Nothing Changes If Everything Changes

The Only Room Without a Screen

I had an idea in the shower the other night. That's not unusual—it happens a lot, actually. The cliché is real.

Nobody was pinging me. I wasn't toggling between tabs or checking what had just come in. It was just me, the water, and whatever my brain decided to do with ten uninterrupted minutes.

The thought wasn't fully formed. It didn't arrive as a thesis. But it connected to something that had been nagging at me, and by the time I got out I could feel the shape of an idea emerging. 

Then I started brushing my teeth—and immediately grabbed my phone to write it down. Couldn't even make it to the next room.

I couldn't tolerate thirty more seconds of uninterrupted thought.

Which is its own kind of punchline. The shower gave me the space to think. I couldn't wait to escape it. 

Why the Shower Works

There's actual science behind this. Researchers call it the default mode network—a set of brain regions that activate when you're not focused on a specific task. It's where your brain connects ideas it has no business connecting.

The conditions are surprisingly narrow. Total rest doesn't trigger it. Neither does intense focus. What lights it up is a task that requires just enough engagement to keep you present but not enough to consume your attention. Showering. Walking the dog. Driving a familiar route.

The shower is one of the last environments where most of us aren't reachable. No screen, no notifications, no input stream. You're doing something physical and habitual enough that your brain has nothing to fight for.

We've accidentally built a creativity lab in the one room we go to get clean.

Learning to Drive

When I was learning to drive, everything felt too fast.

I remember taking a curve with my mother and brothers in the car and feeling as though events were arriving faster than I could process them. It wasn't that the road was difficult. I simply hadn't developed fluency yet.

It wasn't that I was a bad driver. I just hadn't calibrated to the speed yet. Everything was happening faster than I could process. I was reacting to each moment individually instead of reading the road ahead.

That's what early driving feels like. Every intersection is its own decision, every lane change requires conscious thought. The task isn't that complicated. You just haven't built the ability to process it. 

Later, once the mechanics become automatic, you stop reacting to each moment individually and start reading farther down the road. You can drive faster and handle more because you're not spending cognitive effort on the basics anymore. You've absorbed the fundamentals. You've metabolized the speed.

I think about this a lot right now, watching how people and organizations are adopting AI. 

The tools are fast. The people using them are still calibrating. And the gap between the speed of the tool and the fluency of the user is where mistakes compound—not because anyone is doing it wrong, but because they haven't had time to get comfortable at the current pace before the pace increases again.

People are operating tools at speeds they haven't metabolized.

Too Many Variables

In any well-designed experiment, you change one variable at a time. That's the whole point. You hold everything else constant so you can see what the change actually did. If it worked, you know why. If it didn't, you know where to look.

Many organizations are currently running the opposite of a controlled experiment. They're changing tools, workflows, team structures, reporting lines, vendor relationships, and internal processes and redefining expectations—simultaneously. All in the name of AI transformation.

And then someone asks: is it working?

Maybe it is.

The changes might be fine. But when every variable is moving simultaneously, it's almost impossible to know which one is impactful. You can't attribute the outcome to any single decision, can't separate what helped from what was irrelevant.

That's not transformation. That's turbulence with a strategy deck.

In Awareness Is the First Casualty of Convenience, I wrote about how automation replaces noticing—small decisions you used to make consciously get absorbed into systems until you forget they were decisions at all. Same problem, wider aperture. When you change everything at once, you lose the ability to notice what any single change is doing.

What Sticks and What Doesn't

Even when a change works, it takes time to stick. A new tool doesn't become part of how a team operates the day it's deployed. It becomes part of the team’s operating model when people stop thinking about whether to use it. When it moves from conscious choice to default behavior. That transition can't be rushed.

Transformation isn't adoption. Adoption is its own challenge, but it's the part people plan for. Transformation is what happens after—when the new way of working has to survive contact with pressure, deadlines, ambiguity, edge cases, and the accumulated habits of everyone involved.

That's why sequencing matters.

In Fast Isn't the Point, I wrote about the difference between speed and rhythm—how the real skill isn't moving faster but knowing when to move. In transformation, it's how deliberately you sequence the changes so each one has time to take root before you pile on the next.

The challenge isn't simply moving quickly. It's allowing one change to become durable before introducing the next.

The Shower and the Spreadsheet

Last week in The Limits of Magic, I wrote about how constraining ChatGPT made it more useful—how the magic and the failure mode are the same mechanism.

The shower is the same idea, running quietly in the background of your evening (or whatever time of day works for you).

Think about what stays constant in a shower. The water, the temperature, the room, the routine. None of that changes. The only variable that moves is what's happening in your head. That's why it works. 

The environment creates the conditions to notice a single thing changing.

Your brain gets to run one process without competing against everything else for resources.

That's what a controlled experiment looks like. You hold everything still except the thing you're trying to learn about.

Most organizations right now are doing the opposite. Every variable is in motion. New tools, new workflows, new structures, new expectations—all at once. And when the results arrive,  nobody can confidently explain which change produced them.

It's not that the changes are wrong. It's that there's no control group. There's no constant to measure against.

Constraints aren't the thing slowing you down. They're what make learning possible.

The principle isn't to do less. It's to hold some things still long enough to see what moving one thing actually does. Not every transformation can proceed sequentially. Competitive pressure rarely allows that luxury.

But the more variables you move at once, you change everything and learn nothing.

The shower is a ten-minute experiment with a single input: you. Most organizations are running a hundred experiments at once and wondering why they can't tell what's working.

Berkson's Bits

I have a customer service joke. Would you like to wait to hear it, or would you like me to let you know when I'm available to tell you?

What I'm Listening To...

Scientists have been studying the human body for a long time. Still, they find new things. About 8 years ago they identified a network that is described in this Radiolab episode as “ the interstitium, a vast network of fluid channels inside the tissues around our organs that scientists have just begun to see, name, and understand.” 

The study was published in 2018, this podcast is from 2023, and I just listened to it this month. There’s always something new to learn. 

The shower isn't magic. It's just the one of the last places where the conditions for thinking haven't been optimized away.

Looking forward to continuing the conversation...

Alan