Comparing Data, Information, and Knowledge: A Hierarchy of Human Intellectual Assets

Walk into any large research library and you are surrounded by three different things at once, even if they all look like books and screens to the casual eye. There are bare facts waiting to be read, neatly organised reports that answer specific questions, and the deep understanding that a scholar carries in their head after years of study. In everyday speech we lump all of this together and call it “information.” But in Library and Information Science, data, information, and knowledge are three distinct layers of human intellectual assets, each built on the one below it. Understanding how they differ – and why knowledge sits at the top – is one of the first conceptual skills every information professional needs.

A hierarchy of intellectual assets 🔗

The most widely used way to picture this relationship is the DIKW hierarchy, short for Data, Information, Knowledge, and Wisdom. It is usually drawn as a pyramid with data forming the broad base, information resting above it, knowledge higher still, and wisdom at the narrow apex. The model has been used widely across information science and knowledge management as a way to explain how raw observations are gradually refined into something genuinely useful. For our purposes, the first three levels matter most, because data, information, and knowledge are the assets that libraries and information centres collect, organise, and serve to users every day.

The framework is most often traced to the organisational theorist Russell Ackoff, whose 1988 address and the resulting paper argued that data can be used to create information, information to create knowledge, and knowledge to create wisdom. The core idea is simple but powerful: value increases as you move up the pyramid. Each step adds meaning, context, and human interpretation, so the higher you climb, the more useful – and the scarcer – the asset becomes.

Data as raw material 🔗

At the bottom of the pyramid sits data. Data is made up of raw, unprocessed facts and figures that have been recorded but not yet interpreted. It can take the form of numbers, words, symbols, images, measurements, or observations gathered through surveys, experiments, sensors, or simple counting. On its own, data has no context and answers no questions. The pyramid model holds that raw data without context lacks inherent value, which is exactly why we rarely stop at this level.

What raw data looks like 🔗

Think of a string of numbers like 23, 27, and 31. By themselves they tell you nothing. They could be temperatures, ages, exam scores, or house numbers. A reading on a rain gauge, a single entry in a hospital register, or the price printed on one grocery bill is all data. It is real and it is recorded, but it is isolated. Data is the building block of everything stored in a library, yet a patron almost never wants data alone. They want something done with it. Ackoff described data as symbols that represent the properties of objects and events – accurate, but inert until someone works on them.

From data to information: adding context and meaning 🔗

Data becomes information when it is processed, structured, and given context. Processing might mean sorting, categorising, calculating, summarising, or simply arranging the data so that it answers a question. Information typically responds to the questions who, what, where, and when. It is data that has been made meaningful and useful for a particular purpose.

How processing creates information 🔗

Return to those numbers – 23, 27, and 31. Labelled as “the average daily temperatures of a city across three consecutive days,” they suddenly mean something. The same numbers, when arranged to show a population rising over a decade, become a meaningful trend. Ackoff offered a classic public-administration example: census takers collect data, and the statistical office then converts that data into the tables and figures published in official abstracts. The raw responses gathered from households are data; the population totals, literacy rates, and growth percentages presented in a census report are information.

This is the everyday work of any information centre. A library does not just store facts; it organises them so users can find answers. A statistical yearbook, a bibliography, an indexed database, or a catalogue record all represent data that has been processed into a form people can actually use. The key shift from data to information is the addition of context and structure.

Knowledge as organized information 🔗

Climb one more level and you reach knowledge. Knowledge arises when information is organised, connected, and internalised through experience, reflection, and application. If information answers who, what, where, and when, knowledge begins to answer how and why. It involves recognising patterns across many pieces of information, forming generalisations, and building the theories and principles that let a person act competently in new situations.

From facts to theories and generalisations 🔗

Consider a meteorologist who has studied years of temperature, humidity, and pressure information. They do not merely know that it rained on certain dates; they understand the patterns that produce rainfall and can forecast tomorrow’s weather. That capacity to generalise and predict is knowledge. The knowledge pyramid is treated as a canonical model precisely because it is simple and intuitive, showing how each level grows more complex and more valuable than the last.

