussion of the book, along with several other things under the title Flexing Mental Muscles, Evelyn Rodriguez provides the following quote:

• The first kind of learning, which is far more common and more easily achieved, is to deepen our knowledge within an existing mental model or discipline.
  

• The second kind of learning is focused on new mental models and shifting from one to another. It does not deepen knowledge in a specific model but rather looks at the world outside the model and adopts or develops new models to make sense of this broader world…Learning about new mental models is much more challenging and complex, but crucial in an environment of rapid change and uncertainty.

Johnnie Moore also discusses the ideas in his post Changing Mental Models.

For many years now I’ve read about and been involved in discussions about the impending retirement of baby boomers, the effect this will have on institutional memory, and what can be done about it. Most of my interest in this at the time concerned the impact on the federal government workforce, which will be very hard hit since the retirement age is a bit lower than the populace in general.

Though I’ve not yet read it, the book Lost Knowledge by Dave DeLong addresses this problem in great detail (more on the book can be found here, here, and here). A snippet from the book’s website:

Dr. David DeLong, a research fellow at MIT’s AgeLab, has just created the first comprehensive framework to help leaders retain critical organizational knowledge despite an aging workforce and increased turnover among mid-career employees.

Like most discussions of the topic I’ve been involved in, the book seems to focus on the negative aspects of people leaving, and taking their knowledge with them. However, I have been reading James Surowiecki’s The Wisdom of Crowds and think that we may be missing out on an opportunity to actively reinvent the corporate knowledge as we try, probably in vain, to keep the old knowledge around.

Granted, there is some information and there are many processes that must be recorded and retained. This the basic infrastructure of how an organization functions. But if you simply take the knowledge of people who are leaving and transfer that to the people that are replacing them, you are effectively eliminating the value of the “new blood” coming into the organization. Or, in the words of Surowiecki, you are maintaining homogeneity at the expense of diversity.

Organizational memory, like human memory, can be a stubborn thing to change and often results in the this is how we’ve always done it syndrome. An excellent description of memory formation can be found in Tony Buzan’s The Mind Map Book (sorry for the lengthy quote, but it bears repeating in whole):

Every time you have a thought, the biochemical/electromagnetic resistance along the pathway carrying that thought is reduced. It is like trying to clear a path through a forest. The first time is a struggle because you have to fight your way through the undergrowth. The second time you travel that way will be easier because of the clearing you did on your first journey. The more times you travel that path, the less resistance ther will be, until, after many repetitions, you have a wide, smooth track which requires little or no clearing. A similar function occurs in your brain: the more you repeat patterns or maps of thought, the less resistance there is to them. Therefore, and of greater significance, repetition in itself increases the probability of repetition (original emphasis). In other words, the more times a ‘mental event’ happens, the more likely it is to happen again.

When you are trying to learn something, this is obviously a good thing. However, the very nature of this learning process makes it more difficult to learn something new, especially if it is very different (“off the beaten path”). By pointing new people down the paths of the people that are retiring, you are ensuring that the well known paths will continue to thrive and that it will be harder to create new paths through the forest.

That’s fine if your goal is to continue on the path you are on, but it brings to mind an old proverb I saw somewhere: If you don’t change the path you are on, you’ll end up where it takes you.

…to be continued…

Found this one languishing in my drafts pile from July 16. Ooops.

In response to a comment I made to Maybe All KM is Personal KM, Tom Collins at Knowledge Aforethought wrote the following:

That means we have to build team collaboration tools and enterprise information systems (not to mention tools for connecting outside the enterprise, as well). My concern comes from the well-documented fact that vast resources have been poured into monolithic “enterprise solutions” without much work on making them accessible or usable by the individual knowledge worker.

Not too long ago I wrote a paper for my graduate studies along these lines. The specific question I asked was, “If an enterprise process can result in such large efficiencies at the enterprise level, why is it so difficult to get individuals to use the process?” In the specific case I investigated, I found that the problem was that while the new enterprise process resulted in a significant (huge would be an understatement) time savings enterprise wide, the process also resulted in a significant (about double) increase in the amount of time required of individuals to access the information from the process.

The “new” enterprise process: In a large (5000+), multi-tiered (5 layers) organization, use a web based document repository to publish/disseminate a telephone directory for the organization staff to the entire organization.

The “old” process: Use e-mail to disseminate the directory.

