It’s a strange feeling to use a tool every day and have no idea if the person who built it is even alive.

A few years ago, on a massive, fast-moving project, we stumbled across this tiny piece of software that solved a very specific problem our main system couldn’t fix. The software had no branding, no real company footprint, no onboarding flow—just a download link, a bare little window, and a seven-day trial that ticked like a kitchen timer.

We figured we’d test it, buy a license, move on.
Except we couldn’t buy anything.

The upgrade button didn’t work.
The website felt abandoned.
The “contact” page might as well have been decorative.

We tried emailing. Nothing. Not a bounce, not an out-of-office—just radio silence. Meanwhile, the project kept barreling forward, so every week we did a clean wipe, reinstalled the trial, and kept going, like this was a normal part of modern software distribution.

And somewhere between the third reinstall and the tenth unanswered email, I realized this wasn’t a weird one-off.

If something this small could turn us into non-customers, then what frictions had I built—on my website, in my pitch, in my process—that were doing the same to someone else?

Once you see friction in the wild, you start seeing it everywhere.
The tiny stuff.
The stuff teams assume is harmless.

Some time ago, we reached out to a law firm for representation. Straightforward inquiry: here’s who we are, here’s what we need, are you taking clients? They got back to us six months later. No apology, no explanation—just a reply as if the thread were warm.

And then there was the message I sent in April of 2020—right in the middle of COVID—about a professional opportunity with real budget behind it. Something they could’ve benefited from if they’d answered within the same geological era.

Two years later—May 2022—I finally got a reply:

“Hi Graig, can you call me tomorrow?”

They misspelled my name.
They missed the moment.
They missed the entire world in which that conversation could’ve existed.

That’s the part people underestimate: when you wait too long to show up, you don’t just lose the deal—you lose the context the deal depended on.

People shift. Needs shift. Urgency shifts.
By the time a delayed reply lands, the story has already moved on.

The more I paid attention, the more I realized the biggest leaks in business rarely come from big mistakes. They come from things so mundane you barely register them.

A checkout page that makes you create an account before you can buy.
A mobile site that looks fine on desktop but breaks the moment someone taps a menu icon.
A contact form that errors out and never tells the sender.
A calendar link with no openings for three weeks.
An onboarding flow that assumes everyone knows your terminology.
A pitch that insists “anyone can use this,” which is usually a red flag that no one specific will.

People love to talk about funnels, but most funnels don’t leak from the top—they leak from the seams no one is monitoring.

And let’s talk honestly:
Most founders hate marketing.
Most of us want the product to “speak for itself.”

I spent the majority of my life posting basically nothing on social media. Then the last six months hit, and I’ve posted more on LinkedIn and Substack than in the previous decades combined. I didn’t suddenly become extroverted. I finally accepted that if I don’t make myself visible, I can’t expect anyone to know I exist.

People can’t walk through a doorway they can’t see, and they won’t go hunting for one.

Visibility = availability.

This applies just as much to brick-and-mortar businesses as online ones.

How many restaurants have you walked past because the entrance looked closed even when it wasn’t?
How many stores have you skipped because the hours on the window didn’t match the hours on Google?
How many small businesses never get the chance simply because their phone number goes to a full voicemail box?

And how many times has someone tried to hire a freelancer or service provider and hit a paywall of friction: outdated portfolios, dead links, vague offerings, unclear pricing, no sense of who the work is actually for?

Everyone wants more customers.
Very few people make it easy to become one.

That’s the realization the seven-day trial forced into focus.

The product was amazing. We had the budget to license it for the entire project run and then some. But they lost us because everything around the product made reaching them feel like trying to catch fog.

And when I traced that feeling back into my own work, it wasn’t fun.

I started asking myself questions I’d conveniently avoided:

How many people bounced from a page I thought made sense but actually didn’t?
How many mobile users gave up because my site wasn’t optimized?
How many potential clients needed clarity on what we offer, and I left it implied instead of explicit?

The deeper I sit with this, the more I think most businesses misjudge where trust is built. They assume it’s in the product, or the pitch deck, or the pricing grid. But trust usually forms earlier—long before substance, long before features—at the moment someone decides whether you seem easy enough to engage with.

Most companies don’t need a better product.
They need fewer escape routes for the people already trying to get in.

If I could go back to that year-long seven-day trial, I wouldn’t change the tool at all. I’d change the doorway. I’d add a human. I’d give the creator a calendar link, a working form, a single button that leads somewhere real.

People don’t wait in hallways.
They don’t jiggle locked doors.
They don’t assume you’re worth the trouble.

