Why SMART goals so often produce nothing

SMART came out of a 1981 management memo about organisational objectives, and the acronym has never been stable. Where the underlying research actually applies, the four failure modes, and what to add so a SMART goal becomes something you can do.

You have almost certainly been taught SMART. Specific, Measurable, Achievable, Relevant, Time-bound. It is on every corporate training slide, in every self-improvement book, and in the onboarding flow of half the goal apps on the store. And you have almost certainly written a properly SMART goal at some point and then done nothing about it.

That outcome is common enough to be worth explaining rather than excusing. The framework is not nonsense — it removes a specific kind of vagueness. But it is asked to do a job it was never designed for, and that gap accounts for most of the disappointment.

Where SMART actually came from

The acronym first appears in a 1981 piece by George T. Doran in a management journal, in the tradition of management by objectives. It was written for managers, about how to phrase organisational objectives so they could be written down, agreed and reviewed — a memo-writing aid for people setting departmental targets, at a time when a lot of corporate objectives were unfalsifiable sentiment.

That context matters more than it sounds. In the setting Doran was writing about, several things are true that are not true of your Tuesday. Someone else assigns the objective and has authority to hold you to it. There is a review cycle with real consequences. There is a team, a budget, and existing machinery for doing the work. Almost none of that transfers to a person deciding on a Sunday evening to get fit.

The letters have never been stable

This detail gets skipped, and it is telling. A stands for Achievable, or Attainable, or Assignable — Doran's original had Assignable, meaning "state who does it", sensible in a company and meaningless for a person acting alone. R is Relevant, or Realistic, or Results-based, or Resourced. Some versions bolt letters on the end: SMARTER, with Evaluated and Reviewed.

A model quietly redefined for forty years, letter by letter, by whoever happened to be teaching it that decade, is not a validated instrument. It is a mnemonic that survived because it is easy to say. That is not an accusation of fraud — Doran never presented it as a theory of human behaviour. It is a caution against treating a memory aid as though it had been tested.

To be plain about the evidence: there is no strong body of research showing that SMART specifically outperforms other ways of framing a personal goal. It is widely taught, widely liked, and thinly evidenced as a package. Its credibility is borrowed from a different and much better body of work.

What the real research says, and where it stops

The serious tradition here is goal-setting theory, developed by Edwin Locke and Gary Latham over several decades. Its central finding has held up across many studies: a specific and challenging goal produces better performance than "do your best". Vagueness lets you settle for whatever happened; a defined target does not.

So the "specific" part of SMART stands on real ground. But look at the conditions that research was mostly done under. Short, well-defined tasks — proofreading, brainstorming, entering data, hitting production targets. Clear and rapid feedback. A goal usually assigned by an experimenter or a manager rather than chosen. Horizons of minutes, hours or a quarter.

Now consider your goal. It runs for months. Nobody assigned it and nobody will notice if you drop it. The task is not well defined — "get fit" contains a hundred possible activities and you have to choose among them. Feedback is slow, noisy and easy to misread. The theory does not stop being true here, but the distance between the evidence and the application is large. The honest summary: specificity helps, the acronym is packaging, and the packaging has been oversold.

Four ways a perfectly SMART goal produces nothing

Measurable is not the same as actionable

This is the central failure and it is worth stating slowly. "Lose six kilos by 30 June" is specific, measurable, arguably achievable, relevant and time-bound. It scores five out of five. And you cannot do it. There is no hour of any day in which the action "lose six kilos" is available to you.

What you can do is cook on Sunday, walk to the station, stop buying the biscuits, book the class. None of those appear anywhere in the SMART formulation, because SMART is a test applied to the statement of an outcome. It makes the outcome checkable; it produces nothing you can perform, and a goal made entirely of outcomes has no surface to put your hands on.

Achievable quietly selects for the easy

Ask someone to make a goal achievable and watch what happens: they lower it until it feels safe. That is the opposite of what the underlying research supports — difficult goals outperform easy ones, given commitment and capability. The defensible reading of A is "not impossible and not dependent on things outside your control". The common reading is "comfortable". Those produce very different goals.

Time-bound gives you a deadline and nothing in between

A date six months out is not structure; it is a cliff you can see from a long way off and ignore for five of those months. Deadlines work when they are close enough to change today. A distant one provides a period of calm followed by a period of panic, and if the goal is the kind that cannot be crammed — fitness, a language, a body of writing — the panic does not save it.

It says nothing at all about falling behind

This is the omission that matters most, because falling behind is not an edge case — it is the normal condition of any goal running longer than a fortnight. SMART has no instruction for the week you got ill, the month work exploded, the point at which the original estimate turned out wrong by half. It describes how to write the sentence and then leaves the room.

What to add

None of this means throwing the acronym out. Keep the specific, keep the measure, keep the date — those do real work. Add three things it does not contain.

1. A phase layer. Two to four stretches of weeks, each with an end you would recognise. This converts a flat six months into a sequence of arrivals, and it is the layer almost everybody skips.
2. Actions you directly control. For each phase, two or three recurring things you will actually perform, plus a short queue of one-off unblockers. If you cannot do it on an ordinary Tuesday morning, it belongs at the goal or phase layer.
3. A review point. Weekly, and again at the end of a phase: look at what happened rather than what you intended, and change the plan rather than your self-image. SMART has no letter for this, and it determines whether the goal survives an ordinary month.

Northstar is built around this shape: it asks for a deadline and a measure and then refuses to stop there. A goal holds phases; a phase holds steps you can perform, each marked once, daily or weekly. It computes a health reading from the days remaining against how far along you are, so a goal you have not touched in three weeks says something while there is still time to act — which is the signal a bare deadline cannot give you.

A worked example: "Run a half marathon in under two hours by 1 October" is a fine goal statement, and it becomes executable only once you add phase one (run three times a week at any pace, six weeks, without getting injured), phase two (extend the long run to fifteen kilometres), phase three (a session at target pace, then taper) — and then steps for phase one alone: three runs a week, one strength session, buy shoes that fit, enter the race. None of that came from the acronym.

What SMART is still good for

It remains an excellent filter for one specific defect: goals written as feelings. "Be healthier", "get on top of work", "be a better parent" cannot be started because there is no moment at which they become true. Running them through SMART forces a date and a measure, and that alone rescues a lot of goals from permanent vagueness.

And it is well suited to its original job. Setting objectives for a team, with a review cycle and an owner for each, it does what Doran meant it to do. The mistake was not inventing it — it was exporting a memo-writing aid into personal change, presenting it as a method, and then blaming the people it did not work for.

Frequently asked questions

Related guides

• Goals versus systems: what the argument gets right and where it breaks
• How to break a big goal into steps you can do this week
• How detailed should a goal plan be?

All guides