When Your Gut, Your Data, and Your AI All Disagree

Five Questions to Examine How You Actually Make Decisions Under Pressure Week 21: Decision Making Under Pressure | The Mindful Leader

Research shows that eighty percent of aviation accidents are traced to pilot error, and when those errors are analyzed, the pattern is clear: difficulty making good decisions under time pressure. Not lack of skill. Not lack of training. Not lack of instruments showing them data. Difficulty choosing wisely when the clock is ticking, information is conflicting, and the stakes are real. Leaders face the same challenge in different contexts. Different altitude. Same pressure. Same tendency toward poor decisions when stressed. Same challenge navigating what to trust when gut instinct, data analysis, and algorithmic recommendations all point in different directions.

The Gap Between How You Think You Decide and How You Actually Decide

When Damola Adamolekun took over Red Lobster in August 2024, the company was in bankruptcy with over a billion dollars in debt. Ninety-three locations had already closed. The pressure to make dramatic, visible decisions was enormous. Everyone was watching. Employees needed reassurance. Customers wanted answers. Investors demanded action.

And one of his first major decisions was to refuse to bring back the Ultimate Endless Shrimp promotion that customers loved but had cost the company eleven million dollars in losses. It was unpopular. It generated criticism. But it was right.

Here’s what’s interesting about that decision: Adamolekun’s gut probably told him to bring it back. Customers wanted it. It drove traffic. It was part of Red Lobster’s identity. The emotional pull toward keeping a beloved promotion must have been strong.

The data was clear. Eleven million dollars in losses. Unsustainable. No path to profitability. The numbers said kill it, even if instinct said save it.

But here’s what makes this moment particularly relevant for leaders right now: If Adamolekun had access to AI-powered customer sentiment analysis, it might have shown positive engagement with Endless Shrimp content. Predictive models might have projected traffic decline without it. Competitive analysis might have shown similar promotions working elsewhere.

So what do you do when gut says one thing, data says another, and AI recommendations add a third perspective? How do you decide what to trust?

He chose financial fundamentals over everything else. That’s not always the right call. Sometimes gut is right and data is misleading. Sometimes AI identifies patterns humans miss. But in this case, choosing discipline over instinct and long-term sustainability over short-term satisfaction was what the situation demanded.

Most leaders believe they make decisions rationally, especially important ones. We tell ourselves we weigh evidence, consider options, and choose logically. But research on decision-making under pressure tells a different story. When stakes are high and time is short, we default to pattern matching, emotional responses, and cognitive shortcuts. We think we’re being rational. We’re actually being reactive.

And what I’m seeing in foodservice leadership right now is that having more information sources—data dashboards, AI recommendations, predictive analytics—hasn’t made this easier. It’s made it more complex. Because now you’re not just navigating gut versus data. You’re navigating gut versus data versus algorithmic recommendations, and trying to figure out which one to trust when they conflict.

The question isn’t whether you use gut, data, or AI. Everyone uses all three. The question is: Do you know which one is driving your decisions, and are you deliberately choosing or just defaulting?

Five Questions to Examine Your Decision-Making Patterns

This weekend, take time to examine how you actually make decisions under pressure. Not how you think you decide. How you actually decide when the clock is ticking, information is conflicting, and the stakes are real.

These questions aren’t comfortable. They’re designed to reveal the gap between your stated decision-making process and your actual decision-making behavior. Sit with them. Be honest. No one else needs to see your answers.

QUESTION 1: WHEN YOUR GUT, YOUR DATA, AND YOUR AI ALL DISAGREE, WHICH ONE WINS? AND WHY?

Think about the last five significant decisions you made under pressure. When intuition pulled one way, evidence pulled another, and AI recommendations (or data dashboards, or analytics) pulled a third direction, which did you follow?

If you consistently choose gut over everything else, why? Is it because your gut is genuinely wise from years of experience? Or because you’re uncomfortable with ambiguity and gut decisions feel more confident? Or because you don’t fully trust data you don’t completely understand?

