92% of New Year's resolutions fail before February. It's not a motivation problem — it's a design problem.
How you frame your goal largely determines whether you'll achieve it. This isn't an opinion: it's what 35 years of research and over 1,000 studies by psychologists Edwin Locke and Gary Latham demonstrate.
In this article, I'll show you exactly what makes a well-designed goal — and why some popular methods like SMART are actually flawed.
Why "doing your best" doesn't work
When you tell yourself "I'll exercise more" or "I'll eat healthier," you're giving yourself a wide range of acceptable outcomes. Three sessions in January? Technically, that's "more." And your brain validates that compromise without a second thought.
A vague goal produces vague results. It's mechanical.
Locke and Latham demonstrated that 90% of studies confirm the superiority of specific and difficult goals. The performance improvement ranges from 40 to 80% depending on the context — a considerable effect.
Four mechanisms explain why this works:
- Attention — A precise goal directs your focus toward the actions that matter. You naturally filter out what's irrelevant.
- Energy — Difficulty mobilizes effort. You invest more in a challenge than in an easy task.
- Persistence — When you know exactly what you're aiming for, you hold on longer in the face of obstacles.
- Strategy — A clear goal pushes you to find ways to get there, instead of operating on autopilot.
The 3 characteristics of a well-designed goal
1. Specificity
A good goal eliminates ambiguity. You know exactly what you're aiming for, and you'll know exactly whether you've achieved it.
| ❌ Vague | ✅ Specific |
|---|---|
| Get back in shape | Run 5K in under 30 minutes |
| Save more money | Save $400 per month |
| Learn Spanish | Reach B1 level on the DELE exam |
| Eat healthier | 5 home-cooked meals/week with 3 servings of vegetables |
| Advance at work | Get PMP certification and apply to 3 project manager positions |
| Increase sales | Increase quarterly revenue by 15% vs Q3 |
The test: could an outside observer objectively verify whether you've reached your goal? If the answer is no, it's not specific enough.
2. Calibrated difficulty
This is where research contradicts common sense — and some popular methods.
People think a "realistic" and "achievable" goal is preferable. That's wrong. Locke and Latham showed that performance increases linearly with difficulty — up to a certain point.
The sweet spot: the 90th percentile of your capacity.
That's a goal you have about a 10-20% chance of achieving without changing anything, but that becomes realistic with sustained effort.
| Difficulty level | Effect on performance |
|---|---|
| Too easy | Demotivating, minimal effort |
| "Realistic" / comfortable | Average performance |
| Difficult (90th percentile) | Maximum performance |
| Impossible | Discouragement, giving up |
The test: your goal should scare you slightly, but not paralyze you. If you're 100% sure you'll achieve it, it's too easy. If you don't believe in it at all, it's too hard.
Critical point: beyond the 90th percentile, the effect collapses dramatically. The correlation between difficulty and performance drops from 0.82 to 0.11 when the goal is perceived as impossible. That's why "impossible goals" meant to "inspire" generally don't work.
3. Time horizon
A goal without a deadline is wishful thinking. The deadline creates urgency and allows you to measure progress.
| ❌ No time frame | ✅ With time frame |
|---|---|
| Run 5K in 30 min | Run 5K in 30 min within 12 weeks |
| Save $4,800 | Save $400/month for 12 months |
| B1 level in Spanish | Pass the DELE B1 in September |
Which time horizon to choose?
| Goal type | Recommended horizon | Why |
|---|---|---|
| Habit change | 30-90 days | Time to anchor an automatic behavior |
| Concrete project | 3-6 months | Long enough to progress, short enough to maintain urgency |
| Major transformation | 6-12 months | Realistic for a deep change |
| Life vision | 3-5 years | General direction, to be broken into milestones |
The complete formula
A well-designed goal follows this structure:
[Measurable action] + [Precise quantity/threshold] + [Deadline]
Examples:
- "Run 5K in under 30 minutes by March 1st"
- "Save $4,800 ($400/month) by December"
- "Get PMP certification before June"
- "Publish 12 articles (1/week) over the next 3 months"
- "Reach 1,000 newsletter subscribers within 6 months"
How to calibrate difficulty: a practical method
Step 1 — Establish your baseline
Where are you today? Measure concretely.
| Domain | How to measure your baseline |
|---|---|
| Fitness | Time a run, count your max push-ups, note your weight |
| Finances | Calculate your average savings over the last 3 months |
| Productivity | Track your time for a typical week (Toggl, RescueTime) |
| Sales | Current conversion rate, average monthly revenue |
| Writing | Number of articles/words published last month |
| Learning | Take a level test (languages, technical skills) |
Recommended duration: measure over at least 1-2 weeks to get a representative baseline, not just a good or bad day.
