The Problem With Decisions Under Pressure
The decisions that matter most in business rarely come with complete information, unlimited time, and no emotional loading.
They come with: partial data, a deadline, conflicting stakeholder interests, and the weight of being the person who has to call it.
I've made a lot of consequential decisions in situations that look like this. As a founder operating without co-founders, without advisors, and without the safety net of institutional backing, every major call landed on me. Which clients to take. Which to decline. Whether to make a hire. How to respond to a client crisis. Whether to invest in a bet that wouldn't pay off for a year.
I've gotten some of them wrong. I've developed, over years of that kind of exposure, a framework that helps me get more of them right.
The First Principle: Separate the Decision From the Emotion
Under pressure, the emotional state and the decision-making process become entangled. The anxiety about being wrong, the desire to protect existing relationships, the fear of missing out — these things don't provide decision-relevant information. They just add noise.
The first step in any high-stakes decision is to notice the emotional state and explicitly bracket it: "I'm anxious about this. That anxiety doesn't tell me which option is correct. Let me set it aside and think about what the information actually says."
This isn't about suppressing emotion — it's about not letting it hijack analysis. The anxiety might be telling you something important (pay attention to the risk) or it might just be the discomfort of uncertainty (not useful). You have to distinguish between the two.
The practical tool: write down what you're feeling about the decision before you analyze it. Externalizing the emotional state creates distance from it, which makes the subsequent analysis cleaner.
The Pre-Mortem
One of the most useful decision-making tools I've adopted is the pre-mortem — a technique developed by psychologist Gary Klein.
Instead of asking "will this work?", ask: "It's 18 months from now and this decision has failed completely. What went wrong?"
The pre-mortem is powerful because it bypasses the optimism bias that affects forward-looking analysis. When you ask "will this work?", your brain is biased toward generating confirming evidence. When you ask "what went wrong in the failure scenario?", you unlock a different cognitive mode — one that's much better at identifying second-order risks and hidden assumptions.
After running the failure scenario, run the inverse: "It's 18 months from now and this decision succeeded beyond expectations. What had to be true?"
The gap between what had to be true for success and what you can actually verify is the risk map for your decision.
The Reversibility Test
Not all decisions carry equal stakes. The ones that deserve the most careful analysis are the irreversible ones — decisions where a wrong call creates a situation that's very costly or impossible to undo.
Before investing cognitive resources in a decision, ask: is this reversible or irreversible?
Irreversible decisions (major hires, significant long-term client commitments, major product pivots, public statements with reputational stakes) deserve slow, careful analysis and a high bar for confidence before proceeding.
Reversible decisions (which campaign approach to test, how to structure a proposal, what content to publish next week) can be made faster with a lower confidence threshold. The cost of being wrong is low — you can adjust, and you've probably learned something useful.
Most founders I've observed apply roughly the same cognitive intensity to both categories. They're either too slow on reversible decisions (because the process is the same for everything) or too fast on irreversible ones (because they're used to making quick decisions on everything else).
Calibrating decision speed and rigor to reversibility saves cognitive bandwidth and improves outcomes simultaneously.
The Information Threshold Question
A persistent failure mode under pressure is waiting for information that won't arrive before the decision needs to be made.
The question to ask: "What additional information, if I had it, would change my decision — and is there a realistic way to get that information in the time available?"
If the answer is "there's information that would change things, and I can get it in 48 hours," then wait 48 hours.
If the answer is "there's information that would change things, but I can't get it in time," then the decision has to be made on incomplete information — which means acknowledging the uncertainty, identifying the highest-risk assumption, and building in a contingency for the scenario where that assumption is wrong.
If the answer is "I'm not sure what information would change things," then the problem isn't information — it's clarity about what you're actually deciding. Start there.
The Two-Year Test
When a decision feels too close to call, I run the two-year test: which option will I be more satisfied with if I'm looking back from two years from now?
This is deliberately not a five-year or ten-year test. Those timelines are too abstract for most real-world business decisions. Two years is close enough to be concrete and far enough to escape the immediate emotional loading.
The two-year test is also good at separating preference from judgment. Sometimes what I want to do in the moment is different from what my two-year-future self would have preferred. Identifying that gap is useful information.
What I've Learned From Getting Things Wrong
The decisions I've gotten most wrong in business share a pattern: I made them in a state where I was emotionally invested in a specific outcome before I'd done the analysis.
The client I took because I was excited about the logo, before evaluating whether the scope was right. The strategy I committed to because I'd already announced it internally, before the data was in. The hire I made because I was lonely running operations solo, before I'd defined clearly what success looked like for the role.
In each case, the emotional commitment preceded the rational analysis. The analysis then became, unconsciously, a process of justifying the already-formed preference rather than evaluating the options.
The corrective is to make the emotional state explicit before doing the analysis. It doesn't prevent bias, but it makes bias visible — and visible bias is at least something you can account for.
The Decision Log
The practice that's most improved my decision quality over the long run: keeping a decision log.
Before any significant decision, I write down: the decision I'm making, the options I considered, the information I had, my reasoning for the choice I made, and what I expect to happen.
Six months or a year later, I review. Was the reasoning correct? Was the outcome what I expected? If not — was the disconnect in the reasoning or in something I couldn't have known?
The log creates accountability to your own thinking. It also, over time, reveals patterns in where your reasoning tends to fail — which is the most valuable calibration you can have for future decisions.
Key Takeaways
- Separate the decision from the emotion — bracket the emotional state before analyzing options; emotion adds noise, not signal
- Run the pre-mortem — "what went wrong in the failure scenario?" unlocks risks that optimism bias hides in forward-looking analysis
- Apply the reversibility test — irreversible decisions deserve slow, high-bar analysis; reversible ones can be made faster
- Ask what information would change your decision — if you can get it, wait; if you can't, acknowledge the uncertainty and build contingencies
- The two-year test separates immediate preference from considered judgment
- Keep a decision log — accountability to your own reasoning reveals your systematic blind spots over time