Who Really Invented DIKW? From a T.S. Eliot Poem to Ackoff — and What It Means for Your Decisions

The DIKW hierarchy — Data, Information, Knowledge, Wisdom — is not a management consultant’s invention. Its roots reach back to a 1934 religious pageant play, passed through information science in the 1980s, and were formalised by a systems theorist in 1989. Knowing who built the ladder matters less than knowing which rung you are standing on when you make a decision. For founders and operators drowning in dashboards, the hierarchy is not a curiosity of intellectual history — it is a diagnostic for why so many well-resourced teams still make poor calls.

Why the origin story of DIKW is a founder’s problem, not a historian’s hobby

Every week, a founding team somewhere installs a new analytics platform, hires a data analyst, and waits for clarity that never quite arrives. The data multiplies; the decisions do not improve. This is not a tooling problem. It is a layer problem — a failure to distinguish between what the organisation has (data), what it understands (information), what it knows how to do (knowledge), and what it judges wisely (wisdom). The DIKW hierarchy names those layers. Tracing its genealogy honestly reveals something useful: the framework was never designed to describe a data pipeline. It was designed to describe the conditions under which human judgment becomes sound.

The thesis here is precise: DIKW is a decision architecture, not a data architecture. Founders who treat it as the latter will keep buying more data. Those who treat it as the former will start asking harder questions about what their organisation actually knows and whether the people making calls have the wisdom to act on that knowledge well.

Movement I: The poet who accidentally founded information science

In 1934, T.S. Eliot wrote a pageant play called The Rock, performed to raise funds for the building of new churches in London.¹ The play was a meditation on faith, modernity, and the spiritual costs of progress — not a treatise on epistemology. Yet buried in its choruses are three lines that would eventually seed an entire academic field. ² The lines ask, in descending order: where is the life lost in living; where is the wisdom lost in knowledge; where is the knowledge lost in information.

The sequence is striking in retrospect. Eliot was lamenting a loss — the sense that accumulating more of a lower-order thing crowds out the higher-order thing it was supposed to serve. More information, less knowledge. More knowledge, less wisdom. He was writing theology, but he was also, inadvertently, writing the first critique of information overload — nearly three decades before the term existed. ³ Crucially, Eliot’s original formulation contained no “data” layer. Data was a later addition by information scientists who needed a floor beneath information — a raw, uninterpreted substrate. The poet had started one rung higher.⁴

Movement II: The information scientists who built the scaffold

The lines from The Rock sat largely undisturbed in literary circles until the early 1980s, when the emerging field of information science began grappling with a practical problem: how do organisations turn raw data into something actionable? Harlan Cleveland, writing in The Futurist in December 1982, produced what is now regarded as one of the earliest formal treatments of the information–knowledge–wisdom progression in an organisational context, explicitly opening his argument by quoting Eliot’s hierarchy.⁵ Cleveland’s framing was explicitly managerial: he was interested in how information, unlike physical resources, does not deplete when shared — a property with significant implications for how organisations should govern what they know.

A few years later, Milan Zeleny, writing in 1987, articulated the hierarchy in the knowledge management domain and mapped its layers to a memorable set of epistemic states: “know-nothing” (data), “know-what” (information), “know-how” (knowledge), and “know-why” (wisdom).⁶ Zeleny’s formulation predates Ackoff’s more famous 1989 paper, and some scholars argue he deserves co-credit for the hierarchy’s formal entry into management literature.⁷ Zeleny later proposed adding a fifth layer — “enlightenment” — above wisdom, though this extension has not gained wide traction.⁸

Movement III: Ackoff formalises the pyramid — and makes it operational

Russell L. Ackoff, a systems theorist at the Wharton School, delivered what became the canonical statement of the hierarchy in his 1988 Presidential Address to the International Society for General Systems Research. That address was published in 1989 as “From Data to Wisdom” in the Journal of Applied Systems Analysis.⁹ Ackoff’s contribution was not the invention of the hierarchy but its operationalisation — he gave each layer a precise functional definition that made the model usable in management and systems design.

