How to think from first principles
You do not need a complicated framework. Start with one problem and a few honest questions.
1. Name the outcome
Write what you want to achieve, without assuming a particular solution. “I need a new app” names a solution. “I need customers to complete this process more easily” names an outcome.
If the goal is a personal decision, state what matters to you. There is no reason to pretend your preferences are objective laws. A clear goal helps you compare possible answers.
2. Separate what you know from what you think
List the facts you can support and the assumptions you are relying on. For each important claim, ask what observation, definition, or evidence supports it.
User reports can be useful without being independently verified. Keep that distinction visible. If an unknown fact would change the answer, do not fill it in with a confident guess.
3. Identify the real limits
Some limits are structural: available resources, a required approval, or a necessary physical condition. Others come from a chosen process or a habit and may be changed.
Be specific about the decision and its timeframe. A constraint that holds today may be negotiable later. Calling everything “unchangeable” can hide the option you are trying to discover.
4. Question one expensive assumption
You do not need to doubt everything at once. Pick the assumption that most affects cost, effort, or outcome. Ask: must this be true for the goal to be achieved?
For example, a team may assume that every decision requires everyone in a meeting. That might be appropriate for some decisions and unnecessary for others. Inspect the actual approval requirement before redesigning the process.
5. Build the smallest useful answer
Use the facts and necessary limits to propose an answer that serves the goal. State the condition that matters most. If two options remain plausible, identify what information would distinguish them.
A short explanation should help someone judge the answer. It should not merely repeat the conclusion in different words. Prefer one concrete next step over a long list of general advice.
6. Test the answer
Choose a small action that could show the assumption is wrong. A demo can test whether a benefit is understandable. A workflow trial can test whether a missing approval causes a delay. An observation can test whether the proposed cause is present.
Change the recommendation when evidence changes. The point is not to defend a clever first answer; it is to arrive at a more reliable one.
A reusable prompt
“I want to achieve [outcome]. I know [facts]. My limits are [constraints]. Which assumption matters most, what follows from the facts, and what is one small way to test the answer?”
You can also describe your situation in ordinary language. The First Principles tool uses the same basic structure while showing the key answer first.
Have a question of your own?
Find the key answerBackground and related guides
Philosophical background: Stanford Encyclopedia of Philosophy: Aristotle’s Logic. The practical examples and method on this page are our illustrative explanations, not claims from that source.