Critical Thinking and Better Decisions

Smart people make poor decisions all the time, usually not because they lack intelligence but because they never checked what they were assuming. Critical thinking is a set of small habits that catch those gaps before they get expensive.

What critical thinking is, and why it matters

Critical thinking means examining claims and reasoning before accepting them, including your own. In practice: asking what the evidence is, what else could explain it, and what would change your mind. It matters at work because most decisions are made with incomplete information, under time pressure, and with opinions presented as facts.

Separate facts, assumptions and interpretations

Much confusion comes from mixing three different things. A fact is something you can verify. An assumption is something you are treating as true without checking. An interpretation is the meaning you attach to facts. Sorting your notes into these three columns is one of the cheapest thinking upgrades there is.

FACTS (verified):
- Signups: 420 -> 310 week over week.
- New form shipped Tuesday.

ASSUMPTIONS (to check):
- Traffic volume unchanged.
- The form works on mobile browsers.

INTERPRETATIONS (hypotheses):
- New form is too long.
- A marketing campaign ended.

Next step: check traffic by source, then mobile error rate.

Common thinking traps

Psychologists have catalogued many cognitive biases. Two are worth knowing well because they appear constantly in work. You cannot fully remove them, but naming them helps you notice them.

Confirmation bias

Confirmation bias is the tendency to look for, notice and remember information that supports what you already believe, and to discount what does not. Example: you suspect a bug is in the payment module, so you read the logs hunting for payment errors and overlook an error in the inventory service that appears right before them. A countermeasure is to ask 'What would I expect to see if I were wrong?' and then actively look for it.

Sunk cost fallacy

The sunk cost fallacy is continuing something because of what you have already invested, instead of what it will cost and return from now on. Example: a team has spent three months on a feature that testing shows users do not want, and keeps going because 'we have come too far'. The past effort is gone either way. The useful question is: knowing what we know today, would we start this now? If not, stopping may be the right call, though there can be legitimate reasons to continue, such as contract commitments or learning value.

Reversible vs irreversible decisions

Not every decision deserves the same effort. A useful sorting question is: if this turns out wrong, how hard is it to undo? Jeff Bezos has described this idea as one-way and two-way doors. Reversible decisions can be made quickly by a person or small group, because the cost of error is low and you learn from trying. Irreversible or expensive-to-reverse decisions deserve more evidence, more voices and more time.

Pre-mortem

The pre-mortem, popularised by psychologist Gary Klein, flips the usual review. Before starting, you imagine that the project has already failed and ask what went wrong. Because the failure is treated as certain, people feel freer to name risks they would otherwise keep quiet about.

Setup (2 min):
"Imagine it's three months from now and this launch has failed badly. Take 5 minutes and write down every reason you can think of why."

Collect (5 min):
Each person adds reasons, silently, on sticky notes or a shared doc.

Group and rank (5 min):
Cluster similar reasons. Vote on the 3 most likely or most damaging.

Act (3 min):
For each top risk: one thing we do now to prevent it, or one early warning sign to watch.

Example output:
Risk: Data migration takes longer than planned. Action: run a trial migration on a copy this week. Warning sign: trial takes more than 4 hours.

Write the decision down

A short written record forces clarity, helps others understand why, and protects you from hindsight rewriting later ('I always knew that would happen'). Engineering teams often call this a decision record, but it works for any decision.

Decision: What we decided, in one sentence.
Date and decider:
Context: What problem or situation prompted it?
Options considered: A, B, C, with one line each.
Reasoning: Why this one? What facts, and what assumptions?
Trade-offs we accept:
Reversibility: Easy / hard to undo? How would we undo it?
Review date: When will we check if it was right, and what will we look at?

Common mistakes

• Treating assumptions as facts because everyone repeats them.
• Searching for evidence only for your favorite explanation.
• Spending a week on a decision that could be undone in an hour, or five minutes on one that cannot be.
• Continuing a plan because of past effort instead of future value.
• Judging a decision purely by its outcome. A good decision can have a bad result, and a lucky one a good result. Judge by the information and reasoning available at the time.
• Keeping decisions in meetings and heads, so they are re-argued repeatedly.

How this shows up in interviews and on the job

Interviewers ask 'Tell me about a difficult decision you made' or 'Tell me about a time you changed your mind', and they listen for how you handled uncertainty: what you knew, what you assumed, who you consulted and how you reviewed the result. Good answers acknowledge what you got wrong. Use the STAR method to structure them. Many 'common interview mistakes' are really thinking mistakes, such as assuming what the interviewer wants without asking; see Common Interview Mistakes to Avoid. On the job, being the person who writes down the assumptions and runs a quick pre-mortem makes decisions better and makes you noticeably more senior.

How to practice

Pick one pending decision, large or small. Write the three-column fact, assumption and interpretation note, then do a five-minute solo pre-mortem, then fill in the decision record. Revisit it on the review date and note what you would change in your process.