Sending a Word document through e-mail to a 5 layer organization, layer by layer, can take a while. In my example, it came out to about 87 hours of actual time, including the time required for people to forward, to detach the document so they have it locally, etc. It is important to note that this time does not include propagation delays in getting from layer to layer. Of course, the time required to publish the same document to a web based document repository and have it available to everyone in the organization was much smaller, in this case about 2 minutes.

87 hours 2 minutes. What’s the problem again?

The problem, as it turned out, was that when an individual in the organization would actually want to use this directory, accessing it on the web based system took nearly twice as long (16 seconds vs. 7 seconds) as simply opening it from their local hard drive.

Though the results here are for a specific case, personal experience tells us all that individuals – especially individuals who don’t care for “high tech” stuff in the first place – will do what they know and what works for them, no matter what the “higher-ups” want them to do. The failure is not (necessarily) in the enterprise system or the enterprise wide goals, but not taking into account the needs/wants of the individuals that make up the organization.

Update: Since I first wrote the above there has been a continued interesting discussion concerning Personal KM, for instance Personal Knowledge – an oxymoron?? and Gained in Translation.

Is there such a thing as too much communications? Just like so many things in an organizational setting, you have to look at it from (at least) two different perspectives: the organization as a whole and the individuals that make up the organization. From both perspectives, I think the answer to my question is a resounding, “Yes.”

From an individual knowledge worker standpoint, I think that it is reasonable to try to have as many potential means of communication as possible. However, from a practical standpoint I think that individuals need to limit the number of actual communications channels they have open and/or pay attention to at any given time. One reason is information overload, another is the fact that at some point you have to stop taking input and giving feedback and actually accomplish the task at hand.

From an organizational standpoint the problem is organizational entropy, which I wrote about a few months back. Simply stated, the problem is that as an organization gets bigger and the number of connections between members of the organization increases, the ability of these members to make efficient decisions decreases, unless steps are taken to decrease the entropy.

Again, a good model in which to see this in action is the brain. There you have a whole bunch of individuals – neurons – that make up an organization – the mind. Each neuron has a set number of communications channels, with specific weighted values for each, that it uses to bring in information, act, and send out information. Just imagine the chaos in your mind if each neuron connected to as many other neurons it could find….

The beauty of neuronal connections in the brain is that the number of connections is not permanent, but it is finite at any given point in time. As the mind learns, connections are formed, broken, formed again, and finally settle into a somewhat steady state. When something new is learned, new connections are formed and old ones may go away.

So it should be with organizations. Of course, the challenge is that we are trying to build an organization that is optimized, while organic things like the brain have quite a long time to grow their capabilities.

Not too long ago I had a boss who, not unlike many leaders in organizations, had an established set of operating guidelines for the organization he was responsible for. Among the list of about 10-12 general principles was: Process is important, but it is results that matter. Which got me thinking about what “results” means.

In this particular case, results meant something along the lines of, “Get done what needs to get done now.” There was little or no emphasis on any of the hallmarks of process, such as repeatability, standardization, improvement over time. Every time a task came up, it was as if it were the first time. Though there were projects, they were managed more on a task by task basis than an overall project basis. Your status was measured not on your past project success, but whether you were able to complete the most recent task.

Worse, to my mind, was that these short term results were chased with incredible vigor with very little concern for future requirements or the big picture of an overall project. While achieving a specific result can be (and usually is) a good thing, if that result is achieved at a large cost relative to the overall importance of the task in the big picture it is definitely a bad thing. The more resources you expend on achieving a specific short term result, the fewer resources you have available for the rest of the project. I hadn’t thought about this situation for several months, until last night when I came across my favorite golf movie, Tin Cup with Kevin Costner, on one of the movie channels.

[Spoilers ahead] To make a long story short, a driving range golf pro (Roy McAvoy) with a tendency to “lose it” in the big events works hard to overcome this tendency and qualifies to compete in the US Open. The par 5 18th hole presents an interesting, hard to resist challenge: get to the green in 2 shots. The problem, of course, is that the green seems to be just out of reach with a water hazard in play for any that come up short.

In the first three rounds, Roy tries and fails to make it across the water and onto the green. With his newly found calm head, he accepts this small defeat in pursuit of the US Open title, takes the “drop”, and finishes out the hole with a decent score. In the fourth and final round, Roy finds himself in the same situation. This time, however, there is an important difference: All he needs is par to tie, birdy to win, which he could do easily if he took the easy way and didn’t go for the green in two.