An important feature of knowledge is that it is partly personal and dynamic. The same report can be read by ten people, but each builds slightly different knowledge depending on what they already understand. Knowledge is not just stored on a shelf – it lives in people who have absorbed and interpreted information. This is why, in library science, information is seen as something that can be easily recorded, retrieved, and shared, while knowledge is the deeper understanding a person gains after processing that information and applying it to real situations.

Why knowledge matters most in society 🔗

If value rises as we move up the pyramid, then knowledge is the most precious of the three assets for individuals, organisations, and nations. Data is abundant and cheap to generate, especially in a digital age where every transaction and click is recorded. Information is plentiful too. But knowledge – the ability to interpret, decide, and act wisely – is comparatively rare and difficult to build. This is the logic behind the term knowledge economy, where competitive advantage comes less from physical resources and more from the ability to create and use knowledge effectively.

Knowledge as a “stock,” information as a “flow” 🔗

A useful way to grasp why knowledge is so valuable comes from economics, which distinguishes between a stock and a flow. A stock is a quantity of something accumulated and held at a point in time, like the water sitting in a reservoir. A flow is the rate at which something moves, like the water running through a pipe. Applied to organisations, knowledge stocks represent accumulated knowledge while flows represent how it moves and evolves through teams over time.

Information behaves like a flow. It streams constantly through an organisation – emails, reports, notifications, news – and much of it is consumed and forgotten quickly. Knowledge behaves like a stock. It is the accumulated reservoir of understanding, skills, and expertise that an organisation has built up and can draw on whenever it needs to make decisions or solve problems. Researchers describe a firm’s knowledge management capability as its ability to accumulate critical knowledge resources and manage their use, treating these resources in exactly the stock-and-flow terms borrowed from economics. A company can lose access to today’s information flow and survive; if it loses its knowledge stock – its experienced people and their understanding – it is in deep trouble.

For libraries and information centres this distinction has practical weight. Many institutions, especially special and corporate libraries, increasingly position themselves as knowledge management centres rather than mere storehouses of documents. Their task is no longer only to manage the flow of information, but to help the parent organisation identify, retain, and share its accumulated knowledge stock so that expertise is not lost when individuals move on.

Where the model is debated 🔗

It is worth knowing that the neat pyramid is not accepted by everyone. The philosopher Martin Frické has been a prominent critic, arguing in a well-known paper that the elements may not be as cleanly related as the hierarchy suggests, and that data need not always be processed to yield information. Others point out that the leap from knowledge to wisdom is left vague, and that much human knowing is tacit and experiential rather than something simply extracted from data. The Springer reference work notes that some theorists in library and information science have used DIKW to account for the logical and conceptual constructions that interest them, while business managers have valued it for tackling practical information challenges. Treating the model as a helpful map rather than a perfect description of reality keeps it useful without overstating its precision.

For a student of information science, the takeaway is steady even amid the debate. Data, information, and knowledge are genuinely different things, the value rises as you ascend, and the highest of the three is the one that lets people and societies act with understanding. Recognising which level you are dealing with shapes how you collect it, organise it, and serve it to those who need it.

What do you think? Can you trace a single example from your own studies – perhaps your exam marks – as it moves from raw data, to information, to genuine knowledge? And if knowledge is an organisation’s most valuable “stock,” what should a library do differently when it sees itself as a keeper of knowledge rather than just a manager of information?

References

1. https://www.isko.org/cyclo/dikw
2. https://faculty.ung.edu/kmelton/documents/datawisdom.pdf
3. https://www.ebsco.com/research-starters/library-and-information-science/dikw-pyramid
4. https://www.mdpi.com/2673-9585/3/2/14
5. https://www.thoughtworks.com/radar/techniques/knowledge-flows-over-knowledge-stocks
6. https://www.sciencedirect.com/science/article/abs/pii/S0378720611000905
7. https://link.springer.com/rwe/10.1007/978-3-319-32001-4_331-1