One Defining Moment

Originally published Aug 24, 2020

15 min readAug 24, 2020

I participated in a recent discussion in the KM4Dev community: “Data — Information — Knowledge Continuum.” This led me to review my thinking and that of others on the subject of definitions.

Here is the complete KM4Dev discussion:

From Kristin Strohecker on July 30, 2020

I posted this question internally within the World Bank Group also. Forgive me if there is already a lot of historical discussion on this on the KM4Dev website; I admit that I haven’t checked but plan do to so.

I’m curious if folks have thoughts about definitions in the data/information/knowledge continuum. I know there are tons of articles, blogs, book chapters about it (some including wisdom), and to be honest, I hadn’t felt the need to make that distinction so explicitly until now. We are working on an IEG data strategy and also a KM plan, and for that work, it’s important to define the scope of these exercises, which means we need to define the terms. It would be interesting to hear any thoughts from this community.

From Stan Garfield on July 31

Here are my thoughts: Yet Another Myth: The DIKW Pyramid Scheme

I see no need for creating pyramids, hierarchies, or other similar, meaningless representations.
I define knowledge as information in action.

From Ivan Butina on July 31

WOW, this is such an amazing collection of articles and perspectives on the topic! Thanks for having everything so organized and ready to share on KM topics, Stan! I plan to go through them, but it will take me a while.

Meanwhile, based on a quick read of your post I see that you and many KM seasoned practitioners and experts don’t agree with the pyramid. I see the point and all the arguments are good arguments (at least the basic ones I read in your post). However, we do need clear ways of explaining to colleagues in our organization what is knowledge in practical terms, from the day-to-day non-KM work perspective, what is data, and what is information. I agree with the person that commented on your post — Roy Follendor — that while the pyramid might not be the ultimate truth or the reality as it is 100% (is anything like that?), we do need simple ways of communicating, especially in a business or organizational context.

I have seven years of experience in KM applied in two big international organizations and we are SO BEHIND when it comes to KM. We lack the basics, people are still confused and don’t understand what KM is, what purpose it serves, etc. (yet we keep hiring staff and consultants with the KM title). How it relates to the data work, to the information management in ICT departments, to innovation and agile offices and initiatives, to the internal research departments, not to talk about learning etc. I suspect that a big aspect of this is defining and connecting these terms in practical terms, and what they mean in the context of an organization and its business/day-to-day work: what is data for us? what is information for us? what is knowledge for us? what do we mean by learning? how do we define innovation and what it means for us? and most importantly, what is the interplay between all of these, because it ends up being translated in organizational barriers and silos (departments, offices, etc.) that often don’t communicate with one another.

I appreciate and enjoy (to an extent) the theoretical debates and complex definitions, but we do need simple and clear definitions that could be adapted to a specific organizational context in practical terms. This is important to get out of the KM bubble and make sure every single employee understands what KM is and what its role within an organization is.

From Peter J. Bury on July 31

Ivan, could it be age, could it maturing, could it be arteriosclerosis, something else? What I do seriously wonder about is what makes you, the World Bank, and others, to look for simplicity, the need for clear definitions? Isn’t it much more important that people understand what true inclusive and respectful collaboration is all about? Isn’t human adventure after all about how we behave towards each other?

Once we agree and understand its importance and implications all the rest will seem so much easier to address and cope with.

From Kristin Strohecker on July 31

Stan, Ivan, and Peter,

Thanks for the input and discussion on this topic so far. As I mentioned in my original post, for many years I hadn’t given too much weight to making these distinctions before now. So far, I’ve mostly found them artificial and not particularly helpful. However, since I’m working on a couple of strategies which will result concrete action plans, I now see the importance of defining the scope of each, so it’s clear what is included in the actions. These will have impact on our leaders’ expectations, on our work programs, and also on how some of our staff understand our roles within the organization at a very concrete level. We need clarity for our teams but also for efficiency, to avoid duplication of effort or working at cross-purposes.

For a simple example, extracting knowledge from existing reports for re-use and sharing is related to but different from extracting data for the same purposes. There is overlap but different staff with different skills may be involved, and potentially different technologies and processes may be employed. So where to make the distinction? If others see the need for definition and distinction differently, that is also fine and interesting to learn about.

Anyway, I’m enjoying the conversation and resources so far and will hopefully arrive at something useful that I can share.

From Patrick Lambe on August 1

My thought is that it’s a very unreliable “continuum” — looking at the roots, it was originally put together in the 1970s by data science folks seeking a political justification for getting more corporate resources put into data management in the enterprise — that is why the linking “upwards” to knowledge and decision making (by executives) was so important.

In fact, if you look at what the descriptors actually describe, the typology is of categories that are of radically different natures from each other, with radically different affordances and constraints, that cannot sit logically on a continuum. The best outline of the arguments against it (and there are a few now) is David Williams’s 2014 JEMI article Models, Metaphors and Symbols for Information and Knowledge Systems. David also has some useful practical suggestions for how to approach the categories.

I think they can be defined in relation to each other, but NOT in a continuum, and without any inference that one “progresses” from one to the other up a hierarchy.

