What is a mental model?

A mental model is a simplified idea of how something works that you can carry from one problem to another. Opportunity cost, second-order effects and the margin of safety are examples. Charlie Munger argued that a few dozen of them, drawn from different fields and used together, beat expertise in any one.

Which model should I use?

Start with the suggestions. They come from the kind of decision, the words in your question and how hard it is to undo, and each says why it was picked. A decision you cannot reverse always gets inversion and the margin of safety near the top. You can add any of the twelve at any time.

How many lenses does one decision need?

It depends on how hard it is to undo. Jeff Bezos split decisions into two-way doors, which you can walk back through, and one-way doors, which you cannot. The desk asks for two lenses on a decision that is easy to undo, three on one that is costly to undo and four on one you cannot undo, and says when you have fewer.

What does it mean when the lenses disagree?

That the decision is not as clear as it felt, which is the point of using more than one. When lenses that look at different things point the same way, you can be more confident. When they split, the desk names the lenses on the smaller side, because that is where the thing you have not resolved usually is.

Why write down my confidence and a review date?

Because memory rewrites decisions once the outcome is known. Annie Duke calls judging a decision by how it turned out resulting. Writing down what you expected and how sure you were lets you look back fairly. The review asks two separate questions, how it turned out and whether it was a good call given what you knew, and over time the desk compares your confidence with your results.

No. The desk asks questions and keeps your answers in order; it does not know your situation and does not tell you what to do. For a decision with legal, medical or financial consequences, talk to someone qualified as well.

Where are my decisions saved?

In this browser, as you type, and nowhere else unless you are signed in, in which case they are also saved to your account so they follow you to another device. You can copy a decision as text or download it as Markdown.

Keep this tool for later

Tools
Mental Model Explorer

Look at a decision through more than one lens.

Write the decision as a question, and the desk suggests mental models that fit it: a pre-mortem, a consequence ladder, three cases with odds. Each lens says which way it points, and the desk shows where they agree and where they split. Then write down what you decided and how sure you were, and come back to review it.

Open the desk Start from an example

12 mental models
8 worked examples
Free, no account

Should I take the job offer in Lisbon?

Work · Costly to undo · 4 lenses, 3 needed

Look ahead · Consequence ladder

Second-order thinking

If you do it

A hard first six months, in a new language
A wider network and a role I can grow into
Options I would not have had if I stayed

If you don’t

Nothing changes, which is comfortable
The same role and a year more bored
I start to resent the job I kept

Staying is easy now and costly later. Going is the other way round.

For it
Count the cost · Three questions

Opportunity cost

What does saying yes make impossible?
Being near my sister when the baby comes in spring

What is the best other use of the same time, money or attention?
Pushing for the promotion here, which might come in a year

Would you still choose it if you had half the free time you have?
Yes. The role is the reason, not the city

The best alternative is a maybe. This one is a yes on the table.

For it
See the risk · Needs against slack

Margin of safety

Savings33% short Has 4 of 6 months of costs

Time to find a flat50% short Has 3 of 6 weeks

Notice at my jobno margin Has 4 of 4 weeks

Two of the three buffers are short. Ask for a later start.

Unsure
See the risk · Pre-mortem

Inversion

I took it for the city and the work was the sameGuard: meet the team and see the first project before I sign

The money ran out before the first paycheckGuard: ask them to pay for the move

I was lonely by the winterGuard: book two trips home before I go

Three ways it fails, and I can guard against only one of them today.

Against it

Where the lenses pointThe lenses disagree2 for and 1 against. Before you decide, look again at Inversion.

Second-order thinkingfor

Opportunity costfor

Margin of safetyunsure

Inversionagainst

How it works

Ask the question, look at it several ways, then keep a record you can check later. All on one screen, saved as you go.

Should I take the job offer in Lisbon?

Work Easy to undoCostly to undoCan't undo

Second-order thinking Opportunity cost Margin of safety Inversion

Ask the question

Write the decision as a question, pick what kind it is, and say how hard it would be to undo. The desk suggests the lenses that fit, and says why it picked each one.

See the risk

Inversion

Imagine it has already gone wrong, and work out why.

It is a year from now and this went badly. Why?

It went wrong because…Guard against it by…

I took it for the city and the work was the sameMeet the team and see the first project before I sign

The money ran out before the first paycheckAsk them to pay for the move

I was lonely by the winterBook two trips home before I go

Look through the lenses

Each lens is a short worksheet: reasons it failed with a guard for each, what happens next and after that, three cases with odds. Each one ends with which way it points, and the desk reads them together.

Which way does this lens point?

