Leading the Product in the Vibe Code Apocalypse

It's 10:47 AM on a Tuesday, your product manager has mocked up a new onboarding flow—"took me 20 minutes with Claude, looks promising." Your designer shared a radically different checkout experience—"AI helped me explore some wild ideas, this could be huge." Your head of growth just pinged you a landing page variant—"ran it through Cursor, we should test this immediately."

Each looks compelling. Each one claims to solve a real customer problem. Each team member is genuinely excited about their creation and wants your blessing to "just quickly test it with users."

Welcome to the new reality of leadership in an AI-accelerated world.

The conventional wisdom tells you this is fantastic—cheaper experiments, faster iteration, more shots on goal. Just say yes to everything and let the data decide. But here's what nobody talks about: you're drowning in the overwhelm of uncertainty, messy strategy, and the pressure to win. And not helped by three new competing “solutions” on a Tuesday morning, while your engineering team is already stretched across two other "quick tests" from last week.

The uncomfortable truth is that AI hasn't just lowered the cost of experimentation—it's fundamentally changed what leaders need to be good at.

It's not about speed anymore. It's about wisdom when everyone can build anything, anytime.

Leadership muscles that matter now more than ever

While everyone's obsessing over prototype velocity, the real challenge is happening one layer up—How do you make good decisions when the volume of testable ideas has exploded 10x overnight?

While we can 10x what is built, we can't 10x our customers' attention. “Experiment on everything” is not a strategy, because attention is precious and customers don’t forgive quickly when we waste theirs. They’ll make up their mind once. When you're sketchy product launches too early and misses the mark, it’s so much harder to get them back in the room a second or third time. For that reason, better alignment, judgement, and decision-making up front will always be important. The explosion of vibe-coded prototypes makes good product thinking more valuable, not less. Here’re some muscles worth building:

Strategic experiment alignment

Amazon doesn't just run thousands of experiments because they can—they run them because each one ladders up to a clear hypothesis about customer behaviour or business performance. Every test has a role in strategy deployment—from tactical tweaks at the interface layer, all the way to launching entire businesses. Many companies are running experiments in isolation, hoping something sticks.

Organisational design for abundance

Remember when you had one Product Manager for every 8 engineers? That ratio is shifting fast—think more like 1:3 now. That's because the pace of testable ideas requires more product thinking, more strategic filtering, and more decisions. Some thinking and decisions take time—and as well they should. You need thinkers who build, and builders who think. This mindset is more critical now than ever before—and 'builder' doesn't just mean engineering.

Resistance to shiny object syndrome

This is where most leaders fail. When every new idea can be prototyped in minutes, everything becomes promising and urgent. The discipline to hold your strategic line—to say no to compelling prototypes that don't serve your mission—has become even more critical than it always has been.

While AI has made experiments cheaper, it's made good leadership more expensive. You need more strategic thinking, not less.

The Pivot Trap

I've learned this multiple times, always the hard way. Once in a pre series-A startup, exceptional team, killer product velocity. We were building fast, shipping faster, but couldn’t generate enough customer traction. Then the shiny objects started multiplying.

New market opportunity here. Compelling sales requests there. A prospective partner doing something interesting that we "should probably explore." Each idea felt urgent. Each pivot felt strategic. Each course correction seemed logical in isolation.

What I didn’t realise—we were slowly killing our strategy. Every direction change chipped away at our conviction about the core challenge we were solving. Every "quick exploration" pulled focus from the deeper platform work that actually mattered. Every prototype kept us dancing at the interface layer instead of building the foundations.

We were slowly killing our strategy. Every direction change chipped away at our conviction about the core challenge we were solving... every prototype kept us dancing at the interface layer instead of building the foundations.

Building a product or platform takes time and requires discipline. You've got to be confident about the core challenges you're solving—and most of the time, they’re different from the more immediate and obvious things we might see at the interface layer with prototypes. We kept getting seduced by surface-level fixes when the real solutions lie deeper in the stack.

The fatal mistake wasn't any single decision—it was the pattern:

• Reacting to market pressures over strategic missions
• Succumbing to shiny objects over investing in foundation steps
• Pivoting too hard and too fast over properly considering platform strategy

It's hard to get this right, even when blessed with a small, very high-performing team doing exceptional work. Now imagine that same leadership challenge amplified by AI: 10x more prototypes, from different groups, with different agendas, all arriving in your Slack at the same time.

Do more right things with a Decision Stack

So how do you navigate this new reality without losing your strategic mind? You need something that helps you focus experiments in the right places, evaluate success and failure in the context of your broader strategy, and resist the siren call of every compelling prototype.

Enter The Decision Stack—a mindset and a tool for product leadership that works because it fits naturally with rapid experimentation. Leading thousands of experiments with all kinds of teams over a couple of decades of customer development taught me just how hard it is to achieve strategic alignment at all altitudes of a business.

I use this not because it’s better—all models are wrong. I use it because it’s useful for making strategy actionable, communicating intent, and providing practical guidance for decision making in the field. In other words, more of the right things happen, better decisions get made, and everyone is on the same page. That means we can 10x our experimentation with AI while moving in the same direction. Here's how:

Strategic Principles

Principles establish how you'll operate. They're not abstract values that sounded good at the off-site. They’re practical "even over" statements that guide real decisions when you're under pressure from investors, co-founders, or market opportunities that "we absolutely must explore right now."

These principles become your North Star when everyone has a persuasive argument and vibe prototype.

Signal vs Noise Filters

Most importantly, the Decision Stack gives you language and structure for what matters and for who. Whether you're evaluating a co-founder's pivot proposal, a designer's interface prototype, or an investor's market suggestion, you can ask:

• Does this change our understanding of the core challenge we're solving?
• Will this provide learning that is relevant to our strategy?
• Is this signal about customer behaviour or just market noise?
• Are we being asked to dance at the interface layer when the real work is deeper?

It's an antidote to drowning in data and opinions that look important but don't inform decisions that matter.

This is not about saying no to everything—it's about having the strategic clarity to say yes to the right things, at the right time, for the right reasons.

The choice ahead

AI has handed every leader a superpower—the ability to test almost any idea, almost instantly. But AI superpowers without wisdom are just expensive ways to make mistakes faster.

The leaders who thrive in this new reality won't be the ones who run the most experiments—they'll be the ones who ask the best questions about what those experiments actually mean. They'll build cultures that can handle abundance without losing focus. They'll have the courage to kill ideas that don't serve the mission, before they get to prototype.

Most importantly, they'll remember that the goal isn't to experiment—it's to learn. And learning requires the discipline to dig deeper than the interface layer, the wisdom to distinguish signal from noise, and the strategic clarity to stay focused when everything feels urgent.

Your Tuesday morning doesn't have to be chaotic. With the right strategy and mindset, those three competing initiatives become strategic choices, not overwhelming distractions.

The question isn't whether you can afford to experiment more. It's whether you can afford not to experiment better.