The research behind Goal-Setting Theory
"Do your best" sounds like reasonable advice. It encourages effort without putting unnecessary pressure on the outcome. Yet decades of research on goal setting suggest that, in many situations, people perform better when the target is considerably more specific.
Psychologist Edwin Locke began studying the relationship between conscious goals and performance in the 1960s. His early experiments and later work with psychologist Gary Latham eventually developed into Goal-Setting Theory, one of the most extensively studied theories of motivation in organizational psychology.
The central finding is often summarized as specific, difficult goals lead to better performance. That is broadly consistent with the research, but it leaves out several important conditions. A goal needs to be sufficiently challenging, but it also needs commitment, feedback and a realistic relationship with the person's ability and the complexity of the task. In some situations, focusing too aggressively on a performance target can even make learning harder.
The research therefore tells us something more useful than simply "set bigger goals."
THE RESEARCH
Locke's early experiments examined whether different kinds of goals affected how hard people worked and how well they performed. Rather than treating motivation as a vague internal quality, he studied whether changing the target itself could change performance.
Across a series of laboratory studies and later field experiments, a consistent pattern emerged. People assigned specific, challenging goals frequently performed better than people given easy goals, vague goals or instructions simply to "do their best."
Latham later extended this work into real workplaces. In one well-known series of field studies involving logging operations, workers were given specific production goals for loading trucks. The researchers found that challenging goals could produce meaningful improvements in productivity compared with less specific expectations.
This mattered because the effect wasn't confined to artificial laboratory tasks. Goals appeared capable of changing behavior in environments where performance had real operational consequences.
Locke and Latham continued studying goal setting across different tasks and organizations for decades. By the early 2000s, they were able to review a large body of experimental and field research examining when goals improved performance and when they did not.
The overall pattern was strong, but it was not unconditional. Specific and difficult goals tended to improve performance when people were committed to the goal, possessed or could develop the necessary ability and received enough information to understand how they were doing.
THE FINDING
Why would a specific goal work better than simply asking someone to try hard?
One reason is that "do your best" does not define what your best is supposed to produce.
Imagine two people trying to save money. One decides to "spend less this month." The other decides to reduce discretionary spending by $300.
The first person can interpret almost any improvement as success. The second has a reference point against which behavior can be evaluated.
Specific goals reduce ambiguity. They make it easier to determine whether current behavior is sufficient and whether something needs to change.
Difficulty matters for a related reason. If a goal can be achieved with minimal effort, there is little reason to increase effort beyond that level. A more challenging goal creates a larger gap between current performance and desired performance.
But Goal-Setting Theory does not suggest that making a target arbitrarily difficult will continuously improve results. At some point, the target can exceed a person's capacity or become so unrealistic that commitment deteriorates.
The useful goal is therefore not simply the hardest imaginable outcome. It is one that is challenging enough to require meaningful effort while remaining credible enough to pursue.
THE THEORY
Locke and Latham identified several mechanisms through which goals can influence performance.
First, goals direct attention. A clear target makes information related to that target more noticeable and gives competing demands less priority. If your goal is to complete a proposal by Friday, activities that move the proposal forward become easier to distinguish from activities that merely make you feel busy.
Second, goals can mobilize effort. More difficult goals generally require more effort than easy ones, assuming the person remains committed to achieving them.
Third, goals affect persistence. A person who knows what they are working towards has a reason to continue allocating effort until the target has been reached.
Finally, goals can encourage the development of strategies. When the current approach is not producing sufficient progress, the discrepancy between the goal and actual performance can prompt someone to search for another method.
That final mechanism becomes particularly important when tasks are complicated.
If you already know how to perform a task, a performance goal can help you direct and sustain effort. If you do not yet know how to perform it, simply increasing the target may not solve the problem.
You may need to learn first.
NOT THE SAME THING
A difficult goal is not automatically a good goal
This distinction gets lost easily in popular discussions of ambition.
Suppose someone has never run before and sets a goal to complete a marathon in six weeks. The goal is certainly specific. It is also difficult. That does not make it effective.
Goals operate within the constraints of ability, knowledge, resources and time.
Task complexity matters especially because difficult tasks often require people to discover effective strategies before high performance becomes possible. Research by Locke, Latham and others has shown that in these situations, learning goals can sometimes be more useful than immediate performance goals.
A performance goal asks for an outcome: reach a certain sales number, achieve a particular score or complete a task within a certain amount of time.
A learning goal focuses on acquiring the knowledge or strategies necessary to eventually produce that outcome.
Someone beginning a new business, for example, might immediately set a target of generating $100,000 in revenue. The number is specific, but if they do not yet understand customer acquisition, pricing or sales, the target provides little guidance about what they need to learn.
A better early goal might involve identifying and testing several acquisition channels, interviewing potential customers or developing proficiency in a particular sales process.
Sometimes the obstacle is insufficient effort.
Sometimes the obstacle is that you don't yet know what you're doing.
Goal-Setting Theory works considerably better when those two problems are not confused.
WHY IT MATTERS
Many goals in ordinary life are surprisingly difficult to evaluate because they are never clearly defined.
People want to exercise more, read more, spend less, become more organized, improve their relationships, grow their business or get better at something.
These intentions can be perfectly sincere. The problem is that they contain no obvious threshold for success.
What counts as exercising more? Once this week? Four times? Twenty additional minutes? What does becoming more organized actually look like? What evidence would tell you that your relationships had improved?
Without a clear target, evaluation becomes subjective. You can feel as though you are constantly working on something without knowing whether your behavior is producing meaningful progress.
Specific goals create a reference point.
But the research also provides a useful warning against turning every aspiration into a number. Goals can narrow attention towards whatever is being measured. If the measure captures only one part of what matters, people can become very effective at improving the metric while neglecting the larger purpose behind it.
A business focused exclusively on revenue could sacrifice profitability. Someone pursuing a lower number on the scale could adopt behaviors that undermine health. A person trying to read fifty books in a year could become more concerned with finishing books than understanding them.
Specificity is useful when the target represents something you genuinely care about.
Otherwise, precision can simply make you better at pursuing the wrong thing.
TRY THIS
Choose one goal you currently describe using words such as more, less, better, healthier, stronger, successful or consistent.
Ask yourself what observable result would tell you that the goal had actually been achieved. If you want to "save more," determine how much. If you want to "read more," decide what amount would represent meaningful progress. If you want to "work on your business," identify what needs to exist at the end of the period that does not exist now.
Then ask a second question: Do I already know how to achieve this?
If the answer is yes, a specific performance goal may help direct your effort.
If the answer is no, don't simply make the target more ambitious. Identify what you need to learn and make that the immediate goal instead.
Finally, consider how you will know whether your current approach is working. A target without feedback leaves you unable to adjust until the deadline has already arrived.
A good goal doesn't just describe something you would like to happen. It gives your attention and effort somewhere specific to go while still leaving room to change the strategy when the evidence tells you to.
References
Latham, G. P., & Baldes, J. J. (1975). The "practical significance" of Locke's theory of goal setting. Journal of Applied Psychology, 60(1), 122-124.
Locke, E. A. (1968). Toward a theory of task motivation and incentives. Organizational Behavior and Human Performance, 3(2), 157-189.
Locke, E. A., & Latham, G. P. (2002). Building a practically useful theory of goal setting and task motivation: A 35-year odyssey. American Psychologist, 57(9), 705-717.
Seijts, G. H., & Latham, G. P. (2005). Learning versus performance goals: When should each be used? Academy of Management Executive, 19(1), 124-131.