For itUnsureAgainst it

In one lineThree ways it fails, and I can guard against only one of them today.

Decide, then review

Write what you decided, what you expect, how sure you are and your first step. The desk sets a review date from the stakes, and on the day asks how it turned out and whether it was a good call.

12 lenses in 4 groups

Each one looks at something the others miss, and each names its own blind spot. The worksheet a lens opens is shaped to its question.

See the risk3 lenses
- Inversion Pre-mortem
  Imagine it has already gone wrong, and work out why.
  
  Charlie Munger, after Carl Jacobi, and Gary Klein’s pre-mortem (2007)
- Probabilistic thinking Three cases
  Treat the decision as a bet, with odds on each outcome.
  
  Annie Duke, Thinking in Bets (2018)
- Margin of safety Needs against slack
  Leave room for things to go worse than planned.
  
  Benjamin Graham, The Intelligent Investor (1949)
Count the cost3 lenses
- Opportunity cost Three questions
  Every yes is a no to something else.
  
  Friedrich von Wieser named it in 1914; Greg McKeown’s Essentialism (2014) applies it to a life
- Sunk cost Three questions
  What you have already spent is gone either way.
  
  Hal Arkes and Catherine Blumer, The Psychology of Sunk Cost (1985)
- Value equation Four levers
  Value rises with the outcome and its odds, and falls with the wait and the effort.
  
  Alex Hormozi, $100M Offers (2021)
Look ahead3 lenses
- Second-order thinking Consequence ladder
  Ask “and then what?” more than once.
  
  Howard Marks, The Most Important Thing (2011)
- Explore vs exploit Three questions
  Try something new, or double down on what works?
  
  Brian Christian and Tom Griffiths, Algorithms to Live By (2016)
- Systems vs goals Three questions
  The goal sets the direction; the routine gets you there.
  
  James Clear, Atomic Habits (2018)
Know your part3 lenses
- Circle of control Three rings
  Sort what you control from what you can only influence or must accept.
  
  Epictetus, the Enchiridion, and Stephen Covey, The 7 Habits of Highly Effective People (1989)
- Identity-based habits Three questions
  Choose by who you want to become, not only by the result.
  
  James Clear, Atomic Habits (2018)
- Attention economy Three questions
  Your attention is limited, and plenty of things are built to take it.
  
  Herbert Simon (1971)

Keep going

Four of the lenses have a longer guide of their own, and the dungeon tests how fast you spot which model a problem needs.

Inversion guide Second-order thinking guide The value equation The dichotomy of control Play Mental Model Dungeon

Where they agree, where they split, and what you decided

Once you know how it turned out, it is hard to remember what you knew when you chose. Writing it down first is the only fair way to look back.

Where the lenses point · Should I take the job offer in Lisbon?

The lenses disagree

2 for and 1 against. Before you decide, look again at Inversion.

For it2

Second-order thinking Staying is easy now and costly later. Going is the other way round.

Opportunity cost The best alternative is a maybe. This one is a yes on the table.

Unsure1

Margin of safety Two of the three buffers are short. Ask for a later start.

Against it1

Inversion Three ways it fails, and I can guard against only one of them today. The smaller side

When they split, look at the smaller side

Lenses that look at different things and still agree are worth more than one lens that is very sure. When they split, the desk names the lenses on the smaller side, because that is usually where the unresolved part is.

The decision

What I decided: Say yes if they move the start to March and pay for the move
What I expect to happen: A hard first winter, then a role I am glad I took
How sure I am: 65%
First step: Email the recruiter about the start date and the move
Review on: Thu, January 28, 2027in 90 days

Confidence you can check

Each decision records how sure you were. Once three are reviewed, the desk compares the confidence you wrote with how things turned out, so you learn whether your 80 percent means 80 percent.

Review

On Thu, January 28, 2027, the desk asks two things.

How it turned out: Better than I expectedAbout as I expectedWorse than I expected
Was it a good call?: A good call, given what I knewPartlyWith the same facts, I would decide differently

Two questions, kept apart

Annie Duke calls judging a decision by its outcome resulting. A good call can turn out badly, and a bad one can get lucky, so the review asks how it turned out and, separately, whether it was a good call given what you knew.

More than one lens, on purpose

One model makes every problem look like the one it was built for. A few, used together, catch what each would miss.

A latticework, not a hammer

Charlie Munger’s advice was to carry a few dozen models from different fields and hang your experience on them. “To the man with only a hammer, every problem looks pretty much like a nail.” The desk keeps to twelve, so each one is something you would actually reach for.