I tend to resist the data folks’ inclination to categorize any content as “data” whether “unstructured” or structured. I think it’s much easier to think of data as structured elements in designed formats for designed purposes. Information is the explicable stuff that can float around our environment between people in many communication forms, and “knowledge” is principally located in people’s heads and can be modeled from person to person through observation, learning, information transfer, communications, etc.

But please scrap the idea of the continuum or hierarchy idea, it’s really a logical dead end that doesn’t serve any practical purpose from a management point of view — except to persuade people with budgets, that if they fund data management, knowledge (or wisdom, horrors) will somehow automagically follow.

From Stephen Bounds on August 1

To be honest, the definitions that you should use are the ones that work for your organization. I think it’s possible to spend a lot of time arguing over semantics for no particular outcome. I would pick concepts that help your organization fix the problem it is facing.

Here are some possibilities:

Data and information are synonymous, simply meaning “anything explicitly captured within a system”
Data is rigid, like a database, while information is unstructured, like a document
Some use data to mean unenriched data (“facts”), while information relates to deductions obtainable from those facts (“insights”)
Data is the encoding of a signal; information is the utility to the recipient from transmission of that data (here the opposite of information is noise)

Lam’s typology of knowledge describes four aspects of knowledge based on whether it is individual or collective, tacit or explicit. From this we can similarly draw the lay boundary of the term “knowledge” several ways:

Knowledge is implicit and only relates to the individual skills and knowledge possessed by a person
Knowledge must relate to problem solving, whether memorized or written down
Knowledge is every aspect of a person or organization that determines how problems are solved, including embedded knowledge (also encompassing what is often termed “culture”)

I would caution, by the way, against any attempt to structure data, information and knowledge in a hierarchy unless you are knowingly seeking to implement a data processing / business intelligence hierarchy (definition #3 above). The concepts are rarely cleanly separable, especially if you are attempting to describe operations of a multi-agent system.

If you do want to incorporate the idea of information feeding into knowledge, I would recommend the Williams AKI model which is circular rather than hierarchical instead.

From Rajeev Bali on August 1

Apologies for the “fly by” nature of this short posting…my starting point would be that

Data consists of (almost meaningless) raw facts and numbers
Information adds context to this and organizes the data into something easily understandable
Knowledge attempts to answer the “so what”? question for the Information aspect (making sense of the information) — this leads to accurate decision-making

A contextual example may help: 5000 numbers representing student Course marks (Data); these same numbers presented as a graph which are perhaps split into gender/geographical area (Information); We can see that Males in South America seem to have struggled with this Course (Knowledge — we can now investigate the “So What?” aspect and make informed management decisions on this: do we need to allocate more resources in South America? Are female students performing well in South America? If so…let’s dig deeper).

Hope this helps, rather than hinders 😊.

From Ivan Butina on August 1

I think that it’s very important to define data vs. information vs. knowledge and I would love to hear from colleagues who have been defining these three terms and how they differ in practical rather than theoretical terms.

My sense is that there’s a lot of confusion on how these three terms are used — at times interchangeably — and that doesn’t do any good for management and colleagues to understand what KM is and what its role is within the organization. Also, if we don’t define and communicate well — practically — these terms and the interplay between them, it becomes more difficult to bring down the silos and collaborate between offices that work on data, information management, knowledge management, research, evaluation, etc.

From Pavel Kraus on August 1

Within Germany, Austria and Switzerland the six leading KM communities have updated the KM Glossary in 2020. In the meantime, we have also French and Italian translations, due to be published this Fall. You might look there as well. I hope we get the resources to make also an English translation. For now deepl.com will have to do.

From Ivan Butina on August 1

Rajeev, Pavel, Stephen, and Patrick,

While the question was asked by Kristin, I do want to thank you for your helpful answers and the resources you shared with us. This is definitely helpful to me for the reasons I shared.

In particular, I appreciate Rajeev’s simple definitions and practical example. I do think that we need to simplify the KM language when communicating outside the bubble.

I completely agree with Stephen when he writes: “To be honest, the definitions that you should use are the ones that work for your organisation.” I was going to come back to this thread and suggest the same. In fact, I would love to ask colleagues in my organization to define the three terms and how they relate in their own words, and then see the differences between colleagues from the different areas of work: data, information management, ICT, KM, monitoring, evaluation, research, programs (WASH, Nutrition, Education etc.). Then we could build our own shared definitions, that work for us and our work.

Patrick, thanks a lot for sharing and alternative model to the pyramid. As for your comments on data people expanding their definition of data, I have come across that only recently. I never asked myself what data is: it’s one of those terms we think we understand because it’s ever-more part of our day-to-day life (I see it on my phone every day) until we feel that someone has used it in a “wrong” way. In my previous organization, I never came across the issue of confusion between data, information, and knowledge. But where I am now I was struck when I realized that at times data is used to describe what to me would be information without any doubt. However, I’m not sure where this comes from (I wouldn’t necessarily “blame” my data colleagues).

And Pavel, thanks a lot for the resource you shared — my German is VERY rusty but glad to see resources that are in languages other than English. We need more of that!