If you consistently follow data or AI recommendations, why? Is it because you’ve learned to trust systematic analysis? Or because you’re using technology as cover for decisions you’re afraid to own? Or because “the algorithm said so” feels safer than personal accountability?

Here’s what I’m watching happen in foodservice: Leaders who grew up trusting instinct and experience are now getting AI recommendations they don’t fully understand. When AI suggests something that contradicts their decades of experience, they face an uncomfortable choice: Trust the algorithm or trust themselves?

Neither approach is always right. Sometimes gut is pattern recognition from deep experience. Sometimes it’s anxiety masquerading as insight. Sometimes data reveals truth. Sometimes it obscures it by quantifying the wrong things. Sometimes AI identifies non-obvious patterns. Sometimes it optimizes for metrics that don’t align with your actual goals.

The question is whether you’re deliberately choosing based on context or reflexively defaulting to one mode regardless of situation.

Adamolekun chose financial fundamentals over customer sentiment for Endless Shrimp because the situation demanded it. Financial crisis. Clear losses. Unsustainable model. The numbers were reliable.

But he chose speed and responsiveness over extensive analysis when customers requested menu items via social media. Context determined the approach, not a fixed preference for one type of input.

Reflect honestly: Do you have a default mode when gut, data, and AI conflict? When do you override it? When should you override it more often? What makes you uncomfortable about the input sources you tend to dismiss?

QUESTION 2: WHAT KIND OF PRESSURE MAKES YOU GRAB FOR AI AS A CRUTCH VS. A TOOL?

Not all pressure affects you the same way. And technology doesn’t affect all leaders the same way under pressure. Some types of pressure make you over-rely on AI recommendations because “having data” feels safer than trusting judgment. Other types make you ignore AI entirely because human instinct feels more reliable.

Think about your own patterns:

When do you over-rely on AI/data?
• When you’re afraid of being wrong?
• When the decision affects people and you want objective justification?
• When you don’t trust your instincts in this specific domain?
• When you’re being questioned and want “data-backed” answers?
• When you’re exhausted and don’t want to think hard?

When do you dismiss AI/data entirely?
• When it contradicts what you “know” from experience?
• When you don’t understand how the algorithm reached its conclusion?
• When it suggests something that feels wrong culturally or ethically?
• When you’re emotional and trust feeling over analysis?
• When you need to decide fast and processing data feels slow?

Neither pattern is inherently wrong. But understanding your patterns helps you compensate for your tendencies.

What I’m seeing in foodservice leadership: Leaders using AI recommendations as decision camouflage. “The data said to do it” becomes a way to avoid accountability for judgment calls. Or the opposite, leaders dismissing perfectly good insights because “I’ve been in this business 30 years and no algorithm knows better than me.”

Both are failures to integrate AI appropriately. One abdicates judgment to technology. The other dismisses potentially valuable information because of ego.

Reflect honestly: What specific types of pressure make you grab for AI as a crutch instead of a tool? What types make you dismiss it entirely? How can you recognize these patterns in the moment and compensate?

QUESTION 3: WHO ARE YOU TRYING TO IMPRESS WITH YOUR DECISIONS? AND IS THAT HELPING OR HURTING?

Every leader has an audience they’re performing for, even unconsciously. Your boss. Your board. Your team. Your peers. Your own self-image. That audience influences your decisions in ways you might not recognize.

And here’s what’s changed: The presence of AI in decision-making creates new performance dynamics. Some leaders feel pressure to appear “data-driven” and “tech-forward.” Others feel pressure to demonstrate that human judgment still matters. Both are performing.

Sometimes this helps. If you’re trying to impress your team with thoughtful, principled decision-making that integrates multiple perspectives, that standard raises your game.