They move on—to whatever is easiest, clearest, closest, or simply present.

Showing up doesn’t guarantee the opportunity.
But not showing up almost always kills it.

And most of the time, you won’t even know what you lost.

It was one in the morning.
I was in bed, running loops through my head—markets, strategies, flipping variables.
I wasn’t tossing or turning, just wired, trying to see the pattern inside the noise.
I wasn’t chasing an answer so much as listening for one.

And then it hit me—an idea so wild it didn’t belong in a notebook.

I picked up my phone and texted my co-founder:

“You’re going to hate me. But you’re going to love this. Come over tomorrow.”

In business, you can often tell in an instant who’s ready—and who isn’t.

A few months earlier, I’d gone to a networking mixer and posted about it on LinkedIn. Unlike my usual feed, it attracted a new crowd—founders, operators, people who actually build things.

One connection request stood out. On paper, he was impressive: co-founder of a fast-growing platform I’d started seeing in ads and Reddit threads. The kind of pedigree that usually comes with a bit of gravity—investors, press mentions, panels, all of it.

So when he reached out, I thought, interesting—this could be worth a conversation.
Then I opened his message.

“Saw your post.”

That was it. No curiosity. No bridge. No invitation.

I actually went and looked him up again, just to make sure I wasn’t missing something—maybe he was busy, or there was context I didn’t catch. I read about his company, skimmed their case studies, looked at what they were building. It was good work. It made sense.

So I decided to do the decent thing. I wrote back. Not a throwaway line, but a proper reply—mentioned his company, what I’d been working on, and how some of our work might overlap. I gave him an easy opening.

His reply came quickly:

“Love to hear that.”

And that was it.

No follow-up, no question, no spark.

That’s when it clicked: he’s not ready for business.
Because business isn’t about status—it’s about curiosity.

The best operators I know treat curiosity like oxygen. It’s what keeps the conversation alive.
The moment it’s gone, the deal dies. And maybe that’s why the people who stay curious the longest start to look a little unhinged—because they’re still asking questions long after everyone else has moved on.

My co-founder came in the next morning.
He arrived, calm as always, and we went through the usual pleasantries. He walked over to the table where we usually sat and worked and pulled out a chair.

That’s when I stopped him.
“You’re gonna want the couch.”

I didn’t plan it. I just started talking.
Logic, trajectories, imagination.
Extrapolation, storytelling, conviction.
One idea unfolded into the next, each piece locking into place faster than I could process it. I wasn’t pitching anymore—I was building the thing in real time.

He just sat there, listening, nodding, occasionally throwing in a question or two—small objections, small pivots.

I was shocked at how long I’d been talking.
No one had ever just let me go like that before—just sit there and listen while I built an entire world out loud.
Maybe I was borderline manic, but I was convinced. I believed in this idea more than I had believed in anything else.

And that’s when I realized something else.
I finally understood what crazy was.

I wasn’t just the guy with the crazy idea.
I had become the crazy one.

Because I’d just done something I swore I’d never do.

A few years earlier, when I was briefly in real estate, I attended a Q&A session between the brokerage’s CEO and his chief of staff. They were trading stories about how they worked together—how they’d learned each other’s rhythms, how they’d built this shorthand you only get from running at the same speed for years.

At one point, the chief of staff laughed and said he used to get texts from the CEO in the middle of the night. Out-of-nowhere ideas. Half-formed strategies. Random flashes of clarity. Half the room laughed; the other half groaned.

I remember thinking: that’s insane.
But also… kind of amazing.

There was something about the rawness of it—the pure velocity of someone who couldn’t wait until morning to think out loud. I admired that energy. I thought, one day, I’d love to have that kind of fire.

But I also promised myself I’d never do that to someone else.

And then, years later, there I was—doing it.

Eventually, he bought in.

I don’t text him in the middle of the night anymore. That was a one-time thing.

But it taught me something. The people we call crazy are often just the ones thinking a few frames ahead—too fast for the rest of us to recognize the pattern yet.

The truth is, every industry has its version of that 1 a.m. glow—the moment when curiosity outruns consensus. Most people shut the laptop and promise to revisit it later. A few can’t. They chase it. And that chase is what eventually redraws the map for everyone else.

And if you’re lucky, every once in a while, you catch a glimpse of what they’re seeing.
Long enough to follow it.
Long enough to build it.
Long enough to realize that maybe being called crazy is just the cost of paying attention before everyone else does.

After all that, I finally asked my co-founder, “Am I crazy?”
He didn’t even pause.

“Yeah,” he said. “But in the best way.”