As you may have guessed (if you haven’t seen the movie), Roy goes for it. But unlike the previous three rounds, when the ball ends up in the water Roy does not take the drop but instead insists on hitting the ball again from where he is. The next, and the next, ball goes into the water until Roy only has one ball left in his bag. If this one goes in the water, he is disqualified and can not turn in his score card.

The goal, the project, in this case was to win the US Open. Achieving this goal required the adequate completion of 72 individual tasks, getting the ball in the hole in the fewest number of strokes possible. Unfortunately for Roy, he lost track of the big picture goal and focused on a short term goal related to a single task. While he eventually got the ball in the hole (in a very dramatic way), he used up more resources than he could afford on this one hole and failed to achieve the original goal, victory at the US Open.

Back to my original point: Results are important, but you need to have a process in place in which these results make sense. While achieving short term results may have short term benefit and gratification, it may not help the big picture and may in fact hurt.

Another example of the push and pull between individuals and the larger organization – which can also be seen as a question of decentralization vs. centralization – is discussed in Organization in the Way: How decentralization hobbles the user experience by Peter Merholz and linked to by Karl on his Information Management Weblog.

Putting, for example, web publishing capabilities in the hands of many de-centralized units within a business is the goal of many Content Mgt projects. But, without strong central leadership, organizations often lose site of the goals and audience. So, how best to structure web content creation, especially given that this role is often an add-on in many organizations?

From Learning and Training Innovations magazine is the article Making Time to Learn. The author presents several approaches to overcome the problem “lack of time is the major impediment to learning,” including:

• Make learning an official, voiced company or unit goal (iow, Give your employees permission to learn)
• Link training explicitly to individual goals
• Tie training to annual performance reviews

A key quote from the article:

Many companies understand the links between ongoing learning, a knowledgeable workforce, high performance, and ultimate business success. Truly enlightened companies also walk the talk. They make professional development a voiced, company goal, and they support the alignment of corporate, unit, and individual goals (emphasis added by me). Continued learning is an expectation and often a requirement- whether a course required upon becoming a manager, training for specific skills required for a new task, or a training program tied to a company event, such as budgeting, or to a company initiative, such as a change effort.

A lot of discussion of KM involves the relative value/importance of organizational vs. personal KM. A focus on aligning all aspects of the KM spectrum, as in the learning example here, I think is a more appropriate approach than a focus on any single aspect. Obviously, each individual area needs a focus to improve performance at that level, but only if it is not at the expense of the whole effort.

The other day my 13-year old (Zeke) decided he was going to sit down and “beat” the game SuperMario 64 on the Nintendo 64. We’ve had the game since it first came out with the N64 (I think in ’96) and he played the game quite a bit over the years, but he had never beaten the game before. (In point of fact, he never expressed much interest in beating it and simply enjoyed playing it.)

If you’re not familiar with the game, it is like many others: Your character is on a mission to save the princess, and to do that you have to navigate a bunch of worlds on a bunch of levels and overcome the enemy to collect coins and stars. If you collect all 120 stars, you’ve beaten the game. Not only do you have to figure out how the character works (moves, jumps, etc), you have to figure out how to get around.

To help him in his quest, Zeke used the “official” strategy guide for SuperMario 64. The strategy guide is basically a beginning to end description of the game, including maps of the various worlds/levels, location of stars and other special items, and tips on how to overcome enemies, etc. An excellent example, in other words, of best practices in the form of explicit knowledge of the game.

If you’re not familiar with these types of strategy guides or you’re not a gamer, your first thought may be something along the lines of, “But that’s cheating, if you know where everything is and how to do it.” As we all know, however, just because you have the instructions for something doesn’t mean you have the ability to execute those instructions.

Out of curiosity, I convinced Zeke to let me try one of the more difficult levels towards the end of the game. Strategy guide in hand, Zeke watching and giving advice, it still took me many (many many) tries before I had accumulated the skill (tacit knowledge) to do what I needed to do.

What Zeke brought to the game that allowed him to beat the game relatively quickly (in an almost non-stop all day marathon) was his tacit knowledge of the game including such things as how to control the character, best ways to get around in the worlds, and hitting the jump and punch button at just the right time to get done what you need. He hadn’t played this game in years, but it all came back to him as if it had never left.

I guess the short point of this long story is this: Best practices are good, and sometimes essential, but you can’t always rely on them alone to achieve what you need to do.