A senior team agrees in a Wednesday meeting that the new pricing tier will launch by end of quarter. The decision is recorded in the minutes. Three weeks later, nobody is sure whether the launch is still on, who is doing what, or which open question is blocking it. By month-end, the pricing tier has quietly been deferred. The decision was made. The decision did not survive.
This is one of the most expensive patterns in modern operations and one of the least visible. The cost does not appear on any P&L line. It shows up as missed quarters, repeated discussions, executive frustration, and the slow erosion of trust in whatever the leadership team last agreed on. The instinct, when this pattern becomes visible, is to fix it with discipline — better minute-taking, stricter follow-up, more accountability culture. The real cause is structural, not behavioral. And the fix requires two specific levers, not more willpower.
Most organizations treat decisions as events. The meeting happens, the call gets made, the decision is announced, and the matter moves off the agenda. What is missing is the structural connection between the decision and the work that has to follow. There is no defined owner outside the meeting room, no deadline that the process actually tracks, no metric that signals whether the decision is producing the expected outcome, and no recurring forum that asks the unglamorous question: is this thing still on track, or did we quietly forget about it?
The result is a strange paradox. Leadership teams that pride themselves on being decisive are often the most exposed to this dynamic. They decide quickly, frequently, and confidently — and a meaningful fraction of those decisions evaporate within weeks because no system carries them forward. The faster the decision velocity, the heavier the backlog of unfollowed-through decisions, and the deeper the cynicism inside the organization that nothing announced in a leadership meeting actually changes the operating reality. Two structural levers fix this without slowing decision-making down: a deliberate decision artifact and a recurring decision review cadence.
Walk into the recurring leadership meetings of a mid-sized organization and you will usually see the same pattern. The agenda is long. Each item gets ten or fifteen minutes. Conversations stay alive long enough to surface a position, and then a decision is announced — sometimes by the most senior person in the room, sometimes by exhaustion. The note-taker captures the outcome in a sentence. The next item begins. Nobody is responsible for asking, in the moment, the four questions that turn a decision into a decision the process can carry: who owns this, by when, with what success criterion, and how will we know it is on track.
What happens instead is that the decision becomes a performance. It exists in the moment of the meeting. It exists in the minutes. It exists in the immediate week of follow-up activity. Then it begins to fade. The original owner, if there was one, is not reminded. The deadline, if there was one, slides quietly. The success metric, if it was ever named, is not tracked anywhere. Three months later, somebody asks in a different meeting "didn't we decide this in March?" and the answer is yes, but nothing is different. The decision lived only as long as the meeting held it.
Operationally, this is indistinguishable from not having made the decision at all. Worse: the organization has spent the cognitive energy of debating it, the political capital of aligning on it, and the time of every person in the room — and produced no change in behavior. The cost is not the meeting. The cost is the absence of what was supposed to happen next.
When you trace this pattern across a year, the numbers get uncomfortable. A leadership team that makes four to six meaningful operational decisions per week, with a meaningful follow-through rate of 50 to 60 percent, is producing somewhere between 100 and 150 decisions per year that consumed leadership time and produced no operational change. Each of those decisions cost three to five hours of executive attention to make and another two to four hours of downstream activity in the week that followed before the energy dissipated. Even at a conservative blended rate of senior labor, that is somewhere between 800 and 1,500 hours of leadership capacity per year burned on decisions that quietly did not survive.
The harder cost is not the time. It is the second-order effect. When a meaningful fraction of decisions do not land, the organization adapts. People stop treating leadership announcements as the actual operational signal. They wait for the second or third repetition of the same decision before they reorganize their work. They become more conservative about volunteering to own anything because ownership without follow-through is just exposure. The decision-making culture starts to look fast and decisive on the surface, while underneath, execution velocity slows down because nobody is sure which decisions will actually matter. This is a structural problem rooted in process design, not a behavior problem rooted in attitude or laziness. (We have explored how this dynamic shows up specifically in the form of escalation — where unmade or unsurvived decisions get routed upward — in When Escalation Becomes the Default Way to Get Things Done.)
Two structural levers consistently move this needle without making decision-making slower or more bureaucratic.
The first lever is to require, at the moment a meaningful decision is made, that it produce a small structured artifact — a decision record — captured in a system that the organization actually uses. The artifact answers four questions in concrete terms. What is being decided. Who owns the execution. By when. What signal will tell us whether it landed. This is not a memo. It is not a slide. It is a structured entry in a tool — a BPM workflow, a project system, a dedicated decision register — that has fields, an owner, a state, and a place in a queue.
The discipline is in the moment of capture, not in the document itself. Most leadership teams accept the idea of capturing decisions and then fail at the four-question test in practice. They name the decision but not the owner. They name the owner but not the deadline. They name the deadline but not the success metric. Each omission is the seed of a decision that does not survive, because each omission removes one of the hooks that the process needs to carry the work forward. The discipline of forcing all four answers, in the room, before the agenda moves on, is what turns a decision from an event into a process input.
