The three kinds, compared
| Kind | Example | Problem |
|---|---|---|
| Outcome | "Get a 2:1 this year" | Not under your direct control; doesn't say what to do today. |
| Time | "Study three hours a day" | Rewards presence, not work. Satisfied by three hours of highlighting. |
| Output | "Answer 20 practice questions and mark them" | None serious. This is the one to use. |
Outcome goals aren't useless — they set the direction and are worth writing down once a term. They just can't be the thing you check against daily, because nothing you do today visibly moves them and the gap between effort and feedback is where motivation dies.
What makes an output goal good
- Countable. You can say yes or no at the end of the session with no interpretation.
- Produced, not consumed. "Read chapter 6" is consumption and can be faked by moving your eyes. "Write 10 questions from chapter 6" cannot.
- Sized for a bad day. Set it at what you'd manage when tired, not at your best day. A goal you hit 90% of the time builds momentum; one you hit 40% of the time teaches you that goals are decorative.
- Attached to a time and a place, not just a day. "After Wednesday's lecture, in the library" is a plan; "on Wednesday" is an aspiration.
- Failable visibly. If you can't tell whether you hit it, it isn't a goal.
Converting the goals you have
| What you'd normally write | The output version |
|---|---|
| Revise pharmacology | 20 cards made from lecture 8, then answer 15 mixed questions and mark them. |
| Work on the essay | Write the thesis sentence and three paragraph claims. 300 words. |
| Do some past papers | One timed section, marked against the scheme, losses categorised. |
| Catch up on lectures | Convert weeks 4 and 5 slides into questions. Two lectures, not the backlog. |
| Learn the anatomy | Label the diagram from memory three times, correctly, without looking. |
| Study for four hours | Clear the review queue and complete one past-paper section. |
Three levels, and only one you check daily
- 1
One term-level outcome, written once
"Pass everything with at least a 2:1 in the two modules that matter for next year." Write it, put it away, look at it at the weekly review if at all. Its job is direction, not motivation.
- 2
Three weekly outcomes, chosen every week
Concrete enough to check on Friday: "every week-6 lecture converted to questions", "one timed past paper marked", "essay plan finished". Three, not ten.
- 3
One daily output, decided the night before
The single thing that makes today count. Decided in advance, because deciding at the moment of sitting down is expensive and gets negotiated away.
- 4
Check daily, adjust weekly, ignore the term goal
The daily goal is the only one you interact with. If you're missing it three days running, the goal is too big — halve it rather than concluding you're undisciplined.
- 5
Record hits and misses somewhere visible
Not for gamification — for calibration. After two weeks you'll know your real capacity, which is almost always lower than your planning assumed and higher than your worst weeks suggest.
Set the goal at bad-day size
The commonest goal-setting error is calibrating to your best day. You had a great Sunday, did four hours, and set that as the daily target. Now every ordinary day is a failure, and after a fortnight of failing you stop setting goals — which is worse than having set a small one.
The rule that works: set the goal at what you'd do on a bad day, and let good days overshoot. A goal hit nine times in ten builds a habit; the overshoot on good days gets you the volume anyway. This is also why streak-based systems are risky for anyone with irregular capacity — see the resumption argument in how to study with ADHD.
Goals for things that aren't countable
Some work resists counting — understanding a difficult concept, thinking through an argument, reading something dense. Output goals still apply, they just measure a different product.
- Explain it out loud without notes, recorded. Countable: you either could or couldn't.
- Write the three sentences that state the argument. A product, not a duration.
- Produce one worked example of your own, not from the book.
- List the three questions you still can't answer. Genuine progress on hard material often looks like better questions rather than more answers.
When to change a goal
Missing a goal three times running is data about the goal, not about you. Halve it. Missing it once is noise and needs no response at all — the instinct to compensate by doubling tomorrow is what turns one missed day into an abandoned system.
And revise the whole set at the weekly review, not mid-week. Mid-week goal changes are almost always negotiations with a version of you that doesn't want to work right now, and they compound. Once a week, calmly, with the evidence of what actually happened.