Goals set on outcomes leave you nothing to do on Tuesday

Two ways to write the same ambition

The distinction is between a target expressed as a result — a weight, a revenue figure, a finished manuscript — and one expressed as a behaviour repeated on a schedule. Both can describe the same underlying wish. They differ in what they tell you to do at any given moment, and in what they do to you when things go badly.

An outcome goal is a specification of success and nothing else. It leaves the entire question of what actions produce it unanswered, which is fine when the route is obvious and unhelpful when it is not. A process goal specifies the action and leaves the result uncontrolled, which is uncomfortable and matches how most outcomes actually work.

Control is the argument

Nearly all outcomes are produced jointly by your actions and by conditions you don’t control. Setting the target on the joint product means your assessment of yourself moves with things you cannot influence, and it means that doing everything correctly can still register as failure. Over time that decouples effort from feedback in the worst way.

Setting the target on the action keeps the assessment inside your control. It also produces a cleaner signal: if you did the thing consistently for a season and the result did not follow, that is genuine evidence that the theory connecting the action to the result was wrong, which is information you can act on. When the goal was the result, the same period yields only that you failed.

What the evidence supports, with the usual caveats

Research on goal setting is one of the larger literatures in applied psychology, and the general finding — that specific, challenging goals are associated with better performance than vague encouragement to do your best — has held up across many studies and settings. That much is reasonably solid.

The finer claims are less settled. Whether process goals outperform outcome goals in general, for whom, and under what conditions has produced mixed results, with some work suggesting the combination works better than either alone and some suggesting the advantage depends on how much control the person has over the outcome. Treat the process-goal recommendation as a well-motivated heuristic rather than a demonstrated superiority.

It is also worth noting the documented downsides of aggressive outcome targets in organisations: narrowed focus, neglect of unmeasured aspects of the job, and in some cases distortion of the numbers themselves. Those failures are better established than the fine-grained comparisons between goal types.

Writing a process goal that’s actually checkable

The requirement is that at the end of any given day you can say yes or no without interpretation. "Write more" fails. "Write for forty minutes before opening anything else" passes, because it names a duration, a trigger and an order. The specificity is not pedantry; it’s what makes the goal capable of being kept or broken rather than merely aspired to.

The second requirement is that the level be sustainable at your worst realistic week rather than your best. A process goal calibrated to a good week generates a long series of failures during an ordinary one, and the failure record is what usually ends the attempt. Setting the floor low and exceeding it is structurally more robust than setting it high and missing.

Keeping the outcome in the picture

None of this argues for ignoring results. The outcome is why the process was chosen, and a process that is not producing anything should be revised rather than defended. The role of the outcome is to serve as a periodic check on whether the theory is right, not as the daily scoreboard.

A workable arrangement is to review the outcome on a slow cycle — quarterly, seasonally, whatever matches how fast results could plausibly appear — and to leave the daily attention on the behaviour. Reviewing the outcome too frequently mostly imports noise, since short-run movement in almost any result is dominated by things unconnected to what you did this week.

Where outcome goals are the right tool

When the route is well known and the constraint is simply whether you do it, an outcome target can be perfectly good, and it has the advantage of not prescribing a method that may not suit you. Somebody who knows exactly how to reach a result and just needs to commit doesn’t need the action broken down.

They also work better when the outcome is largely under your control and arrives quickly, since the feedback then behaves like process feedback anyway. The mismatch appears specifically where results are slow, noisy or jointly determined, which unfortunately describes most things people set goals about.