Goal setting theory: what Locke and Latham found, and why OKRs quietly depend on it

Ask anyone why goals should be 'specific and measurable' and you'll get a shrug, a reference to SMART goals, or nothing at all. It's accepted wisdom, repeated so often that nobody remembers where it actually came from.

It came from somewhere specific: goal setting theory, the body of research built by psychologists Edwin Locke and Gary Latham starting in the late 1960s. By 1990 they'd pulled together the results of hundreds of studies, run across industries as different as logging, insurance sales and engineering, into what's still one of the most replicated theories in industrial-organisational psychology. It isn't a productivity hack. It's an explanation for why some goals reliably produce better performance and others quietly fail, no matter how inspiring they sounded in the all-hands meeting.

If you've ever wondered why a stretch goal motivates one team and demoralises another, or why a target with no deadline slowly evaporates, the answer is in this research. Not vibes. Forty years of data.

What goal setting theory actually says

Locke's original insight, published in 1968, was simple to state and hard to argue with: people given specific, challenging goals perform better than people told to 'do your best.' That last phrase matters. Locke's early experiments compared specific hard targets directly against the vaguest instruction available, and 'do your best' lost, consistently. A vague goal lets people define success down after the fact. A specific one doesn't.

Latham joined Locke's research programme in the 1970s, and together they spent the next two decades testing the idea against real work rather than just lab tasks. Their 1990 book, A Theory of Goal Setting and Task Performance, synthesised all of it into a single claim: specific, difficult goals lead to higher performance than easy goals or vague ones, but only when four other conditions are also true.

Those four conditions are the part that gets dropped every time this becomes a listicle.

The five mechanisms that actually make a goal work

Goal setting theory doesn't say 'write a hard goal and you're done.' It identifies five things that all have to be present before a goal changes behaviour: clarity, challenge, commitment, feedback, and task complexity. Drop any one of them and the goal stops working, even if the others are solid.

The commitment mechanism is the one most goal-setting advice skips entirely, and it's also the one with the most research behind it: goals set participatively (the person doing the work helps set the number) produce roughly the same performance as goals assigned top-down, but only when the assigned goal comes with a clear rationale. Hand someone a number with no explanation and no say, and commitment quietly erodes before the quarter even starts.

Why this is the real backbone of OKRs

Here's the part that doesn't get said enough: OKRs didn't invent any of this. Andy Grove built the OKR format at Intel in the 1970s, around the same years Locke and Latham were publishing the studies that became goal setting theory, and John Doerr popularised it decades later at Google. The format works, when it works, because it happens to implement all five mechanisms the research identified: an objective and key result are specific and measurable (clarity), key results are meant to be ambitious (challenge), a cadence of weekly check-ins keeps progress visible (feedback), and splitting annual goals into quarterly ones is exactly the task-complexity fix the theory calls for.

The mechanism most OKR programmes get wrong is commitment. A key result handed down from leadership, with no input from the team that owns it, can be perfectly specific and appropriately difficult and still fail, because nobody actually committed to it. That's the same failure mode behind the sandbagging problem: when scores get tied to compensation, people rationally negotiate the number down before they ever write it in the sheet, a pattern management by objectives ran into decades before OKRs existed.

This is exactly the operational layer StratOps is built to own: not just picking the numbers, but running the cadence, the check-ins and the review rhythm that keep clarity, challenge, commitment and feedback all live at the same time, quarter after quarter, instead of alive for one meeting and then forgotten.

Where most goal-setting advice quietly ignores the theory

Most 'how to set better goals' content stops at clarity. Make it specific, make it measurable, done. That's SMART goals, and it's genuinely useful as far as it goes, but it covers roughly one of the five mechanisms the research actually describes.

Skipping commitment and feedback is why so many well-written goals fail anyway. A goal can be perfectly specific and still get ignored if the person accountable for it never bought in, which is the same ground covered in our piece on intrinsic motivation: a goal only sustains effort if it's meaningful to the person holding it, not just clearly worded.

Challenge is the other principle that gets misapplied most often. Locke and Latham's data shows performance rising with difficulty, but only up to the point the goal still feels achievable. Push past that and you don't get more effort, you get disengagement, or worse, the kind of quiet risk-aversion covered in the psychology of ambitious goals: when a stretch goal feels punishing rather than aspirational, people protect themselves by aiming lower next time instead of higher.

A worked example

Two support teams are both given the same target: cut average response time by 20% this quarter.

Team A gets the number in a message from their director. Nobody asked whether 20% was realistic, nobody checks in until the quarterly review, and the goal quietly becomes wallpaper by week three. Specific, sure. Nothing else.

Team B sets the same number, but the team lead brings it to the group first, asks what's realistic given current staffing, and agrees on 20% together, with a stretch option if things go well. They check in weekly on whether response time is trending the right way, and adjust resourcing when it isn't. Same target, same difficulty, wildly different result, because Team B has all five mechanisms running at once: clarity, challenge, commitment, feedback, and a cadence that breaks the quarter into something smaller than 'wait and see in twelve weeks.'

Common mistakes when applying goal setting theory

• Writing a specific goal and stopping there, on the assumption that clarity alone will change behaviour.
• Handing down a difficulty number without ever asking the team if they believe it, which quietly kills commitment before the quarter starts.
• Treating 'ambitious' and 'achievable' as opposites to pick between, instead of a balance to calibrate.
• Reviewing progress once, at the end of the quarter, instead of building in feedback while there's still time to act on it.
• Setting one big number for a genuinely complex goal instead of breaking it into a shorter, checkable cadence.

None of these are new mistakes. They're the same five mechanisms, just missing one at a time.

Where Tability fits

Goal setting theory is nearly fifty years old, and it still explains why some teams' OKRs stick and others quietly die around week six. The fix isn't a better goal-writing template. It's an operating cadence that keeps clarity, challenge, commitment and feedback all running at once, which is what weekly check-ins are actually for, not just a status update ritual.

Tools like Tability are built around exactly that cadence: setting the goal, tracking it against the OKR framework, and running the check-in rhythm that keeps feedback alive between quarterly reviews instead of saving it all for one meeting.