The first thing that caught my attention wasn’t lights, or signs, or someone talking — it was , tapping my shoulder.

“Now that’s good marketing,” he said. And pointed.

We were halfway through the trade show floor, surrounded by a desert of a hundred booths that looked almost identical. But this one stood out immediately: one clean sentence, no noise, no slogans about disruption. You didn’t have to think. You just knew what they did. Out of hundreds of companies, they were one of maybe three we remembered.

It was my first time at the Javits Center — thousands of people outside, lines for badges looping around the lobby. The Wi‑Fi barely worked. We lost connection to the floor plan ten minutes in and just started wandering.

What were we looking for? We couldn’t tell you. It wasn’t our job to have the pitch, it was theirs.

But the further we walked, the more impossible it became to tell one company from another. Rows of glowing screens. Sharp suits. Free tote bags. Every message bending toward the same promise:

AI for creators.

The future of media.

Next-gen… something.

The words repeated until they started to dissolve. Everything shimmered, and nothing stayed in focus.

I wanted to engage. I wanted to be pulled in. But nothing invited me. Every message was either too vague or too loud. Each booth seemed optimized for spectacle, not understanding. Some were barebones — a laptop on a folding table, a sheet of printer paper taped to the wall with the company name in black text. Others had massive LED walls, candy bowls, giveaway tables, looping demos. Different extremes, same result: you’d stop for a second, squint, and still have no idea what they did.

You could feel how much money and effort went into all of it — the travel, the staff, the design work. But so much of it still vanished.

I saw the hard truth: most teams aren’t bad at what they do. They’re bad at showing it. They’ve lost the ability to tell when something isn’t landing.

And I realized we weren’t immune to that either.

When we were building the landing page for our first enterprise software product, we thought we’d nailed it. Clean visuals, clear structure, strong copy. We were proud of it. But when we showed it around, even people already using our tools struggled to explain what the company actually did. They understood the product, but not the point.

That disconnect hit the same nerve as the trade show — standing in front of something full of potential that somehow communicates nothing.

It’s an easy trap to fall into. When you’re deep in your own work, it’s hard to remember that most people aren’t. You speak in shorthand, assume context, and convince yourself that a clever phrase will carry the meaning. It rarely does.

We didn’t need another round of polish. We needed to start over, to say what we did as plainly as possible, and then say it again until it stuck.

So we did what we always do when something feels off: we rebuilt it. Line by line, section by section, until it started to breathe. We tested copy, rewrote headlines, pitched the idea out loud, and listened for the pause — that tiny delay where someone still needs to translate. The goal wasn’t to be clever, it was to be grasped in one glance.

We ran it by users, friends, even family. Could they tell what we did in ten seconds? If not, back to the whiteboard.

That became our new baseline. If a potential customer has to think too long, they’re already gone. The world moves too fast for confusion. You only get a few seconds of attention, and in those seconds, you either earn trust or lose it.

When the new page finally clicked, we felt it instantly. People could describe us back to us. They understood the service, the value, and the outcome, all before scrolling. The conversation that followed was faster, smoother, and more honest. You could sense it in the tone of every call after that.

That’s the part most people miss. Clear communication isn’t just about marketing; it changes the relationship. When someone knows what you do right away, the rest of the interaction starts from alignment, not confusion. Every decision downstream gets easier.

Over time, that became a habit. Every week, we block time to review how we’re showing up — our site, our decks, our emails, even our internal docs. We treat communication like part of the product itself, not decoration. Clean language makes sales faster, onboarding smoother, and client relationships steadier. It saves time on the back end because it removes friction on the front. We keep tuning until it feels right in the gut.

Most people don’t notice when something’s off; they only notice when it fails. But we’ve learned to trust that early discomfort that says, this isn’t landing yet.

And that’s what most of those booths were missing: the awareness. They’d spent everything trying to appear and never realized they’d gone invisible.

We left after one day. There wasn’t any more to learn.

The pattern was clear: so much motion, so little meaning.

You don’t need to know when it’s perfect.
You just need to know when it isn’t, and keep moving until you find something real.

I’ve always had a low tolerance for preventable problems. Things that don’t need to be difficult but somehow always are. Forgetting an umbrella. Leaving lights on. Realizing the heat’s been blasting all day while I’m out. None of it’s catastrophic, but it adds up to a background hum of inefficiency that quietly eats your time.

It started with wet socks.

Not metaphorically — literally. I was stepping out of a grocery store in Manhattan when the rain came down like a punishment. Everyone else popped umbrellas in one motion, but I didn’t have mine. I froze, holding two paper bags already starting to wilt. The wind turned sideways, the street overflowed, and within seconds my shoes made that miserable squelch that stays with you for hours.