But sometimes it hurts. If you’re trying to impress your board with how “AI-enabled” you are, you might over-rely on algorithmic recommendations even when context demands human judgment. If you’re trying to prove human expertise still matters, you might dismiss valuable AI insights out of defensive pride.

The question isn’t whether you have an audience. Everyone does. The question is whether that audience is pulling you toward better decisions or pushing you toward performance.

Think about your last major decision under pressure. If you’re completely honest, who were you trying to impress? How did that influence what you chose and how you communicated it? Did you lean on “the AI recommended this” when actually it was your judgment using AI as one input? Or did you avoid mentioning AI entirely even when it provided valuable insight, because you didn’t want to appear dependent on technology?

Reflect honestly: Who is your primary audience when you’re making decisions under pressure? Is that audience pulling you toward better judgment or better optics? How does the presence of AI in your decision-making change who you’re performing for?

QUESTION 4: WHAT ARE YOU OPTIMIZING FOR: SHORT-TERM RELIEF OR LONG-TERM RIGHT?

Under pressure, the instinct is to make decisions that reduce immediate discomfort. The vendor is demanding an answer. Give them one. The team is anxious about direction. Announce something. The customer is upset. Comp their meal. The dashboard is showing concerning metrics. Implement the AI’s recommendation immediately.

These decisions feel good in the moment because they resolve tension. But they often create bigger problems later.

And here’s what I’m watching: AI makes short-term relief easier because it provides fast answers. You’re looking at concerning data, AI suggests an action, you implement it immediately, tension relieved. But did you think through second-order effects? Did you consider whether the metric AI optimized for is the right metric? Did you ask whether this solves the actual problem or just addresses the visible symptom?

Research shows that time pressure increases focus on immediate outcomes at the expense of delayed outcomes. Your brain naturally discounts future consequences when current stress is high. And AI, which provides instant analysis, can amplify this tendency by making immediate action feel more justified.

Think about your recent decisions under pressure. How many were optimized for short-term relief rather than long-term right? How many times did having an AI recommendation make you move faster than you should have because “the data supports it”?

Bringing back Endless Shrimp would have provided short-term relief. Customers happy. Media coverage positive. Employees relieved. Dashboard metrics improving immediately. But long-term? Eleven million dollars continuing to hemorrhage. Unsustainable business model reinforced. Eventual collapse just postponed.

Killing it was long-term right despite short-term pain. Initial customer backlash. Critical social media posts. Uncomfortable questions. Metrics temporarily down. But six months later? Finances stabilizing. Resources available for real improvement. Sustainable path forward.

The discipline is choosing long-term right even when short-term relief is screaming at you, and when AI is making short-term action feel data-justified.

Reflect honestly: How often do you choose the decision that makes the pressure stop rather than the decision that solves the actual problem? Has having AI recommendations made you more likely to take fast action because it feels “data-backed”? What would it take to choose differently?

QUESTION 5: IF THIS DECISION FAILS, WHAT STORY WILL YOU TELL YOURSELF ABOUT WHY?

This is the most uncomfortable question, and it’s deliberately forward-looking. Before you’ve even decided, imagine the decision fails. What story will you tell yourself about why it failed?

This reveals your blind spots and your defense mechanisms.

If your story is “The data was wrong,” you’re probably overweighting data and underweighting judgment. If your story is “I knew I should have trusted my gut,” you’re probably making decisions based on intuition without enough rigor. If your story is “The AI recommendation was bad,” you’re probably abdicating responsibility to technology. If your story is “They didn’t implement it right,” you’re probably not considering implementation difficulty when deciding.

And here’s the new defensive story I’m hearing from leaders: “The AI told me to do it.” As if algorithmic recommendations absolve human judgment. They don’t. AI provides input. You make decisions. If it fails, the accountability is still yours.

The story you’ll tell yourself about failure reveals what you’re not adequately considering when you decide.