In Ackoff’s formulation, data are raw symbols representing properties of objects and events — the products of observation, carrying no inherent meaning.¹⁰ Information answers the questions who, what, where, and when — it is data given context and relevance. Knowledge is know-how: it enables the transformation of information into instructions for action. And wisdom, the apex, is the ability to increase effectiveness — it requires human judgment and, according to Ackoff, cannot be fully encapsulated within any formal system.¹¹

This last point deserves emphasis. Ackoff was explicit that wisdom is not a higher-resolution version of knowledge. It is categorically different: it involves ethics, values, and the capacity to act well under conditions of irreducible uncertainty. No algorithm produces wisdom. No dashboard surfaces it. It is the layer that belongs entirely to the human beings running the organisation.

Movement IV: The critics who made the model more useful

The DIKW hierarchy has attracted serious academic criticism, and founders should know the strongest objections — not to dismiss the model, but to use it more honestly.

Jennifer Rowley, in a widely cited 2007 paper in the Journal of Information Science, revisited the hierarchy across a broad range of textbooks and found that the definitions of each layer vary considerably across authors — there is no universally accepted account of what data, information, knowledge, or wisdom actually are.¹² The model is, in this sense, a family of related frameworks rather than a single precise theory.

Martin Frické, writing in the same journal in 2009, went further, arguing that the hierarchy contains a central logical error and rests on the philosophically dated positions of operationalism and inductivism.¹³ Frické’s sharpest point is that the model implies a strictly bottom-up process — raw data flows upward and is progressively refined into wisdom — when in reality, knowledge and prior understanding shape how data is collected and interpreted in the first place.¹⁴ You cannot observe without a framework for what counts as observable.

These critiques are valid. But they do not invalidate the model for practical use. A map that is not perfectly accurate is still better than no map when you are navigating unfamiliar terrain. What the critics establish is that DIKW should be held as a heuristic, not a law — a prompt for asking which layer a given organisational problem lives on, not a mechanical description of how cognition works.

Movement V: What this means for founders operating in a data-saturated world

Most modern startups have solved the data problem. Cloud infrastructure, product analytics, CRM systems, and financial dashboards mean that even a ten-person team can generate more data in a week than a 1990s enterprise could in a year. The bottleneck has moved. The scarce resource is not data — it is the capacity to move up the hierarchy.

Consider the three transitions the hierarchy demands:

1. Data to Information requires context. The same revenue number means something entirely different depending on whether it is compared to last month, to a competitor, or to the plan. Most analytics tools handle this reasonably well. This layer is largely solved.
2. Information to Knowledge requires pattern recognition and causal reasoning. It is the difference between knowing that churn spiked in Q3 (information) and knowing why it spiked and what to do about it (knowledge). This layer requires experienced operators, not just better software.
3. Knowledge to Wisdom requires judgment under uncertainty, ethical grounding, and the willingness to act when the data is incomplete — which it always is. This is the layer where founders earn or lose their organisations’ trust. It cannot be delegated to a model or a metric.

The practical implication is that most founder decisions fail not at the data layer but at the knowledge-to-wisdom transition. The team has the information. They may even have the knowledge. What they lack is the judgment to act on it decisively, ethically, and at the right moment. Eliot’s lament — where is the wisdom we have lost in knowledge — turns out to be a precise description of the modern founder’s predicament.

There is a second, less obvious implication. Because the hierarchy is not strictly linear — because knowledge shapes what data you collect, and wisdom shapes what knowledge you pursue — founders should invest in wisdom before they invest in data infrastructure. A team with sharp judgment and limited data will consistently outperform a team with abundant data and poor judgment. The former knows what questions to ask. The latter drowns in answers to questions nobody needed answered.

What this means

The forward view: decisions, not feeds

The information environment facing founders in 2025 is not going to simplify. Generative AI is accelerating the production of information and compressing the time required to move from data to knowledge. What it cannot compress is the time required to develop wisdom — the accumulated judgment that comes from making consequential decisions, living with their outcomes, and integrating that experience into a more refined understanding of how the world works.

Eliot’s three questions — where is the life, the wisdom, the knowledge we have lost — were a lament about a civilisation that had confused accumulation with understanding. The founders who will build durable organisations are those who resist the same confusion: who treat their data infrastructure as a means to an end, who invest in the human conditions that produce knowledge and wisdom, and who remember that the pyramid has an apex for a reason.

The hierarchy was never about the data. It was always about the decision.

Business Growth Accelerator (a FounderWise brand) works with founders and operators on the judgment infrastructure behind high-stakes decisions — the knowledge and wisdom layers that no dashboard can replace. If this framing resonates with a challenge your organisation is navigating, start a conversation.