The most expensive meeting in most businesses is the one at the end of the quarter where somebody asks whether the plan worked, and the room goes quiet. Not because the answer is bad. Because there isn't one.Nobody wrote down what success would have looked like in January, so there is nothing to check the year against. The plan gets renewed by default, and the same meeting happens again in twelve weeks.
Gallup's mid-2025 tracking found that 47% of US employees strongly agree they know what is expected of them at work. Read that the other way round and slightly more than half the workforce turns up each day without a firm sense of what would count as doing the job well.
That usually gets filed as a culture problem, something to fix with better communication. It is more useful to treat it as a line item. The money leaves the business in ways that are perfectly countable once you know where to look, and the repair is more mechanical than most owners expect.
Where the money actually goes
Gallup puts the productivity cost of disengagement in the US at roughly $2 trillion, and unclear expectations sit close to the root of it. Inside a single business the leak takes smaller forms. Two people build overlapping versions of the same thing because nobody said which mattered. A decision waits three weeks because no one is sure whether it is theirs. The repair is unglamorous: attach a number to the intention, which is the entire idea behind objectives and key results, and the gap between a goal that reads well and one a team can act on is easiest to see in worked examples laid side by side.
The most expensive version is quieter. A quarter ends, the honest answer to "did that work?" is that nobody can say, and the same plan gets renewed by default. A business that cannot evaluate last quarter has no mechanism for stopping anything, so the cost of a poor decision is not the decision but every quarter it survives.
Why goals go vague in the first place
Rarely through carelessness. Vagueness is usually the sensible response to an incentive.
A goal with a number in it can be missed in public. A goal phrased as "improve customer experience" cannot, because you can always point at something that improved. If missing a target has consequences for someone's review, the safest target is one that cannot be scored, and most people work that out without being told.
The second cause is inheritance. A target arrives already phrased as an aspiration, and each layer that passes it down softens it further. By the time it reaches the people doing the work, it has no edges left.
What the OKR format does to a vague goal
Objectives and key results are less a methodology than a forced separation of two things businesses habitually mash together: what you are trying to achieve, and how you would know. The objective carries the direction in plain language. The key results are numbers with dates on them.
Take "improve onboarding", which fails the only test that matters: would two reasonable people, looking at the same evidence in twelve weeks, agree on whether it happened? Rewritten in that format it becomes something like:
Objective: Make new customers successful in their first month without hand-holding.Key result 1: Raise the share of new accounts completing setup within seven days from 41% to 60%.Key result 2: Cut first-month support tickets per new account from 3.4 to 1.5.Key result 3: Raise 30-day retention from 68% to 80%.
The objective stays readable by anyone in the business. The key results settle the argument in advance and point at three things somebody can go and look at on Monday. None of those numbers has to be one you already track. It has to be one you are willing to find out.
Ambition, and the rhythm that makes it pay
Having found the discipline of specific goals, most businesses then set them too low. Decades of research on goal setting, much of it associated with Edwin Locke and Gary Latham, points the other way: specific and difficult goals tend to produce better performance than comfortable ones, as long as people are committed to them and get feedback along the way.
Those two conditions carry the weight, and they are where most OKR attempts fall over. A difficult goal nobody believes in produces resignation rather than effort, and one reviewed only at the deadline produces a scramble in the final fortnight. Fifteen minutes a week updating where each number stands is what turns the format into something that changes decisions, and it is the part businesses skip because it looks like meeting overhead.
Worth saying plainly: it does not suit everything. Work meant to stay stable belongs on a dashboard, not in a quarterly objective, and attaching attainment to bonuses reliably kills the ambition the format exists to encourage.
A reasonable place to start
One objective for the business and two or three key results underneath it, for a single quarter. Leave everything else as it is. A business that can name a handful of measurable priorities and say honestly whether it hit them is already operating differently from one running fifteen aspirations, and that difference tends to show up in the accounts before anywhere else.
