How CEOs Decide Better Under Pressure

Pressure reveals judgement. It does not create it. When time is short, the data is late and three constituencies want different outcomes, you do not rise to a new intelligence. You fall to the quality of your gates. I learned that carrying P&Ls through hyper-growth and multi-country expansion, and I watch it every week in executive coaching.

The CEOs who look “calm under fire” are not calmer people. They have a way to stop the room collecting opinions as a substitute for a call.

Pressure does not invent judgement—it reveals it

In an easy quarter you can hide a sloppy decision process behind good markets. Under pressure the same sloppiness becomes visible: you reopen closed calls, you invite extra people so no one is solely on the hook, you wait for a model that will not arrive in time.

I do not coach people to feel less. I coach them to notice which of those moves is already their pattern. If you have a blind-spot for optimism, pressure will dress it up as pace. If you have a blind-spot for harmony, pressure will dress it up as “alignment.”

The tell is rarely the decision you made. It is the decision you postponed while looking busy. That postponement is also how burnout risk shows in the business before it shows in a speech.

A four-gate decision protocol

I use four gates. They fit on one page. If a call cannot pass them, it is not ready—or it is being used as theatre.

Frame. What are we actually deciding, by when, and what happens if we do nothing? If the frame is “discuss the market,” you do not have a decision. You have a seminar.

Options. Name at least two real options, including “do nothing.” A single option with a long justification is a sales pitch.

Irreversible costs. What can we not unspend, unhire, or un-announce? Time, reputation and key people are costs even when the spreadsheet is green.

Owner. Who will live with this after the meeting, and what is the first review date? No owner, no decision—only a mood.

I would rather a CEO walk through those four badly than skip them fluently. Fluency without gates is how companies walk into irreversible bets on a Friday afternoon. The guide has a longer treatment of making tough decisions when every option costs something.

This is the same spine I use when CEO coaching has to beat another leadership course: we work live calls, not case studies.

Late data is normal. A missing gate is what turns a corridor of options into a bad night.

What to do when data will not arrive in time

You will not get the dataset you want. Decide what would change your mind if it arrived, then decide without it.

I ask three questions. What is the cheapest fact that would flip this call? Can we get that fact in the time box? If not, which assumption are we accepting on purpose? Write the assumption down. That is how you stay honest when the quarter is loud.

Do not confuse more slides with more truth. Founders in the title-to-OS transition often stall here: they still believe proximity to the product will produce the missing number. Sometimes it will. Often it produces another late night and the same options you had on Monday.

Set a time box before you open the analysis. When the box ends, you choose, you schedule the review, or you kill the work. Those are the only three legal exits.

Decision table: reversible vs irreversible bets

Decision type	Time box	Who must be in the room	Pre-mortem question	Kill criterion
Reversible price or offer test	48 hours	Owner plus one challenger	What would make this embarrassing, not fatal?	No learning after two cycles
Reversible hire at mid-level	One week	Hiring manager and HR partner	What will we know in thirty days that we pretend to know now?	Proof points missed at day 30
Irreversible capital or exit	Days, not weeks, once framed	CEO, CFO, relevant operator, chair if needed	How does this fail in public?	A single assumption we cannot live with
Irreversible culture call (who stays)	Same week you know	CEO and the person’s manager — not a crowd	What standard dies if we wait?	Waiting becomes the decision
Board-facing strategy bet	Before the pack is printed	CEO plus the exec who will own the number	What will non-execs hear that we did not say?	The sentence still needs a paragraph

If you put an irreversible bet through a reversible process—casual corridor, no pre-mortem, no kill—you are gambling with the company’s name. If you put a reversible test through an irreversible process, you are using governance as a blanket.

Team behaviours that improve decision quality

The CEO can run perfect gates and still lose if the table is performing.

Require dissent by role, not by personality. One person is paid to argue the downside. Rotate it so it does not become the “negative” colleague’s identity. That is how you get usable dissent instead of a vibe.

Ban late surprises in the meeting. If a number was known yesterday, it belongs in the pack. Ambush is not candour.

Record the assumption, the owner and the review date in the minutes. Culture is what you write down when everyone is tired. A visible record is also how you earn trust that survives a bad call.

If the team cannot do those three, 1:1 coaching on the CEO will not fix the table. You will just become a better-prepared hero.

Linking decisions to vision and scoreboard

A good call that does not connect to the sentence the company can repeat is just activity. I keep a vision check in the room: does this move the one-line destination, or are we entertaining a clever detour? That is why transferring vision is still job-one.

The scoreboard is how you inspect the call later. If you cannot name the lead indicator that should move, you did not make a decision. You made a speech.

The seven-step process exists so this does not stay as a one-off protocol in a notebook. Frame, options, cost, owner—then the rest of the leadership system has to carry it.

I will not tell you this protocol produces better quarters. I will tell you it produces inspectable calls. That is the only kind worth making under pressure.

Closing

When the data is late, your gates are the product. Write them down. Use them when you are tired. Make the team use them when you are not in the room.

If you want a private pass over the next irreversible call, book a confidential 30-minute conversation.