If you’ve ever been caught in a New York downpour, you know the feeling. It’s not just being wet, it’s being defeated. You can’t sprint home without soaking your jeans, and no amount of scaffolding saves you. By the time I made it back to my apartment, I looked like someone who had fallen into the Hudson River fully clothed.

That night, as I blow-dried my shoes, I promised myself I’d never feel that stupid again.

I didn’t even know where to begin. I wasn’t a developer. I didn’t want to write code. I just wanted a digital version of that Post-it note you leave on the door — don’t forget your umbrella. Somewhere in a forum thread, I stumbled onto Make.com, a no-code automation tool that promised to connect different apps like puzzle pieces. I was cautious, but it felt approachable enough to try.

A few nights later, I had my first working prototype.

The logic was simple, at least in theory:

If the smart lock on my apartment door unlocks, then check the local weather.
If the chance of rain is over 20%, send me a push notification: “Grab your umbrella.”
If the temperature drops below 60°F, ping me again: “You might need a jacket.”

Getting it to work was a different story. I connected the wrong modules, broke the flow, retried. I remember sitting there at midnight watching the Make.com dashboard like a heartbeat monitor, hoping that when I unlocked the door, the system would finally respond. And when it did, my phone buzzed.

“Grab your umbrella.”

I laughed out loud. The code was janky, the data barely synced, but it worked. A tiny miracle. A problem, solved invisibly.

From there, I couldn’t stop.

I moved the setup into Home Assistant running on a Raspberry Pi, and started building more logic on top of it. With one press of the TV remote, the lights dim and the curtains draw themselves closed. Plugging my phone in at night becomes a “goodnight” command — lights off, thermostat down, silence. When my partner leaves the office, I get a small ping on my phone: time to start cooking dinner.

Friends come over and shake their heads. “You’ve lost it,” they say, laughing as the curtains move on their own. But then they stop and ask how it works. Because underneath the absurdity, there’s a kind of elegance. A home that responds, not performs.

I won’t lie: some of it got out of hand. I automated so much that I started automating things I didn’t even need. The system would flick the hallway lights on if it detected motion — even if that motion was my cats. I learned the hard way that automating without purpose doesn’t create freedom. It creates noise.

Not everything should be automatic.

That realization became its own quiet philosophy: find the real pain points, not just the possible ones. The goal isn’t to make everything effortless. It’s to make the right things disappear — the small, predictable frictions that steal focus and attention.

That shift from “make everything automatic” to “make only the right things automatic” became the foundation for how I build systems today.

Now, when I work with teams, I see the same pattern everywhere, their own versions of the umbrella problem. Small, constant frictions that make good work harder than it needs to be. Most teams don’t need new tools, they need invisible systems that clear a path for people to do what they’re already good at.

So we build them.

An employee checks a box, and a real-time Slack notification alerts the next person — no one waiting, no one guessing. When a deliverable goes out, execs who need to know are notified instantly — no chasing, no “just checking in.” Documents that used to take hours to create, sort, and label now generate automatically — clean, uniform, finished in seconds. And for the leaders who need to see the signal through the noise, we build daily briefings, surfacing exactly what matters that morning, and nothing else.

These aren’t magic tricks. They’re acts of respect.

Automation, done right, isn’t about replacing people. It’s about removing the parts of work that make smart people feel stupid.

When you build it right, it feels like the lights dimming as the movie starts — effortless, expected, right on time.

Not too long ago, I was standing on the 70th floor of One World Trade Center at a Make.com mixer. Floor-to-ceiling windows stretched across the skyline, the entire city laid out below in perfect grids of light. It was one of those rare moments where you can see the system you’ve been living inside — the movement, the structure, the potential — all at once.

I caught my reflection in the glass and thought about that soaked version of myself outside the grocery store two years earlier, the one trying to solve a $5 problem.

The same curiosity that wanted to stay dry now powers entire systems that keep teams from drowning in their own noise.

I didn’t plan that arc. I just followed the irritation until it turned into something useful.

Solve for one.
Solve what hurts.
That’s how scale actually begins.

It was my first tapeless delivery — and one of the first MTV had ever received.
Everyone was figuring it out. Half the industry still worked in tape, the other half stumbling through digital delivery like it was a new religion nobody understood.

Two specs landed on my desk.
Both labeled official.
Both written by different people.
Both wrong in different ways.

The tape spec required one kind of slate — the tapeless spec required a completely different one.
Same space. Conflicting instructions. Both required.

