Login / Register ID | EN
This page has no official English version. It was translated automatically and may contain errors. Read the original in Indonesian →
Data, Informasi, dan Pengetahuan: Tiga Pertanyaan Uji dan Batas Model DIKW
Foto: Pexels
Office IT

Data, Information, and Knowledge: Three Test Questions and the Limits of the DIKW Model

Tim Taalenta · 9 Oktober 2026

In meetings, these three words are often used interchangeably: data, information, and insight. However, each refers to something different. As a result, teams collect tables hundreds of rows thick, then become confused about what to do with them. The most common reference for distinguishing them is the DIKW model (data, information, knowledge, wisdom). This article summarizes the original formulation, three test questions you can use at your workspace, and the accompanying scientific critiques.

Original formulation by Ackoff

This model is most often associated with Russell Ackoff, through his 1989 writing titled "From Data to Wisdom." According to the ISKO encyclopedia entry written by Martin Frické, Ackoff's formulation is roughly as follows:

  • Data: symbols that represent the characteristics of an object or event, the result of observation or measurement.
  • Information: data that has been made relevant, usually by answering the questions who, what, where, when, or how much.
  • Knowledge: know-how, for example, knowing how a system works, so that information can be transformed into action.
  • Wisdom: added value through judgment, which is a human function that cannot simply be delegated to machines.

Ackoff also included "understanding" as a fifth level, but this level is usually omitted in concise versions.

Three test questions

The difference between data and information according to Ackoff is functional, not about form. This means the same table can be data for one person and information for another, depending on whether it answers a question. The examples below are fictional for illustration, not real data.

Layer Test Question Illustrative Example
Data What is recorded? 1,240 transaction rows: date, product code, quantity
Information Is one concrete question answered? Sales of product A this month decreased by 18% compared to last month
Knowledge Do you know how to trace the cause and take action? You know how to break down the decline by week and region, then adjust the stock schedule

The way to use it is simple. Before creating a dashboard, first write down one question you want to answer. If the graph is created without a question, the result is just data that has been beautified. If the question is answered but no one knows the next steps, it is merely information.

Critique of the model

This model has many opponents, and it is advisable for you to be aware of them before using it as a law.

Jennifer Rowley, in the Journal of Information Science 2007 (volume 33, number 2, pages 163 to 180), compared the definitions of each level in information systems and knowledge management textbooks. The result: definitions tend to rely on the level below them, while the process of moving from one level to the next is not agreed upon. Wisdom itself is rarely discussed.

Frické, in the same journal in 2009 (volume 35, number 2, pages 131 to 142), argued that this hierarchy is logically and methodologically flawed because it relies on inductivism and operationalism. The ISKO entry also notes its practical criticism: this model encourages people to hoard data in the hope that it will eventually become information. The conclusion of the entry is that the DIKW pyramid is now generally considered unsatisfactory as a theory.

Used as a thinking aid

So treat DIKW as a thinking aid, not an explanation of how knowledge is truly formed. The three test questions above are sufficient to avoid the most common mistake: labeling a pile of numbers as insight. If you prefer to see the explanation directly, there is a short clip on the video page.

Sources

Read also:

Ditulis oleh Tim Taalenta