Career decisions with incomplete information

Career advice generally assumes that with enough reflection the right answer becomes apparent. The more accurate framing is that the information required does not exist yet.

Why introspection is limited here

Research on affective forecasting consistently finds that people predict their future emotional responses poorly.

Specifically, they overestimate both the intensity and the duration of reactions to future events — a pattern usually called impact bias.

Which means predictions about how a job will feel are unreliable in a systematic direction, and confidence in the prediction is not a signal of accuracy.

The practical implication is uncomfortable: thinking harder about which option you would prefer does not reliably improve the answer.

What does improve it

Talking to people doing the work. Substantially more informative than reflection, and the most reliably underused step.

The useful questions are about the texture of the work rather than the summary: what does a normal Tuesday contain, what is the worst part, what surprised you, what do people who leave say.

Sampling the work. A project, a secondment, freelance work, a side arrangement.

Direct experience resolves in weeks what speculation cannot resolve at all.

Looking at what you have actually chosen. Revealed preference is more reliable than stated preference.

What have you spent time on when nobody required it? What did you volunteer for? That record is evidence about your preferences that does not depend on prediction.

The variables that predict satisfaction

Job satisfaction research does not point where people expect.

Consistently associated with satisfaction: autonomy over how work is done; the ability to see the effect of your work; variety; supportive relationships with colleagues and manager; and a reasonable match between demands and capability.

More weakly associated than people assume: salary above a level covering needs comfortably, prestige, and — notably — the subject matter of the work itself.

That last finding surprises people. Evidence on passion-following as career advice is weaker than its popularity suggests, and interest frequently develops from becoming good at something rather than preceding it.

Which suggests that assessing a role on autonomy, colleagues and feedback loops predicts satisfaction better than assessing it on how interesting the topic sounds.

Reversibility

The most useful frame for the decision itself.

Decisions differ in how easily they can be undone. A reversible decision made quickly costs little if wrong. An irreversible one deserves the deliberation.

Most career moves are more reversible than they feel. Moving industry, taking a step sideways, or trying something for a year are recoverable in a career of forty years.

Genuinely irreversible choices are rarer: substantial debt for a qualification, relocating a family, or leaving a field with a closed entry point.

Applying the deliberation appropriate to the second category to decisions in the first is a common and expensive error.

The sunk cost trap

Particularly strong in careers because the investment is years rather than money.

Training, qualifications and accumulated experience feel like reasons to continue, and they are not reasons — they are already spent.

The relevant question is what the best available option is from here, and past investment is only relevant to the extent that it changes what is available.

Frequently it changes it less than people believe, because skills transfer more than job titles suggest.

A workable process

Establish what is actually unsatisfactory. "I want a change" is not actionable; "I have no autonomy and my manager is unreasonable" is, and might be solved without changing employer.

Generate more than two options. Binary framing — stay or leave — is almost always false, and the third option is frequently better than both.

Test cheaply where you can. A conversation, a project, a short course.

Decide on the reversible ones quickly.

And set a review date rather than expecting certainty. A decision made now with a check in six months is more robust than one made after six months of deliberation with no review at all.

General information about work and wellbeing, not career or psychological advice. Consult a qualified professional about your own circumstances.