So I made a professional call.
I used my judgment.
I built a single combination slate that covered both — every piece of information either spec could possibly need.
The show was technically perfect. The delivery clean.
But the leader didn’t match the tape spec exactly.

Rejected.

Anyone who’s ever worked in post knows a rejection isn’t the end of the world. You fix it and move on. It happens all the time.
But that’s not what happened here.

The production company called — relaying the network’s fury from above.
The person on the other end told me the show had been rejected and that I’d “messed up.”

The network executive wanted blood.
She demanded someone be fired — over a slate.
It was absurd.
Everyone in the industry knew deliveries got bounced all the time; it was routine. But this wasn’t about the show anymore — it was about blame.

And because I was young and desperate to prove I could handle it, I did the worst thing you can do when people are looking for a scapegoat:
I volunteered.

“If it’s that bad,” I said, “I’ll resign.”

There was a pause on the line. Not compassion — calculation. The kind of silence you hear when someone realizes how far they can push you.

Then came the verdict.

“Don’t ever use your best judgment again,” she said. “Do exactly what the network says.”

That’s when I realized the system didn’t actually want competence.
It wanted obedience.

I was volunteering nights and weekends, trying to be indispensable to a system that didn’t care about my sanity, my time, or even my insight.

Fine. I thought. I’ll do exactly what I’m told.

The Second Call

A day later, they needed promo materials.
This time, I made sure everything was in writing.

Please confirm exactly how you want this delivered, I wrote.

The reply came back almost instantly — confident, definitive, and completely wrong.
Every line contradicted common sense.

I even asked, Are you sure?
“Yes.”

End of discussion.

So I followed it exactly.
No interpretation. No correction. No fixing what I knew was broken.
I did exactly what they told me to do.

And it failed again.

The second call came from the same production contact.
She was frustrated — running on the panic that comes from being stuck in the middle.

“The network says it’s wrong again. What happened this time?”

I said, calmly,

“I did exactly what you told me to do. Exactly what they confirmed. You told me not to use my best judgment, and I didn’t. I followed instructions. Now it’s wrong again.”

It got quiet on her end — that half-second where someone realizes what they’ve actually asked of you.
Then her voice changed. Softer. Honest.

“You’re right.”

I didn’t blame her.
She didn’t write the specs. She didn’t understand the technical side.
All she knew was she’d gotten a furious phone call, and that the relationship with the network was on the line.
She was doing her job.
And I respected her for that.

When she saw how absurd it was — how I’d been yelled at for thinking and then again for not thinking — she stood up for me.
That almost never happens in this business.
That moment stayed with me.

The Theater of the Call

The network’s executive producer called a meeting: me, the EP, a coordinator, another manager.
They decided to read through the spec with me, line by line, over speakerphone.
Like a classroom punishment.

They were reading the wrong spec.

I let them go for a few pages before interrupting.

“You’re reading the wrong version,” I said. “The correct one says something else.”

Shuffling. Silence.

I walked them through the contradictions: line by line, note by note, which one was newer, which one contradicted the other.

Finally someone muttered,

“No one’s ever pointed that out before.”

No. Shit.
No one reads it. They just forward it and hope the next person takes the hit.

That was the moment I understood the real job.
It wasn’t delivering shows.
It was absorbing confusion so everyone else could pretend the system worked.

The show eventually passed.
No apology. No thank-you. Just the next episode on the calendar.

I stayed late that night — not to prove anything, just to sit there and process it.
I’d been right twice, and it hadn’t mattered either time.

After that, I made my own rules:

• If specs conflict, flag it once, in writing.

• If they confirm something wrong, follow it exactly — and save the email.

• Never argue twice.

• Never expect the apology.

I cared because I was scared — scared to lose my job, scared to be called difficult, scared that thinking would get me punished again.

Years later, I still remember her voice on that second call — tired, honest, human.

“You’re right.”

She didn’t have to say it. It didn’t fix anything.
But it mattered.

Because she could have doubled down, but she didn’t.
She stopped. She listened. She gave me back my dignity with two words.

That stayed with me.

Now, when I’m the one on the other end of the line —
when someone younger, quieter, or lower on the chain tells me something I don’t want to hear —
I try to remember that moment.

And I say it out loud.

“You’re right.”

Not because it fixes everything.
But because someone needs to hear it once.
Because that’s where leadership actually starts.

