Eight judges in Israel. 1,112 parole rulings. One uncomfortable pattern. Prisoners whose cases came up early in a session had roughly a 65% chance of a favorable ruling. Prisoners heard at the tail end of the same session had almost none. After the judges broke for food, the rate snapped back to about 65% and then slid again. The cases were not meaningfully different. The judges were. That finding comes from a 2011 study published in the Proceedings of the National Academy of Sciences, and it is the cleanest evidence we have that the quality of a decision falls as the pile of decisions grows, independent of how smart or how committed the person deciding happens to be.
Managers live inside that pattern every day. The difference is that nobody logs your rulings, so the decline is invisible until something expensive breaks. Decision fatigue in management does not arrive as a crisis. It shows up as the contract you approve at 4 p.m. without really reading it, the hire you have been putting off for three weeks, the performance conversation that keeps sliding to “next Monday.” You are not lazy or checked out when this happens. You are depleted, and depletion makes different choices than a rested mind does.
The afternoon version of you is a worse manager than the morning one
The mechanism is not mysterious. Every judgment call draws from the same limited pool of attention: the budget shift, the tie-breaking opinion a direct report asks for, the Slack thread that wants your read, the candidate you are supposed to evaluate. Spend enough of that pool and you do not stop deciding. You start deciding on autopilot, which usually means defaulting to the safe option, the easy yes, or no decision at all.
This is the quiet cousin of decision-making under pressure. Pressure is one loud, high-stakes moment you can see coming. Fatigue is a hundred small cuts you never notice landing. The parole data makes the point vivid because the stakes were a person’s freedom and the judges were trained experts, and the effect showed up anyway. Your approvals are lower stakes one at a time, but you make far more of them, and no bailiff announces that your capacity just ran out.
Management is built to drain this pool faster than almost any other job
Three forces have made the manager’s version of this problem sharper than it was even a few years ago, and each one is documented.
The first is volume. Microsoft’s 2025 Work Trend Index, built on productivity signals from more than 30,000 workers, found that the average knowledge worker is interrupted every two minutes by a meeting, email, or notification, roughly 275 interruptions in a standard day, on top of 117 emails and 153 Teams messages. Managers sit at the center of that stream, and a large share of those pings are not information. They are decisions in disguise: approve this, weigh in on that, pick between these two.
The second is time. DDI’s Global Leadership Forecast found that 71% of leaders report increased stress since stepping into their role, and 40% of those are considering leaving. Only about 30% feel they have enough time to do the job with the depth it requires. When time is that tight, the first casualty is not the work. It is the deliberation the work deserves.
The third is that the role itself is cracking. Gallup’s State of the Global Workplace 2026 report recorded manager engagement falling from 27% in 2024 to 22% in 2025, the largest single-year drop it has measured, while overall employee engagement sat at 20%, down from a 23% peak in 2022. Managers used to carry an engagement premium over the people they led. That premium has nearly evaporated. A disengaged manager making a high volume of fatigued calls is a compounding problem, because the team feels every one of those calls.
None of this is a willpower failure. It is a systems failure, and it needs a system, not more grit.
The Decision Triage Method
I started building this after watching capable managers drown in their own inboxes, and after doing it to myself more than once. The idea is deliberately plain, because a complicated framework is useless to someone who is already cognitively spent. It has three moves.
Sort every decision into red, yellow, or green
Red decisions are high impact and hard or expensive to reverse. Hiring a key role. Killing a product line. Restructuring who reports to whom. These deserve your sharpest hours and, usually, more than one brain in the room.
Yellow decisions are moderate impact and reversible with some effort. Choosing between two project tools. Approving a budget shift under $10,000. Setting a new meeting cadence. They need your judgment, not your agonizing.
Green decisions are low impact and easy to undo. Routine time-off requests. Which room to book. Which vendor swag to order. These should either never reach you or take under thirty seconds when they do.
