Thirty-five years of research says most goals are too easy
Edwin Locke and Gary Latham spent thirty-five years chasing a single question: what actually makes a goal produce real performance, and what makes another goal go nowhere?
The goal-setting theory that came out of that work is the most heavily validated framework in industrial-organizational psychology. Hundreds of studies. Multiple meta-analyses. The pattern replicates across sales, manufacturing, education, sports, and healthcare.
The headline finding is blunt: specific, difficult goals produce performance 200 to 300% higher than "just do your best" goals.
Not 10% higher. Not 30% higher. Two to three times higher.
The two things that actually make a goal work
Locke and Latham found two separate reasons specific, difficult goals beat vague ones.
Direction. A specific goal points your attention and effort at exactly what matters, and away from everything that doesn't. "Increase sales by 23% this quarter" tells your brain precisely what to optimize for. "Do your best" tells it nothing.
Effort calibration. Difficulty sets how hard you actually try. People tend to put in effort roughly proportional to how hard a goal feels. A hard goal pulls more out of you. An easy one lets you coast.
That difficulty effect climbs in a straight line, right up to the edge of your ability. Past that point, extra difficulty stops helping, but the useful range is wider than most people assume.
The two things that have to be true first
The effect only shows up when two conditions are met.
Commitment. You have to actually believe the goal matters and is at least possible. A goal handed down from above only works as well as you've internalized it. Goals you set yourself, or genuinely helped shape, produce bigger effects than goals someone just assigned to you.
Feedback. You have to be able to see how you're tracking. Without feedback, even a great specific, difficult goal fails, because there's no way to calibrate your effort against it. Sales numbers, workout PRs, word counts, deployment frequency: the metric changes by domain, but the need for one doesn't.
Take either one away and the difficulty effect disappears. Keep both in place, and the 2 to 3x performance gain holds up reliably.
Why SMART quietly loses the effect
SMART (Specific, Measurable, Achievable, Relevant, Time-bound) has been taught in business schools for decades. It's also, in a real sense, a watered-down version of Locke and Latham's work that drops the one variable that matters most.
The problem sits in the A: Achievable. It was meant to stop people from setting unrealistic goals, but in practice it gets read as "comfortable." And a goal that's comfortably achievable with your current capacity is exactly the kind of goal Locke and Latham's research shows underperforms a difficult one.
What the real research actually supports looks more like this:
- Specific: yes.
- Measurable: yes, this is your feedback.
- Difficult: the variable SMART quietly drops.
- Time-bound: yes.
- Committed: the other variable SMART quietly drops.
The better reframe isn't "achievable." It's "difficult enough to demand real effort, with a genuine belief that hitting it is possible."
Where goal-setting goes wrong
The same thirty-five years of research also maps out exactly how goals fail.
Vague goals. "Improve performance" gives you no direction at all.
Easy goals. "Hit 90% of last quarter's number" gives you nothing to calibrate effort against.
Conflicting goals. "Cut costs" and "improve quality" at the same time, with no clarity on the trade-off, splits your effort so you hit neither.
No feedback. "Improve sales" without any regular tracking means your effort has nothing to calibrate against.
Commitment that's all talk. Agreeing to a goal in a meeting isn't the same as actually internalizing it. Without real buy-in, there's no effect to speak of.
Pressure without resources. A hard goal with no extra time, training, or tools produces frustration, then abandonment, and sometimes outright cheating. The Wells Fargo cross-selling scandal is the textbook example of what happens when hard goals meet missing resources.
Where this framework applies, and where it doesn't
Goal-setting theory works extremely well when:
- The path to the goal is clear (make the call, close the deal).
- Effort and outcome are tightly linked.
- The metric is honest.
It works less well when:
- The work is highly creative, where locking onto a direction too early can mislead you.
- You're learning a brand-new skill, where focusing on the process beats focusing on the outcome early on.
- You're innovating, and nobody actually knows the right metric yet.
For learning and innovation, process goals ("complete 20 practice sessions this month") tend to beat outcome goals ("become proficient at X"). That distinction matters more than most goal-setting advice admits.
What this actually looks like in practice
Here's what a working goal looks like, mapped to each piece:
- Specific: "Ship the v2 onboarding flow with 4 redesigned screens, A/B tested, to all new users."
- Measurable: "Improve activation rate from 32% to 45%."
- Difficult: "45% is a stretch. 38% would be comfortable. The goal is 45%."
- Time-bound: "By the end of Q2."
- Committed: "I genuinely believe this matters and is possible."
- Feedback-rich: "Track activation daily, review every Friday."
Notice that the difficulty is deliberate, not accidental. This isn't "the realistic projection." It's a stretch designed to pull effort out of you. The odds of hitting exactly 45% are moderate. The odds of beating 38% are high either way.
What TaskCoach.AI does with this
The Goals system is built to support specific, difficult goals rather than vague aspirations. Every goal carries specific tasks underneath it, a measurable success metric, and a real time horizon. The Goal Review flow walks you through the Locke-Latham criteria before you commit to anything.
The AI coach also pushes back on goals that read as too vague or too easy. Write "get healthier" and it asks for the specific metric and the stretch number. Write "work out 3x a week" when you're already working out 3x a week, and it asks whether 4x or 5x is the actual stretch you're avoiding.
The Habit Momentum chart is the feedback half of the equation. Your weekly review surfaces actual versus target, so the calibration loop keeps running. Without that loop, even a perfectly Locke-Latham-shaped goal would quietly degrade. Feedback is what makes the difficulty effect hold.
The bottom line
Specific, difficult, committed, and backed by feedback: that's the most validated formula in goal-setting science.
SMART drops the difficulty variable and hands you comfortable goals that produce comfortable results.
If your goals don't make you a little nervous, they probably aren't specified correctly. Thirty-five years of Locke and Latham's evidence backs that up.