Master the DEED Framework: Transform How You Ask and Solve Problems

We face many daily problems—work scheduling, teaching design, business judgment, life choices—that can be grouped into four types: Description & Understanding (What), Causation & Explanation (Why), Evaluation & Decision (How), and Estimation & Prediction (What if).

In my book "Model, the Mathematics of Thinking" I call this the DEED framework:

D : Description (What)

E : Explanation (Why)

E : Estimation (What if)

D : Decision (How)

These four problem types form a mental map that guides us to find direction and methods in an uncertain world.

Basic Definitions and Boundaries of the Four Problems

1. What: Description & Understanding

This is the starting point of cognition. We must first ask: "What happened? What is the situation? What are the facts?" These questions build our basic picture of the world.

How severe is the recent heatwave in a certain region?

What is the teacher‑student ratio in a school?

What is the daily active user count of an application?

If the What cannot be described clearly, subsequent explanation, inference, and judgment have no basis.

2. Why: Causation & Explanation

After understanding the facts, we naturally ask: "Why is this so?" These questions aim to reveal mechanisms, causes, and structural drivers.

Why does this product have high user stickiness?

Why is educational resource distribution uneven in a region?

Why do young people tend to change jobs more frequently?

Why questions guide us to build models, identify causal paths, and constitute the first step toward deeper cognition.

3. How: Evaluation & Decision

Having mastered the current state (What) and understood the reasons (Why), we move to the action layer: "How should we respond?" These questions involve value judgments, standards, and action choices.

How to optimize teaching schedules to increase student engagement?

How to allocate resources effectively under budget constraints?

How to devise the optimal recruitment strategy?

How questions are goal‑oriented, requiring integration of the previous two types of information and balancing different interests.

4. What if: Estimation & Prediction

This type jumps beyond the present, contemplating change and the future. It usually starts with "If..." to simulate possible outcomes under altered conditions.

If policy changes, how will corporate tax costs vary?

If a new teaching tool is used, will student performance improve?

If traffic restrictions tighten, will city congestion ease?

What if questions serve as a "testing ground" for cognition and judgment, requiring hypothesis construction, conditional inference, and risk assessment.

Structural Re‑arrangement of Problems

Many people view the four problem types as static categories: What for observers, Why for analysts, How for decision‑makers, What if for futurists. In reality, problems are dynamic and often cross categories.

A good problem usually does not belong to a single class but can be re‑combined and evolved.

For example, a school principal introducing AI‑assisted teaching would ask:

What is the current awareness and acceptance of AI among teachers? (What)

Why do some teachers resist AI? Fear of replacement? Lack of skills? (Why)

If teachers are equipped with AI assistants, will teaching efficiency and student interest improve? (What if)

How to roll out the initiative—pilot first or school‑wide? How to evaluate its impact? (How)

This is not a simple sequence but a logical chain and structural combination forming a cognitive system.

System Pathways of the DEED Framework

From a cognitive perspective, the four problem types can form several typical pathways:

1. Classic Decision Path: What → Why → What if → How

Most common, moving from description to analysis, simulation, and rational choice; applicable to public governance, corporate strategy, education reform, etc.

2. Mechanism Reverse Path: Why → What → How → What if

Some research starts with mechanisms (e.g., medical studies) before describing phenomena and finally offering actions and risk assessment.

3. Prediction‑Driven Path: What if → What → Why → How

In uncertainty‑dominated fields (AI, finance, climate), one begins with future hypotheses, then back‑traces reality and causality to decide optimal responses.

4. Decision‑Push Path: How → What → Why → What if

Often decisions are forced (e.g., emergency response), requiring rapid cognition to fill facts and models.

These pathways can switch, nest, or loop. True expertise lies in flexibly navigating among them, combining as needed to build a closed cognitive loop.

Value of Problem Re‑structuring

If you only handle What, you are a data clerk; only Why, an analyst; only How, an executor; only What if, a dreamer. Mastery of all four and the ability to jump among them makes you a strategic thinker and systems builder.

The key is not just asking questions but connecting them—using what works, linking them cleverly, and forming a coherent system.

Becoming a Cognitive Expert

Complex problems rarely have a single‑type solution. You must move among the four actions:

Sometimes start with What and end with How.

Sometimes start with How and backtrack to What.

Sometimes get stuck at Why and jump to What if for expansion.

Sometimes simulate with What if and finalize with How.

A true cognitive expert not only asks good questions but also traverses, connects, and bridges between them.

Description, explanation, prediction, and decision are the four basic actions humans use to confront an uncertain world.

We spend our lives employing these four question types to understand and change the world, but often do so unconsciously without structuring or validating the underlying assumptions.

Only by consciously managing and linking these four modes can we turn knowledge into understanding and thought into power.