More lenses for a one-way door

Jeff Bezos sorts decisions into two-way doors, which you can walk back through, and one-way doors, which you cannot. The desk asks for more lenses and a later review the harder a decision is to undo.

How many lenses a decision deserves

Easy to undo2 lenses · review in 30 days
A two-way door. If it goes wrong you can walk back through, so decide quickly.
Costly to undo3 lenses · review in 60 days
You could reverse it, but it would cost time, money or goodwill.
Can't undo4 lenses · review in 90 days
A one-way door. Slow down and look through more lenses before you go.

Start from an example

Each opens the desk with its lenses already worked through, ready to change into your own.

Opening an example adds it to your shelf of decisions. The ones you already had stay where they were, and the new one is one Undo away.

What it is built on

Six sources, used for what they show and no more. Each lens names its own in the desk.

A latticework of models Charlie Munger, “A Lesson on Elementary, Worldly Wisdom”, USC Business School, 1994
Carry the big ideas from many disciplines and use them together, because a single model distorts what you see. The desk’s twelve come from economics, investing, psychology and philosophy.
The pre-mortem Gary Klein, “Performing a Project Premortem”, Harvard Business Review, September 2007
Imagine the project has failed and write down why. Assuming the failure, rather than asking what might go wrong, makes it easier to name real risks. The inversion lens works this way.
Decisions as bets Annie Duke, Thinking in Bets, 2018
Treat a decision as a bet on one of several futures, and judge it by what you knew rather than by how it turned out. The three cases and the two review questions come from here.
One-way and two-way doors Jeff Bezos, letter to Amazon shareholders, 2015
Some decisions are consequential and nearly irreversible, and deserve care; most can be changed, and should be made quickly. The desk’s three stakes and the lenses each asks for follow that split.
The margin of safety Benjamin Graham, The Intelligent Investor, 1949
Leave room between what a plan needs and what you have, so a bad turn does not sink it. The buffers worksheet sets each need against what you have and shows the gap.
The pull of money already spent H. R. Arkes and C. Blumer, Organizational Behavior and Human Decision Processes 35, 1985
In their experiments, people who had paid more for something kept using it, even when an alternative was better. The sunk cost lens asks you to decide from here forward.

Questions

What is a mental model?

A mental model is a simplified idea of how something works that you can carry from one problem to another. Opportunity cost, second-order effects and the margin of safety are examples. Charlie Munger argued that a few dozen of them, drawn from different fields and used together, beat expertise in any one.
How does the Thinking Desk work?

You write the decision as a question, say what kind of decision it is and how hard it would be to undo. The desk suggests lenses that fit. Each lens is a short worksheet, a pre-mortem, a consequence ladder, three cases with odds and so on, and it ends with which way that lens points. When you are ready, you write down what you decided, what you expect and how sure you are, and set a date to look back.
Which model should I use?

Start with the suggestions. They come from the kind of decision, the words in your question and how hard it is to undo, and each says why it was picked. A decision you cannot reverse always gets inversion and the margin of safety near the top. You can add any of the twelve at any time.
How many lenses does one decision need?

It depends on how hard it is to undo. Jeff Bezos split decisions into two-way doors, which you can walk back through, and one-way doors, which you cannot. The desk asks for two lenses on a decision that is easy to undo, three on one that is costly to undo and four on one you cannot undo, and says when you have fewer.
What does it mean when the lenses disagree?

That the decision is not as clear as it felt, which is the point of using more than one. When lenses that look at different things point the same way, you can be more confident. When they split, the desk names the lenses on the smaller side, because that is where the thing you have not resolved usually is.
Why write down my confidence and a review date?

Because memory rewrites decisions once the outcome is known. Annie Duke calls judging a decision by how it turned out resulting. Writing down what you expected and how sure you were lets you look back fairly. The review asks two separate questions, how it turned out and whether it was a good call given what you knew, and over time the desk compares your confidence with your results.
Is this advice?

No. The desk asks questions and keeps your answers in order; it does not know your situation and does not tell you what to do. For a decision with legal, medical or financial consequences, talk to someone qualified as well.
Where are my decisions saved?

In this browser, as you type, and nowhere else unless you are signed in, in which case they are also saved to your account so they follow you to another device. You can copy a decision as text or download it as Markdown.

Related tools and books

Decision Filter Decision Compare Matrix Decision Terrain Thinking in Bets Algorithms to Live By The Art of Thinking Clearly

Give the next decision a second look

Your decisions are saved in this browser as you go, and to your account if you are signed in. Nothing you write is shown to anyone else.

Open the desk

Keep this going in the app.

HourLife turns these ideas into a daily system — goals, habits, focus, and reflection in one private place.

See everything HourLife does