I was watching The Studio on Apple TV the other night, a few hours after the news broke that Catherine O’Hara had passed away. The timing felt uncomfortably precise. I’ve been watching O’Hara my entire life. She spent it playing women who understood the performance even as they were trapped inside it — characters sharper than the worlds that needed them, often reduced to spectacle anyway. I’ll never forget how exact her comedy always felt. The timing, the control, the sense that every gesture was chosen. Even at her broadest, she made the work feel deliberate, as though the joke itself knew more than it was letting on.

The Studio operates in that same register. Recently renewed, it treats modern Hollywood as a system that keeps mistaking continuation for survival. It’s a love letter wrapped in satire, a peek behind the curtain of an industry that keeps trying to resuscitate itself through noise, leverage, and denial.

Bryan Cranston plays the studio head, a man who cares nothing for art and everything for outcome. He’s the embodiment of a machine that confuses activity with vitality — endlessly reassuring itself that profit is proof of life.

In the episode, the studio is hosting its big upfronts — that yearly ritual when distributors and theater owners gather to be sold on the new slate of movies.

It’s a circus, but a joyful one.
Movie stars crowd the stage, trailers play in perfect rhythm, speeches swell with just the right amount of hope.

The audience isn’t faking it; they’re having fun.
For a few minutes, everyone remembers why they came into this business in the first place.

And that’s when I hear it, tucked behind the raucous applause: a melody that pulls me straight out of the moment.

My ear perks up like a reflex. What is that?
The rhythm is impossible to mistake. Is it?
And then I laugh, quietly but audibly. Oh my God, it is.
Followed by the inevitable realization: Oh my God, this must be so expensive.

“There’s a Great Big Beautiful Tomorrow…”

They chose it because it fits.
The characters have just pulled off something improbable, and the song seals it — an anthem for survival disguised as a jingle about optimism. For a brief moment, everyone in that fictional theater believes again: that movies still matter, that audiences will still show up, that creativity isn’t dead yet.

The Carousel of Progress

Walt Disney’s 1964 love letter to automation — the ride he designed for the World’s Fair, later moved to Disneyland.

A rotating theater, one family living through each new age of invention: the electric age, the jet age, the imagined home of the year 2000. The family never changes; they simply move forward through time, learning, adapting, smiling.

The optimism is unbreakable.
The future is assumed.

That song, as earnest as it sounds, wasn’t written as propaganda. It was a declaration of faith.

Walt believed automation could end drudgery, not work.
He believed that progress was supposed to buy people back their hours — to relieve them from the mechanical and monotonous so they could spend more of life on what makes it worth living.

Machines, in his mind, were meant to be servants of imagination, not substitutes for it.

He was a capitalist and a complicated one — anti-union, obsessive, ruthless about his standards — but he understood reciprocity. His dream depended on a middle class healthy enough to afford leisure, to take a week off, to buy a ticket to the future he was selling.

He knew you can’t sell wonder to people who can’t afford rest.

That’s what made Carousel of Progress so powerful — it wasn’t just optimism, it was also a promise.

Progress was circular: technology lightened the load, people had time to dream, those dreams created new industries, and the circle began again.

And audiences back then understood that promise. They didn’t fear the machinery; they believed it was there to serve them.

You can see that in Pirates of the Caribbean, the other attraction Walt was building near the end of his life. Those animatronic pirates — rowdy, singing, endlessly repeating their scenes — weren’t a cost-cutting trick.

Walt didn’t automate because he didn’t want to pay performers; he automated because he knew no human could keep that rhythm forever. The actors would tire, their voices would crack, the energy would fade.

The pirates were never meant to replace people — they were built to hold the repetition so that people could stay inventive.

Automation, in Walt’s hands, was an endurance strategy for creativity.

Nobody complained that the pirates had been replaced.
Everyone understood the difference between drudgery and artistry, between the endless motion of work and the human spark of making something new.

That’s the part we’ve lost.

Somewhere between the optimism of Carousel and the automation of today, progress stopped circling back.

We kept the efficiency but forgot the mercy.
We built machines to save time, and then we spent that saved time measuring profit instead of restoring people.

Now AI arrives with the same pitch Walt once sang: shorter workdays, more creativity, a great big beautiful tomorrow.
But shorter for whom? For the worker, or for the owner?
And if it’s for the worker, are they still paid for five?

That’s the real test of progress — whether the prosperity returns to the people who made it possible.

Walt would have understood that immediately.
He wasn’t afraid of automation; he was afraid of forgetting what it was for.

The problem isn’t the technology.

Left unguided, automation follows gravity — toward consolidation, cost reduction, and the quiet erosion of its own audience. Guided with intent, it can still do what Walt meant it to do: carry the weight so people can carry the meaning.

What’s different now is who holds the tools.