This is uncomfortable at first. It slows the meeting down by two or three minutes per decision. Executives who pride themselves on velocity sometimes resist it. The argument is that the cost of friction in the moment far outweighs the cost of decisions that quietly die, and the data supports this. Organizations that adopt this discipline consistently see their landed-decision rate move from 50 to 60 percent into the 80 to 90 percent range within two quarters, because the four-question artifact is operationally what makes the difference between a decision and a wish.
The artifact also has a second effect that compounds. Because every decision is now in a system, the organization can finally see patterns. Which kinds of decisions land. Which kinds quietly fail. Which owners reliably execute. Which decision categories produce the most evaporation. This visibility is the basis for the second lever, which is what actually keeps the system honest over time.
The second lever is a recurring forum — typically once every two weeks for thirty minutes — whose only job is to review the open decision records. Not the operational projects. Not the strategic discussions. The decisions themselves. The agenda is mechanical and short. Walk the decision register. For each open decision, three questions: is it on track, is it at risk, or has it failed. If on track, move on. If at risk, what intervention. If failed, name it as failed and either re-decide or close it.
The cadence does something that no individual meeting can. It builds, across the organization, the lived expectation that decisions are tracked. Owners know they will face the register every two weeks. Sponsors know that abandoning a decision will be visible. The leadership team itself sees the cumulative pattern — how many decisions are landing, how many are quietly failing, which categories are most fragile. Within two or three cycles, the organization's behavior shifts. People take ownership more carefully because they will be asked about it. People are more honest about what is actually going to land before they accept the assignment. The decision culture starts to look slower in the front end and dramatically more effective in the outcome. (For organizations where local optimization is masking system-level decision failure, the same dynamic operates at a different scale — we wrote about that pattern in The Better Each Team Performs, the Worse the System Gets.)
The cadence must be inviolable. The single most common failure mode is that it gets skipped — for an offsite, for a quarter-end push, for a "more urgent" topic — and after two or three skips, it dies. The leadership team has to treat the cadence the way they treat their financial close: a non-negotiable recurring event that runs regardless of what else is happening. If the cadence is skippable, the lever is not real. If the cadence is protected, the organization learns within two quarters that decisions actually have to be made well, because they will be seen for what they are.
The checking principle for both levers together is simple: a decision is not a decision until it has produced an artifact, and the artifact is not real until it is visible in a recurring review. Anything that does not pass both tests is, operationally, a wish.
Of the last twenty operationally meaningful decisions made by your leadership team, how many landed in the form they were originally announced, and how do you know?
How would your decision velocity, decision quality, and execution rate change if every meaningful decision had to produce a four-field artifact in the room before the agenda moved on?
What is the cost, in your organization, of decisions that consumed leadership time, generated a wave of follow-up activity, and then quietly evaporated — and who in the company carries the cumulative emotional and operational cost of that pattern?
How often does a leadership-level decision in your organization get revisited in a later meeting because the original one didn't survive, and what does the repetition cost you in cycle time and credibility?
If you protected a thirty-minute decision review cadence twice a month for the next two quarters and refused to skip it, which currently-evaporating decisions would land that today are not landing?
A decision is not an event. A decision is the start of a small process that has to be carried by an owner, against a deadline, toward a measurable outcome, under recurring review. Organizations that internalize this stop romanticizing decision-making and start engineering it. They give up the performance of decisiveness and gain the substance of execution.
Pick one recurring leadership meeting this quarter and impose the two levers. Every meaningful decision produces a four-field decision record before the next item begins. A thirty-minute review cadence runs every two weeks, no exceptions. Six weeks in, look at the difference between what you are deciding and what is actually landing. The gap is the price you have been paying. The two levers close it.
Because most organizations treat decisions as events rather than as the start of a small process. The decision is announced, the meeting moves on, and there is no defined owner, deadline, success metric, or recurring review that carries the work forward. Without those four structural hooks, even good decisions evaporate within weeks.
A Decision Record is a small structured artifact, captured in a system the organization actually uses, that answers four questions for every meaningful decision: what is being decided, who owns the execution, by when, and what signal will tell us it landed. It turns a decision from an event into a process input that the organization can track, review, and learn from over time.
A Decision Review Cadence only reviews open Decision Records — not projects, not strategy. Its job is mechanical: walk the register, ask whether each open decision is on track, at risk, or has failed, and decide on intervention. Thirty minutes every two weeks. The narrow scope is what gives it the power to expose decisions that are quietly evaporating.
The four-field artifact adds two to three minutes per decision in the room. The data consistently shows that this cost is dramatically outweighed by the reduction in evaporated decisions and the elimination of repeat discussions in later meetings. Organizations that adopt the discipline typically move from a 50 to 60 percent landed-decision rate into the 80 to 90 percent range within two quarters.
That it gets skipped — for an offsite, a quarter-end push, or a more urgent topic — and after two or three skips, it dies. The leadership team has to treat the cadence as inviolable, like a financial close. If the cadence is skippable, the lever is not real. If it is protected, the organization learns within two quarters that decisions actually have to be made well.