How to Keep a Decision Journal (With Template)

What is a decision journal?

A decision journal is a record of your important decisions, written at the time you make them, before you know how they turn out. Each entry captures what you decided, what you expected, and how sure you were. Later, you compare those expectations with what actually happened.

It is a simple idea with an unusual property: it is one of the few ways to get honest feedback on your own judgment. Without a record, your memory of past decisions is rewritten by how they turned out. With one, you can see what you actually thought.

The practice is common among investors, professional forecasters and poker players, all fields where luck and skill are tangled together and you need a way to tell them apart.

Why keep a decision journal?

A decision journal protects you from three problems that make experience a poor teacher.

1. Hindsight bias. After an outcome is known, people tend to believe they saw it coming. Baruch Fischhoff’s research in the 1970s showed that once people learn an outcome, they overestimate how predictable it was. Hindsight bias means you can’t trust your memory of what you expected. A journal entry written before the outcome can’t be edited by it.

2. Judging decisions by outcomes. A good decision can turn out badly because of bad luck, and a bad one can work out by chance. If you only look at results, you learn the wrong lessons. The journal lets you ask a better question: given what I knew, was the reasoning sound?

3. Overconfidence. Most people are more confident than their track record supports. That is overconfidence bias, and you can’t fix it without data. If you write “80% confident” on twenty decisions and only half work out, you have learned something no book could tell you. Research on forecasting, such as Philip Tetlock’s work summarized in Superforecasting (2015, with Dan Gardner), suggests that keeping score is central to getting better at predictions.

There is a quieter benefit too: writing an entry forces you to think. It is hard to write “I chose this because…” without noticing when the “because” is thin.

What to write in each entry

Each entry should be quick to write (10 to 15 minutes) and complete enough to review honestly a year later. These are the fields that matter:

Two fields deserve extra care.

The prediction. Make it specific enough that future you can’t wriggle out of it. “This hire will work out” is vague. “By June, she will own the onboarding project without my help” is checkable.

The confidence %. Use a number, not a word. “Pretty sure” can mean 60% or 95%. Probabilistic thinking starts with the habit of putting numbers on beliefs, even rough ones.

Decision journal template

Copy this into Notion, Obsidian, a notes app or a paper notebook. One entry per decision.

# Decision: [one-line summary]

- **Date:** YYYY-MM-DD
- **Area:** career | money | relationships | business | health | other
- **Decision Weight Score:** _ / 10  (Reversibility _ × Impact _ × Uncertainty _)

## Options
1. 
2. 
3. 

## What I chose and why
- 

## Models used
- 
- 
- 

## Bias Radar check
- Biases I noticed pushing me:

## Prediction
- What I expect to happen, specifically:
- **Confidence %:** 
- What would prove me wrong:

## State of mind
- Energy / stress / time pressure:

## Review date: YYYY-MM-DD

---

## Review (fill in on the review date)
- What actually happened:
- Was the prediction right? 
- Was my confidence too high, too low or about right?
- Luck or judgment (or both)?
- What I would change in the process:

To calculate the Decision Weight Score, rate Reversibility (1 easy to undo, 2 costly to undo, 3 permanent), Impact (1 small, 2 meaningful, 3 life- or business-changing) and Uncertainty (1 mostly known, 2 some unknowns, 3 mostly unknown). DWS = round(R × I × U ÷ 27 × 10), minimum 1. The DWS calculator does it for you.

How to keep a decision journal: step by step

1. Decide what qualifies. Start with every decision that scores a DWS of 7 or more. Add a few medium ones (4 to 6) you’d like to get better at. Don’t journal what to eat for lunch.
2. Write before you act. The entry is only valuable if it’s written before the outcome. If you’ve already acted, write it the same day, before results come in.
3. Fill in every field, briefly. Two lines per field is plenty. A finished short entry is worth more than an unfinished long one.
4. Run a quick stress-test while you write. Writing the “Models used” and “Bias Radar check” fields is a light version of the Stress-test step of the Mentarium Decision Loop. A one-paragraph pre-mortem (“it’s a year later and this failed because…”) fits well here.
5. Put the review date in your calendar. This is the step that most often gets skipped. The entry is only half the habit.
6. Review honestly. On the date, fill in the review section before you reread your reasoning. Then compare.
7. Do a quarterly pass. Every three months, read all entries together and look for patterns.

How to review your decision journal

Single reviews tell you about single decisions. The real value shows up across many entries. In a quarterly review, look for:

• Calibration. Group decisions by confidence. Of the ones you rated around 80%, did about 80% go as predicted? If far fewer did, you are overconfident in that range.
• Areas of strength and weakness. Maybe your business calls are well calibrated and your hiring calls are not. That tells you where to slow down or ask for help.
• Recurring biases. If the same bias keeps showing up in your Bias Radar notes, that is your personal blind spot.
• Models that help. Which mental models show up in your best decisions? Which ones do you list but never really use?
• State of mind. Were your worst calls made when tired, rushed or upset? Many people find a pattern here.

Be fair to your past self. Judge each decision on what you knew and could reasonably have found out at the time, not on what you know now.

Common mistakes

• Writing after the outcome. It turns the journal into a diary of rationalizations.
• Vague predictions. If you can’t be clearly wrong, you can’t learn.
• Skipping the confidence %. Without it, you can’t measure calibration.
• Journaling everything. You’ll burn out in two weeks. Keep it to decisions that matter.
• Never reviewing. A journal you don’t review is just a notebook.
• Being harsh with yourself. The goal is better judgment, not guilt. Bad outcomes from good decisions are part of life.

Making it stick

The hardest part of a decision journal is keeping it going past the first month. A few things help:

• Use a template, so you never face a blank page.
• Keep it where you already work, so writing an entry takes no setup.
• Automate review reminders. A database with a review date field can show you what’s due this week.
• Pair it with a process. The journal works best as the “Decide” and “Review” steps of a full decision-making process, not as a standalone chore.

The Mentarium Decision Journal is a ready-made decision journal for Notion or Obsidian, built around the same fields. You can find it in the shop. If you’d rather have someone walk you through each entry, the free Mentarium Decision Coach runs the Decision Loop inside Claude or ChatGPT and helps you write the prediction and confidence at the end. More free tools are on the tools page.

The Mentarium take

Most people get more experienced with age without getting much better at deciding. The reason is simple: experience only teaches when you can see the gap between what you expected and what happened, and memory quietly closes that gap for you.

A decision journal is the cheapest fix we know. It doesn’t require talent, special software or a lot of time. It requires one honest habit: writing down what you think will happen, and how sure you are, before you find out.

Our advice is to start small. Journal the next three decisions with a DWS of 7 or more, set the review dates, and do one quarterly review. If those three reviews don’t teach you something about your own judgment, stop. We’d bet they will.