For the first time, automation isn’t locked behind corporations and capital. The machinery is no longer hidden inside studios, factories, or boardrooms.

Ordinary people can touch it.

The abstractions are thinning. The systems are learnable. You don’t need permission to design work that removes drudgery instead of removing people.

But that also means the responsibility has shifted.

If automation erases dignity now, it isn’t because it had to.
It’s because someone chose not to learn how to wield it with care.

Learning these systems isn’t technical ambition anymore.
It’s literacy.

If you don’t understand the tools shaping work, someone else will decide what they’re for — and they will optimize for outcomes, not humans.

That’s not a future imposed on us.
It’s one we opt into by disengaging.

When I hear that song now — “There’s a great big beautiful tomorrow, shining at the end of every day” — it doesn’t sound naïve to me. It sounds like a design constraint.

A reminder that progress doesn’t justify itself.
It has to be aimed.

Automation will keep advancing whether we participate or not. The only real choice left is whether we learn enough to guide it — or leave the compass to people who only measure cost.

Walt didn’t build machines because he trusted technology.

He built them because he trusted people — if they were given back their time.

That’s still the bet.

And it’s one worth learning how to make again.

Not sure how AI actually fits into your business?

We offer a focused AI business readiness consultation to evaluate where AI can safely create leverage — and where it’s more likely to introduce risk or noise.

Normally a $500 engagement.
Currently offered at $250.

→ Get the AI Business Readiness Consultation

Everyone keeps asking the same AI question:

“Which one should I use?”

ChatGPT. Claude. Grok. Whatever just showed up in the news this week.

The assumption underneath that question is that there’s a right answer—and that if you pick the wrong one, you’re somehow falling behind.

For most businesses, that framing is already wrong.

AI Is Not a Competitive Sport for Most Businesses

If you run a real company—services, professional work, trades, real estate, accounting, legal, construction, client services—AI is not something you “win.”

It’s not a leaderboard.
It’s not a badge of sophistication.
And it’s not something you should be chasing release by release.

What you’re actually deciding is simpler—and more consequential.

You’re deciding what you want your business to depend on.

The Part No One Wants to Say Out Loud

Most business owners don’t actually want AI strategy.

They want relief.

They want:

• fewer things slipping through the cracks

• fewer follow-ups that only happen because someone remembers

• less mental overhead tracking work

• fewer mistakes caught only by double-checking

AI gets pulled into the conversation because it sounds like leverage.

Sometimes it is.

But here’s the uncomfortable truth:

If your operation already relies on people “keeping an eye on things,” AI does not fix that.

It accelerates whatever system you already have.

Loose systems get looser—faster.
Unclear systems scale confusion.
Manual oversight becomes automated oversight.

That’s why so many people try AI, feel excited for two weeks, and then quietly stop using it.

The problem was the business, not the model.

Why the AI Arms Race Isn’t Your Job

The major AI models are locked in an arms race.

They leapfrog each other constantly.
What looks “better” today will be matched or beaten quickly.

The differences people argue about online—benchmarks, edge cases, micro-advantages—rarely show up in normal business use.

Unless AI is literally what you sell, obsessing over which model is smartest is like a contractor obsessing over which brand of calculator has the best buttons.

It’s not the work.

What actually matters is the company behind the model:

• Will they still be around?

• Will pricing suddenly change?

• Will access tighten?

• Will behavior shift underneath workflows you’ve already built?

That’s the real risk.
Not output quality.

Personality Problems Aren’t Intelligence Problems

A lot of people say they’re “done with ChatGPT” because it’s annoying.

Too agreeable.
Too chatty.
Too much “you’re right to call this out.”

That’s fair.

But it’s not a sign the model is dumb.
It’s a sign you’re seeing defaults.

Most people never adjust instructions or constrain tone. They judge the entire system by the voice it speaks in.

That voice will change.

What doesn’t change is whether your business has a place for the output to land.

If this is useful, subscribe.

If it’s helpful to you, subscribing tells us to keep going.

The Question You Should Be Asking Instead

Instead of asking:

“Which AI should I use?”

Ask this:

What part of my business only works because someone is watching it?

If the answer is “a lot,” then AI isn’t your bottleneck.

Structure is.

AI works when there’s something clear to plug into:

• a defined workflow

• a known handoff

• explicit ownership

Without that, it becomes just another layer on top of the same manual oversight you already had.

That’s where most disappointment comes from.

How We Think About This Ourselves

In our own work, we’ve invested heavily in one ecosystem—not because it’s perfect, but because it was stable enough to build around at the time.

We monitor whether that continues to be true.
If it stops being true, we’ll change course.