Without a baseline, you can't calibrate difficulty. You're navigating blind.
Step 2 — Identify the realistic 90th percentile
| Method | How to do it | Example |
|---|---|---|
| External benchmarks | Look for public statistics, industry standards | "Average beginner 5K time is 35 min, intermediate is 28 min" |
| Experienced peers | Ask someone 2-3 years ahead of you | "How long do you think it would take me to go from 38 to 30 min?" |
| Your best moments | Recall your peak performance | "My best savings was $500/month when I was motivated" |
| Historical progression | Project your learning curve | "I improve ~5% per month, in 6 months I'll be at..." |
| The ×1.5 to ×2 rule | Take your baseline and multiply by 1.5-2 | "I save $200 → target $350-400" |
The problem with SMART
You probably know the SMART acronym: Specific, Measurable, Achievable, Realistic, Time-bound. It's become a standard in business and personal development.
The problem: SMART has never been empirically validated.
A critical review published in Health Psychology Review concludes that there is "virtually no respectable evidence for the quality of the method, just a lot of vague claims."
Worse: the "A" (Achievable) and "R" (Realistic) directly contradict what research shows. Locke and Latham demonstrated that difficult goals — not "realistic" ones — produce the best performance.
A vague goal produces vague results. A "realistic" goal produces average results. Only a specific AND difficult goal produces exceptional results.
SMART is intuitive and easy to remember. But if you want results, rely on what's proven: specific + difficult + time-bound. For a detailed analysis of what works (and what doesn't) in SMART, read our comprehensive guide to SMART goals.
| SMART criterion | Problem |
|---|---|
| Achievable | Encourages goals that are too easy |
| Realistic | Same — limits ambition |
| Measurable | Fine, but redundant with "specific" |
What to avoid: a recap
| Common mistake | Why it doesn't work | Alternative |
|---|---|---|
| "Do my best" | Too many acceptable outcomes | Precise number |
| "Realistic" goal / SMART | Not stimulating enough | 90th percentile |
| No deadline | No urgency, no measurement | Specific date |
| No baseline | Impossible to calibrate difficulty | Measure before setting the goal |
Template: design your goal
My initial vague goal:
_______________
My specific + difficult + time-bound goal:
[Action] + [Quantity] + [Deadline]
My current baseline:
_______________
How I identified the 90th percentile:
- External benchmarks
- Advice from experienced peers
- My past best moments
- Projection of my progression
Verification:
- ☐ Specific — an outside observer can verify
- ☐ Difficult — 90th percentile, slightly scary
- ☐ Time-bound — clear deadline
In summary
A well-designed goal rests on three research-validated pillars:
- Specific — You know exactly what you're aiming for
- Difficult — At the 90th percentile of your capacity
- Time-bound — With a clear deadline
Complete it with a baseline to calibrate difficulty.
Forget SMART and its "realistic." Aim for the 90th percentile.
Designing a good goal is 20% of the work. The remaining 80% is execution — planning, acting, maintaining. That's the subject of our 5-step guide to achieving your goals.
Sources
- Locke, E. A., & Latham, G. P. (1990). A theory of goal setting and task performance. Prentice Hall.
- Locke, E. A., & Latham, G. P. (2002). Building a practically useful theory of goal setting. American Psychologist, 57(9), 705-717.
- Locke, E. A., & Latham, G. P. (2006). New directions in goal-setting theory. Current Directions in Psychological Science, 15(5), 265-268.
- Day, T., & Tosey, P. (2011). Beyond SMART? A new framework for goal setting. The Curriculum Journal, 22(4), 515-534.