Why decisions go wrong and how to check your judgment

To check a decision, separate facts, assumptions and personal preferences. Then ask what could change your initial choice. Feeling certain and having a detailed explanation do not establish that the underlying information is correct or that you considered suitable alternatives.

Imagine choosing how to learn a new work skill. You like an instructor, a friend recommends the course and the first offer looks attractive. Those are reasons to notice the option. The timetable, practical assignments and feedback arrangements may still be unknown. Here is what to examine before enrolling.

Check the experience behind your intuition

In their 2009 paper, Kahneman and Klein connect the likely quality of expert intuition with a predictable environment and opportunities to learn its regularities. Subjective confidence alone is not a reliable indicator of judgment accuracy.

Ask yourself: “Have I faced a similar task?” and “Did I later learn how accurate my expectation was?” Years of attending classes do not automatically tell you whether a particular programme suits a new goal. The relevant question is what you learned to distinguish through that experience.

Record your intuitive preference as a starting point. Name its basis: a clear structure, a useful previous experience or simply a familiar title. You do not have to reject a first impression to examine it more carefully. The note gives you something specific to check.

Notice the first estimate and the memorable example

Tversky and Kahneman described heuristics in judgment under uncertainty, including availability and adjustment from an anchor. An easily recalled example can influence an estimate of frequency, while an initial number can influence a later numerical judgment. This does not mean every quick decision is wrong.

In our example, a friend's excellent result does not show what participants with your starting skills usually achieve. Compare the first quoted price with the content, duration and available alternatives. A large discount does not by itself explain whether you need the programme.

You do not need to name every possible bias. Notice which piece of information is standing in for the question you actually need to answer: does the option fit your task and constraints?

Distinguish a feeling, a preference and a prediction

“I enjoy learning in a group” expresses a preference. “I will definitely finish this course in a group” predicts an outcome. “I feel nervous about meeting people” describes a feeling. You can take all three into account while treating them differently.

A preference helps identify a comfortable format. A prediction needs supporting information: the schedule, independent workload and available support. A feeling can be discussed or considered in preparation without treating it automatically as evidence of future success or failure.

If your conclusion is “Everything will work” or “Nothing will work”, use the questions for a more precise assessment. The aim is to find something you can check, rather than make yourself experience a prescribed emotion.

Look for information that could change your mind

In two experiments by Lord, Lepper and Preston in 1984, considering the opposite corrected biased social judgments more effectively than a general request to be fair. These findings concern particular tasks, not a guarantee that the technique removes all decision errors.

For a course you already like, ask: “What would make this unsuitable?” Perhaps mandatory sessions clash with work, or the promised feedback does not include reviewing your assignments. Seek an answer to that question as well as positive reviews.

Apply the same test to the alternative. If a marketing promise is sufficient for your favourite option, while every detail of another option needs independent confirmation, the comparison is using different standards.

Write a short note before the outcome is known

Record your choice, confirmed reasons, the main unknown and a condition for reconsidering. For example: “I prefer the group with a workable timetable. I will confirm the feedback format before paying. If assignment review is unavailable, I will return to the comparison.”

Later, compare the outcome with what you knew at the time. A good result does not validate every assumption, and a disappointing one does not establish that all preparation was pointless. Record which question would have been useful earlier.

For the comparison itself, use the criteria-based decision worksheet. If circumstances change substantially after choosing, follow the guide to revisiting a plan.