The sorting itself is the insight. Most managers treat every decision as if it were red, which means they spend premium fuel on green noise and have nothing left when a genuinely red call lands in the afternoon.
Give each color its own protocol
Red decisions get scheduled into your peak window and given a deadline, because a red decision left to drift generates more fatigue than making it ever would. Bring the data and one or two trusted judgments, close Slack, and decide.
Yellow decisions get a timer. Ten minutes. What is the upside of option A, what is the downside, and can we reverse it if we are wrong? If it is reversible, pick the option that teaches you the most fastest and move on. This is exactly the distinction Jeff Bezos drew in Amazon’s 2015 letter to shareholders: Type 1 decisions are one-way doors that demand deliberation, while Type 2 decisions are two-way doors that “can and should be made quickly.” His warning was that as organizations grow, they start running the heavyweight one-way-door process on everything, and the result is slowness and timidity. Managers do the same thing to their own calendars.
Green decisions get delegated or automated with a simple rule. Time-off that does not clash with another absence gets auto-approved. Purchases under $200 do not need your signature. Your leads pick the room. Here is where a lot of managers balk, because handing these off feels like abdicating. The opposite is true. A manager who insists on touching every decision becomes a bottleneck and quietly teaches the team that their judgment is not trusted. Giving away green decisions with clear guardrails is one of the most direct ways to build real coaching habits: you are letting people practice real judgment on calls where a wrong answer is cheap.
Guard your first 90 minutes
This is where the method usually dies. Managers know which decisions matter, then spend their freshest hours clearing email and stamping out small fires, so the red call gets whatever is left at 3 p.m. Block the first 90 minutes of the day for red decisions. On the days you do not have one ready, use the time to think ahead about the red call you can see coming. Recognizing which hours are yours to protect is itself a piece of emotional intelligence as a manager, the self-awareness to notice that 9 a.m. you and 4 p.m. you are not the same decider.
What changed when I stopped gatekeeping
For a stretch of my two decades in IT operations, I ran this the wrong way. Managing operations at a large enterprise telecom, I treated being involved in everything as proof I was doing the job. I personally approved change requests, vendor invoices, access tickets, and shift swaps. One night near the end of an eleven-hour on-call stretch, I signed off on a maintenance window change I would have pushed back on at 9 a.m. It was reversible, nothing burned down, but the review I gave it was closer to a reflex than a judgment, and I knew it.
What fixed it was not working harder. It was sorting. The change-request approvals that actually carried risk, the ones touching customer-facing systems, I kept and moved to the morning. The low-risk ones, the routine access grants and the sub-threshold purchases, I handed to senior engineers with a one-page guardrail. In the first month almost none of them needed my override. I had been guarding gates that did not need a guard, and the guarding was what left me rubber-stamping the calls that mattered.
Later, doing fractional COO work through Ops Harmony, I saw the same thing in nearly every founder I worked with. The bottleneck was rarely strategy. It was a leader spending red-level energy on green-level decisions and then wondering why the important calls felt so hard. The fix travels: name the color, match the protocol, protect the window.
A 15-minute audit to run this week
Pull up last week’s calendar and task list and write down every decision you actually made, approvals and tie-breakers and strategy calls alike. Sort each one red, yellow, or green. Most managers find that 60 to 70% of their week was green, which is both deflating and good news, because green is the easiest to offload.
Pick three green decisions and, before the week closes, either hand them to a direct report with a clear rule or build the rule that eliminates them. Then block 90 minutes tomorrow morning and label it for your real decisions. Even if no red call is waiting, spend the time on the one you have been avoiding: the salary conversation you keep rescheduling, the vendor you have not replaced, the hire you keep deferring.
Decision triage is not about making fewer decisions. It is about spending a fixed cognitive budget where it moves the needle. Leading through change asks for clear-headed calls at nearly every turn, and your team is standing around waiting for them. The point of protecting your capacity is simple: so the version of you that decides their fate is the rested one, not the one running on fumes at the end of a long day.