A new breakdown every week — read the newsletter →
PeopleBusinessesTools
Strategies Mental ModelsDecision Tools Business ModelsFrameworksMoats
Learn Book SummariesReading Lists GuidesQuote CollectionsLearning Paths
Artificial IntelligenceNewsletter AboutContact
Decision Tools

Ladder of inference

Trace how you climbed from raw facts to a conclusion — and where you jumped.

Overview

The Ladder of Inference describes the invisible, split-second climb from observable reality to action: we select some data, add meaning, make assumptions, draw conclusions, adopt beliefs, and act — then the beliefs filter which data we notice next, a self-sealing loop. Most disagreements are two people standing on different rungs, each certain the other is blind.

The tool is a way to climb back down. By asking which specific data a conclusion rests on, and what meaning you added along the way, you can find the rung where the leap happened — and test it against reality instead of defending it.

When to use it

A conclusion or conflict feels certain but might rest on assumptions — disagreements, snap judgements about people, ‘obvious’ explanations.

How to use it

Step 1

Notice the conclusion

Name the belief or judgement you are about to act on.

Step 2

Ask what data it rests on

Which specific, observable facts led here — and which did you ignore?

Step 3

Surface the meaning you added

What interpretation or cultural read did you layer onto the facts?

Step 4

Test the assumptions

State the assumptions out loud. Are they facts, or guesses you have stopped questioning?

Step 5

Climb down and re-check

Return to the raw data with the loop visible, and see if a different reading fits better.

Worked example

A manager concludes a report is disengaged because they were quiet in the meeting (conclusion). The data: they said little. The added meaning: silence = disinterest. The untested assumption: everyone shows engagement by talking. Climbing down, the manager asks — and learns the report was processing and later sent the sharpest written feedback of anyone. The ladder caught a leap that would have damaged the relationship.

Common pitfalls

  • Confusing the meaning you added with the data itself.
  • Only using it on others’ reasoning, never your own.
  • Treating a plausible story as proof — plausibility is a rung, not the ground.

Frequently asked questions

Why is it called a ladder?

Because reasoning climbs from concrete facts at the bottom to abstract beliefs at the top, and the higher you go the further from the data you are.

What is the reflexive loop?

Your beliefs steer which data you notice next, which reinforces the beliefs — a self-confirming cycle the ladder helps break.

How do I use it in an argument?

Ask what data the other person’s view rests on, and share yours, so you compare rungs instead of trading conclusions.

Related tools