Goal setting: what the evidence supports and what it does not

Goal setting has one of the larger evidence bases in organisational psychology, and it is applied far beyond the conditions in which it was established.

What the research found

Decades of studies converge on a consistent core: specific, difficult goals produce better performance than vague goals or exhortations to do your best.

The mechanisms proposed are that goals direct attention, mobilise effort, increase persistence, and prompt the development of strategies.

The effect is well replicated and it is genuine.

The conditions attached

Which are stated in the original work and routinely dropped in the popular version.

The person must be committed to the goal. An imposed goal the person does not accept does not produce the effect.

Feedback on progress must be available. Without it, a specific goal is no better than a vague one, because there is no way to adjust.

The task must be within the person's ability or the goal must be a learning goal rather than a performance goal.

This is an important distinction. For a task where the person does not yet know how to succeed, a difficult outcome goal impairs performance — attention goes to the outcome rather than to acquiring the method.

For novel or complex tasks, a learning goal — discover three approaches — outperforms a performance goal.

The task must be reasonably simple and well defined for the largest effects. Complex, ambiguous tasks show weaker results.

The documented downsides

A body of critical work has identified consistent side effects of aggressive goal setting.

Narrowing. Goals direct attention, which means they direct it away from everything unmeasured.

A team with a numeric target reliably optimises for that number, including at the expense of things nobody thought to specify.

Unethical behaviour. Several studies have found that difficult goals with no reward for near-misses increase misreporting.

The corporate cases usually cited — sales targets producing fraudulent accounts, emissions targets producing defeat devices — are extreme versions of an effect that shows up in laboratory settings too.

Reduced intrinsic motivation, particularly where the goal is externally imposed and tied to reward.

Risk-taking, because a goal creates a reference point below which outcomes feel like losses, and people take more risk to avoid losses.

Where SMART goals help and where they do not

The specific-measurable-achievable-relevant-time-bound formulation is a reasonable summary of the specificity finding and it is not the whole of it.

It works well for defined tasks with clear metrics and a known method.

It works poorly for exploratory work, creative work, relationships, health behaviours where the outcome is not controllable, and anything where the right measure is not known in advance.

Applying it to those situations produces goals that are specific and measure the wrong thing, which is worse than a vague intention.

Process goals

The more useful alternative in most personal contexts.

An outcome goal specifies a result: lose a stated amount of weight, get a promotion, run a particular time.

A process goal specifies behaviour: train four times a week, write for thirty minutes daily, have one difficult conversation a month.

Process goals have three advantages. They are within your control. They provide feedback immediately. And they do not fail because of factors you could not influence.

The evidence on implementation intentions — specifying when, where and how — supports this directly, and the effect sizes are respectable.

Knowing when to abandon one

The most neglected part of the topic.

Persistence is treated as unambiguously virtuous, and research on goal disengagement suggests otherwise: the ability to let go of unattainable goals is associated with better wellbeing than persisting with them.

The useful questions are whether the goal still connects to something you want, whether the evidence suggests it is achievable, and what the continued pursuit is costing.

Abandoning a goal that no longer serves is a competence rather than a failure, and it frees resources for one that does.

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