Performance Goals

A performance goal is a dated checkpoint on a process still under control, read in time to change the trajectory. The outcome is later, and luck claims whatever is left. Feedback that arrives only at the result arrives too late to use.

Why hours and pages mislead

A driver chasing a one-minute lap reads the lap mid-flight. A hundred metres before a key corner the speed must show 80 km/h; a hundred metres past it the clock must sit at 43 seconds. Miss either marker and the lap target is already gone, with most of the lap still to drive. The early signal creates room to self-correct while correction is cheap. Without it, weeks of effort can run down a wrong path, and the first honest feedback lands the night before the exam.

Activity metrics count the studying process: hours at the desk, chapters covered, flashcard counts, pages of notes. Learning is a change in what can be retrieved and used, and it moves independently. Five hours with a weak technique can produce less mastery than two hours with a strong one. A checkpoint earns its place only by correlating with the outcome. “Cover this much material by this date” can be maxed out while the goal dies. If the metric can be maxed while the goal dies, it is decoration.

The success avatar

The checkpoints attach to attributes, not to coverage. Imagine a person certain to achieve the goal. Fully hypothetical is fine. The test: if this person could not pull it off, nobody could. List the attributes that make their success near-inevitable. Those are what the checkpoints should measure.

Start with the attributes people skip: time management; focus and procrastination control; learning efficiency (usable learning per unit of time); learning consistency (how regularly the work happens — a separate variable from efficiency); stress management; resilience under discomfort; sleep quality; support network. Add goal-specific ones — communication, networking, the size of the network — where the goal demands them. Sleep as a performance attribute is real and is not a protocol on this page.

Resilience gates skill acquisition. Every new technique feels uncomfortable at first, and retreating to the familiar method at that exact moment blocks the acquisition outright. Stress undercuts the whole profile no matter how strong the other attributes are.

Rate against each attribute, then pick two or three. The gap is the real plan. Deficits across many attributes are normal — more headroom. Seven or eight simultaneous targets overload the plate and tend to fail together. Distrust the rating “this doesn’t apply to me.” Brains assess their own abilities badly. The exemption reflex is strongest in high performers, the group most likely to rush fundamentals and stall later. Rating honestly costs nothing and prevents the expensive version of finding out. The honest self-rating the avatar comparison depends on is the same muscle a Skills Audit trains on technique.

Measure the person you are building

The checkpoints attach to the person being built, not to the coverage being logged. Early targets are supposed to be almost certain because the work is still building the person, not proving the outcome. The process is what sits under control. The outcome is a symptom.

Thirty days, fourteen days, the golden rule

Write baseline → target for each chosen attribute. Both ends in concrete numbers. Specific and measurable beats “do your best.”

A worked time-management line: today, scheduling three days a week, following the schedule about 20%, clearing a third of the daily task list. Thirty days out, scheduling four to five days a week, following it about 60%. Clearly better, nowhere near perfect. That is the right size.

Repeat at fourteen days, smaller. The fourteen-day target should look unimpressive on paper. Its job is momentum: a working improvement system and a visible first gain, so the thirty-day review shows a real step rather than a stall.

The golden rule: set each target so the worst case still hits it and the best case far exceeds it. A missed early checkpoint kills momentum exactly when momentum is the product. A guaranteed one stacks a Marginal Gains cycle — each fourteen- or thirty-day step is one stacked gain — and buys capacity for the next cycle to aim higher. Instrumenting the process (baselines, dates, a fourteen-day review) is the cost; the benefit is correction while it is still cheap.

Without dated baselines the work collapses into the daily grind: weeks of the same activity, no visible progress, rising anxiety — and an outcome that ends up resting on crossed fingers because the process was never instrumented. This page is the forward checkpoint layer. Reverse Goal Setting maps the same attribute gaps backward from the outcome into a full capability plan.

What this is not, and when to quit

On a well-learned, simple task, a difficult specific goal beats an easy one. These early targets are not that kind of goal. They are for building attributes on novel, complex work, where a difficult outcome target can block the learning the attributes require.

The avatar is unvalidated. Fourteen and thirty days are this system’s own teaching defaults, not measured optima. The two-or-three cap is house, shared with the sibling pages, not trialled here. Activity-metric direction is supported; the five-hours-can-lose-to-two comparison is an illustration, not a finding.

If every checkpoint is an activity metric that can be maxed while the goal dies, the page is being used as decoration — change the metric, do not add more dates. If seven attributes are all in play, the two-or-three cap has failed and the next cycle will fail with it. If the fourteen-day mark was set to impress, it is the wrong rule.

A dated baseline and a dated target on two or three attributes, written so a worst honest week still hits the near one, is the checkable form.

The outcome as a symptom

The outcome follows as a symptom of a process good enough and consistent enough, with luck claiming whatever is left. The outcome is not controlled directly. Without the attributes, intermediate goals will not save the work. With them, the goals arrive as a side effect.

Open Questions

When an attribute has no honest number — resilience, a support network — what is the smallest proxy that still correlates with the outcome rather than decorating it?

Sources

• Locke, E. A. & Latham, G. P. 2002. “Building a practically useful theory of goal setting and task motivation.” American Psychologist 57(9), 705–717. Specific, measurable goals; feedback as a moderator; control over direction, effort, persistence, and strategy.
• Kingston, K. & Hardy, L. 1997. Journal of Sports Sciences 15, 187–196. Process-goal training improved self-efficacy and concentration more than performance-standard goals.
• Seijts, G. H. & Latham, G. P. 2005. “Learning versus performance goals.” Academy of Management Executive 19(1). On complex or novel tasks a difficult outcome goal can inhibit learning; a learning goal (strategies, attributes) is the fit.
• Plant, E. A., Ericsson, K. A., Hill, L. & Asberg, K. 2005. Contemporary Educational Psychology. Study hours predicted GPA poorly; deliberate high-quality study predicted it.
• Roediger, H. L. & Karpicke, J. D. 2006. Testing effect: more exposure time is not more learning.
• Goodhart, C. 1975; Strathern, M. 1997. When a measure becomes a target it ceases to be a good measure.
• Kruger, J. & Dunning, D. 1999. Journal of Personality and Social Psychology 77(6). Self-assessment is systematically flawed.
• Dunning, D., Heath, C. & Suls, J. M. 2004. Psychological Science in the Public Interest. Same direction.
• Schwabe, L. & Wolf, O. T. 2009. Journal of Neuroscience. Acute stress shifts control toward habit.