Insights of Others

Here is what others have written about defining knowledge, knowledge management, and knowledge workers.

The Eleven Deadliest Sins of Knowledge Management by Liam Fahey and Laurence Prusak

Error 1: Not Developing a Working Definition of Knowledge

If knowledge is not something that is different from data or information, then there is nothing new or interesting in knowledge management. Yet many managers seem determinedly reluctant to distinguish between data and information on the one hand and knowledge on the other; and, more importantly, they seem reluctant to consider the implications of these distinctions.

The tendency to avoid grappling with what knowledge is should not be surprising. There is little in the education, training, or organizational experience of managers that prepares them for the deep-seated reflection and understanding required by the concept of knowledge. Moreover, this situation is exacerbated by some recent popular management literature that directly advocates not making distinctions between these concepts. The argument advanced by these authors is that contemplation of such distinctions distracts managers from the necessary task of managing. However, reflection upon concepts and the distinctions among and between them is the essence of the process of “knowing” or learning.

This is a critical error. It contributes directly to all of the errors noted in the rest of this article. Also, avoidance of grappling with a working understanding of knowledge leads to a dysfunctional environment for knowledge work. Many executives have told us they were extremely reluctant to even use the knowledge word and that they felt the anti-knowledge culture of their organizations compelled them to do knowledge work by stealth. “We had to disguise our knowledge project within a data warehousing architecture plan” is a true and representative response. In fairness, firms have been assaulted, at least since the 1960s, with multitudes of theories and nostrums that have often proved to be of questionable value. This has made many executives skeptical, if not downright hostile, to new ideas and programs.

“Selling” KM Inside Your Organization by Adriaan Jooste

Lesson #2: “Knowledge” and “Knowledge Management” must be defined in your terms.

What is the scope of knowledge?
What are the core business issues to be addressed?
What are the current pain points to be alleviated?
How do you define success for knowledge management? (don’t pick ROI)

Arian Ward’s “Definition” of Intellectual Capital and Knowledge by Arian Ward

I would like to challenge the concept of needing to define something. I find definitions to be extremely limiting. They become permanent representations of the thing being defined and leave no room for interpretation, adaptation, or evolution over time. Instead I like to think about the distinctions of something. This is an evolving description — a running list of the manifestations and characteristics of the thing. Distinctions allow something to take on different guises in different contexts. They allow you to accumulate a list of those manifestations and characteristics over time. They also allow you to include what something is not in addition to what it is, to set boundaries around the thing being described. They provide people with a much richer sense of meaning and understanding. In short, something moves from being a dead set of words to being alive in the mind of the receiver. Including different people’s definitions of intellectual capital or knowledge in a book or article is a method of beginning to list their distinctions.

David Gurteen: Profiles in Knowledge

Knowledge workers are those people who have taken responsibility for their work lives. They continually strive to understand the world about them and modify their work practices and behaviors to better meet their personal and organizational objectives. No one tells them what to do. They do not take “No” for an answer. They are self-motivated.

SIKM Leaders Community

What I Have Defined

In my career in knowledge management, I rarely, if ever, needed to define data, information, or knowledge. I have found it useful to define other specific terms, provide helpful lists, and answer questions about terminology. Here is a compilation of what I have written about these.

Knowledge & Knowledge Management Defined

I generally don’t spend time trying to define knowledge and knowledge management. Such attempts can lead to long, unsatisfying, and ultimately useless debates.
And I definitely avoid referring to a continuum of data, information, knowledge (and sometimes wisdom), often represented in the form of a pyramid. I don’t find any value in this.

Communities, Community Management, ESNs, and Social Media

Lists

Q&A

Leadership

Posts Containing Definitions

What You Should Define

I recommend not spending a lot of time on defining data, information, knowledge, wisdom, or knowledge management. Instead, devote your energies to defining the following details of how you will implement KM.

1. Identify the Top 3 Objectives for the KM Program.

2. Answer 9 Questions about people, process, and technology. Determine who will participate in the program, which basic processes will be required, and how tools will support the people and processes.

3. Articulate your vision. You must be able to passionately describe the end-state vision for your program. What does KM look like when it’s working? Establish a vision for how knowledge management should work, and relentlessly work towards making that vision a reality. See:

4. Define the KM strategy. These are specific actions that will be taken to implement the program. See:

10 Types of Knowledge Management Strategies

5. Define compelling use cases with clear advantages over existing alternatives. Don’t talk about adoption or rollout of a tool. Talk about the advantages of using it over existing alternatives. See:

6. Define the KM Program Governance. This includes:

Roles and job descriptions for KM leaders, project leaders, and knowledge assistants.
Composition of program staff, virtual teams, and KM communities.
Objectives and schedules for recurring conference calls and meetings.
Processes for creating and updating the plan of record and schedules for implementation, new releases, and reporting.
Process for decision making.

7. Specify the desired modes of knowledge flow. See:

5 Modes of Knowledge Flow

8. Select people, process, and technology components using knowledge management specialties such as Information Architecture, Design Thinking, User Experience, and Agile Development. Define implementation plans for key components such as training, communications, and change management.