If Adamolekun’s decision to kill Endless Shrimp had failed, what story might he have told? “Customers cared more about the promotion than the financial analysis revealed, and traffic declined faster than we could replace it with other draws.”

That potential story revealed the key risk: Can we replace the traffic driver with something profitable? That’s what he had to address before committing. And he did, by simultaneously bringing back high-margin favorites and adding new compelling menu items.

By imagining the failure story in advance, he identified and mitigated the actual risk before implementing.

Reflect honestly: Think about a decision you’re facing right now. If it fails, what story will you tell yourself? Will you blame the data? The AI? Your gut? Other people? What does that story reveal about what you’re not adequately considering? How can you address that gap before deciding?

What This Weekend’s Reflection Reveals

If you’ve answered these five questions honestly, you now know more about your actual decision-making patterns than most leaders ever examine.

You know whether you default to gut, data, or AI when they conflict, and whether that default serves you. You know when you use AI as a crutch versus a tool. You know whose opinion you’re unconsciously playing to, and how that affects your relationship with data and technology. You know whether you’re optimizing for relief or right, and how AI enables the relief path. You know what story you’ll tell yourself about failure, and what blind spots that story reveals.

This awareness doesn’t guarantee better decisions. But it makes better decisions possible. Because you can’t change patterns you don’t see.

The leaders who consistently make good calls under pressure aren’t the ones who never feel pressure or always have perfect information. They’re the ones who understand how pressure affects their judgment, and how technology changes that dynamic, and compensate accordingly.

When gut, data, and AI conflict, they choose deliberately rather than reflexively. When pressure makes them want to grab for technology as justification, they recognize the impulse and ask whether it’s truly informing judgment or just providing cover. When they’re performing for the wrong audience, they notice and redirect. When short-term relief calls, they choose long-term right anyway. When they imagine failure, they identify and address the actual risk rather than defending against imagined criticism.

That’s not heroic decision-making. It’s disciplined self-awareness applied to high-stakes choices in a technology-enabled environment.

Damola Adamolekun demonstrated this at Red Lobster. At thirty-five years old, facing bankruptcy, with enormous pressure to act dramatically and decisively, he could have made spectacular decisions that felt good but wouldn’t work.

Instead, he made disciplined decisions. He chose financial fundamentals over sentiment when fundamentals were reliable. He moved fast when decisions were reversible and slow when they weren’t. He communicated to the right audience with transparency rather than spin. He chose long-term sustainability over short-term popularity. He imagined failure and addressed the real risks.

Six months later, the company exited bankruptcy. The turnaround is working. Not because of genius. Because of discipline.

That’s the lesson. Under pressure, you don’t need to be smarter. You don’t need better AI. You need to be more aware of how pressure affects your judgment, and how technology changes that dynamic, and more disciplined about compensating.

These five questions give you that awareness. Use them this weekend. Then use them Monday when the next pressure decision hits, and you’re staring at conflicting inputs wondering which one to trust.

Because the gap between how you think you decide and how you actually decide is where leadership breaks down. Close that gap, and everything else gets easier—even in a world where AI adds another voice to an already noisy decision-making process.

AI is the context. Leadership is still the work. And the work starts with understanding how you actually make decisions when it matters most.

🧠 The Mindful Leader

Weekend reflection for leaders who want to grow. Every Friday, questions that challenge your assumptions and prepare you to lead differently on Monday.

By Robert Adams • 30+ Years in Foodservice Leadership • EVP, UniPro Foodservice • Marshall Goldsmith Stakeholder Centered Coach

🎧 LISTEN INSTEAD

A Student of Leadership Micro Podcast

Powered by UniPro Foodservice.

📬 SUBSCRIBE TO ALL THREE NEWSLETTERS

Monday: The Leadership Table (Deep frameworks)

Wednesday: Breaking Bread (Daily practices)

Friday: The Mindful Leader (Weekend reflection)

Same weekly theme. Three distinct approaches. Real leadership growth.