In everyday conversations we typically use the terms data, information, knowledge, and learning fairly loosely. And then there are those related, even fuzzier terms that we often toss around such as insights, wisdom, and so on. But it really pays to understand more precisely the distinctions among these terms and exactly how they all fit together because only with this increased clarity can we enable our organizations to perform at their best.
In the previous blog we focused on learning and drilled down to its core characteristic. At its most fundamental, learning is the process of better predicting, and that better predicting is achieved through the reduction of uncertainty about what is being predicted. Importantly, this fundamental nature of learning applies to both people learning and machine learning.
And we saw that actionable learning is distinguished from learning in general in that actionable learning is learning that can potentially change a decision from what it would otherwise be. Furthermore, we know from the field of decision analysis that actionable learning necessarily has positive value—a value that can be rigorously quantified.
But how does learning and actionable learning relate to information and knowledge? Well, let’s first examine how they are different. A distinguishing feature is that information and knowledge can be measured at a point in time. Learning, on the other hand, is dynamic—it’s a process—the process of improving predictions by reducing uncertainty. It can therefore only be measured over periods of time. In other words, information and knowledge are stocks, while learning can be thought of as a flow. Learn is a verb, information and knowledge are nouns. Likewise, nouns such as insights and wisdom are simply labels that we apply to knowledge that is perceived to be particularly powerful.
What about information and knowledge, how are they distinguished? Well, knowledge can be defined as information (which itself is derived from data that has been conditioned appropriately) that is sufficiently transformed to enable actions. That is, there is a causal connection between knowledge and actions. And again, since insights and wisdom are just particular forms of knowledge, there is a causal connection between these forms of knowledge and actions.
How knowledge relates to learning is very simple—knowledge is a result of learning. When we learn, our level of knowledge changes (is enhanced). That is, learning causes a positive change in knowledge. And if the learning causes a change in knowledge that causes a change in actions, then that learning is a special type of learning, actionable learning.
We can succinctly summarize these relationships in a quasi-mathematical form as illustrated in the following diagram. The “–>” symbol represents a causation and the delta symbol represents change.
The diagram depicts that knowledge can cause actions. Learning causes a change in knowledge. Actionable learning is a change in knowledge that causes a change in actions. And our optimizing data-to-learning-to-action (DLA) method causes a change (an increase) in actionable learning, which is equivalent to learning to learn better. And learning to learn better is the way to achieve sustainable competitive advantage.
Note that all these relationships apply whether the learning is associated with people or machines, and whether the knowledge is in people’s minds or is embodied within computer systems. These are universal relationships.
These fundamental relationships can help us more clearly think about value. For example, to be most precise, a given state of knowledge doesn’t generate value; it is the attaining of that given state of knowledge that generates incremental value. But even then, value is generated only if the increased knowledge causes different actions to be taken. A given state of knowledge causes actions, but it always causes the same actions for any given decision and associated environment. The value of all those actions is attributable to the process of getting to that state of knowledge from the prior state of knowledge (if we attributed the value generated to both a given state of knowledge itself and the achieving of that state of knowledge, we would be double counting the value), with the value of achieving the new state of knowledge simply being the difference between the value of the new actions compared to that of the actions that would be taken given the prior state of knowledge. And, of course, that process of getting to the new state of knowledge is learning. Knowledge may be power, but learning is value!
As an example, consider knowledge management. Managing knowledge in of itself doesn’t generate value. The value comes when new knowledge is attained (i.e., learning) by an actor (people or machine) that changes the way they do things (i.e., changes actions) because of the new knowledge.
This more precise understanding of information, knowledge, learning, and value helps us avoid or overcome some common pitfalls. Turning again to knowledge management, for example, if we don’t explicitly consider the direct relationship between knowledge and learning (both in minds and machines), as well as the distinction between learning and actionable learning, there is a danger that knowledge management functions can become overly siloed or misdirected. Similarly, understanding that it is learning that ultimately generates value can help keep a “big data” initiative from becoming untethered from the ultimate drivers of value for an organization.
It should also be clear from these insights why the optimizing data-to-learning-to-action method focuses on learning and decisions and works backwards from there. Any other approach risks putting value in the back seat rather than the driver’s seat.
Leave A Comment