That’s not brand loyalty.
That’s operational judgment.

The goal isn’t to pick right forever.
It’s to avoid anchoring your business to something fragile without realizing it.

If You’re Just Poking Around

If you’re early or experimenting:

• pick one

• use it lightly

• learn where it helps and where it breaks

Don’t confuse experimentation with progress.

But if you’re trying to use AI to actually run parts of your business—intake, follow-ups, approvals, handoffs—the model matters far less than whether the system underneath it can hold weight.

That’s the part most people skip.

And it’s why so many “AI initiatives” quietly die after the novelty wears off.

Not sure how AI actually fits into your business?

We offer a focused AI business readiness consultation to evaluate where AI can safely create leverage — and where it’s more likely to introduce risk or noise.

Normally a $500 engagement.
Currently offered at $250.

→ Get the AI Business Readiness Consultation

I could feel the room before I knew the reason I’d been called into it.

A narrow side office. No windows. Carpet that swallowed footsteps. The kind of place managers use when they want the conversation to feel smaller than it is.

The door closed with that soft, final click — the one that tells you the outcome has already been decided.

The manager sat across from me, hands folded, expression flat. I stayed standing a second too long, realizing the geometry of the room wasn’t an accident: wall behind me, desk between us, nowhere to go.

Then they said it. I had been “stepping on toes.”

The words came out calm and deliberate. The kind of accusation that doesn’t invite a defense.

I hadn’t raised my voice. I hadn’t missed a deadline. I hadn’t ignored direction. I was doing what I’d been hired to do: keep things moving, fix what was broken, make sure the work didn’t stall.

But in that cramped room, under fluorescent light, I was being punished for it.

The rest of the conversation blurred: a steady loop of implication without specificity. I remember nodding, saying little, waiting for the moment when they’d finally let me leave.

When I finally stepped back into the hallway, I felt a sick, sinking realization that I’d seen this before.

Early in my career, I was still figuring out how offices worked — the titles, the politics, the invisible lines that decided who mattered. I showed up early, stayed late, said “yes” to everything. That’s how you learn, or at least how you think you do. Every task felt like a small audition for belonging.

The woman who hired me ignored all that. She saw potential, not polish. She didn’t gatekeep. In an industry where people often hoarded information to make themselves indispensable, she did the opposite: she shared everything she knew. She gave me resources, walked me through what I didn’t understand, and never made me feel small for not knowing. She wanted everyone to succeed.

Her office was the heartbeat of the floor. People came and went constantly: assistants, editors, producers, even execs. It was a revolving door, but not in a chaotic way. It was the place to be. Everyone who passed through seemed to have their own shorthand with her — an inside joke, a shared story, a nickname only she used.

Then, one morning, during what should have been a routine meeting, her name came up in passing.

“She is no longer with the company.”

No pause. No explanation. Said like a calendar update.

The day before, she’d been in her office laughing about a rough cut. The next, she was gone — erased with the same tone used for delivery deadlines.

When the meeting ended, I went straight to a coworker’s office, someone who’d been close to her. I wanted to believe it wasn’t true.

But when he looked up and met my eyes, his face fell.
That was all the confirmation I needed.

A few minutes later, the same manager who’d made the announcement called me in again.

Yes, she was no longer with the company — and someone needed to pack up her office.

That someone was me.

The door was half open when I got there.

Her chair was slightly turned, a sweater still hanging on the back. Papers on the desk, perfectly aligned. A half-finished soda beside a stack of notes.

It looked like she’d just stepped out for a moment. But the boxes were already waiting by the door.

I started folding her things carefully, trying to make order out of something that didn’t feel real. Every sound felt amplified: the rip of tape, the shuffle of papers, the hollow click of a drawer closing.

People passed by in silence. Some slowed down, looked in, then kept walking. No one said a word. Without her, the room felt wrong. Too bright, too empty, too clean.

Six years of work fit into four boxes.

As I went through her shelves, I saw the life she’d built there. Her whiteboard was covered in doodles and cartoon sketches from late nights. Small jokes only the people who’d been there would understand. Her drawers were full of keepsakes: a photo from a wrap party, a birthday card from an intern, a rubber duck someone had left on her desk as a joke.

What struck me most was what wasn’t there. No plaques. No framed awards. No reminders of rank or recognition. Just proof that people mattered to her more than credit did. You could feel her leadership in the details: human and real.

When I said she’d been an incredible boss, someone laughed. They said I only liked her because she gave us “too much time off.”

That was the moment I understood how often leadership gets mistaken for leniency, and how easy it is for power to confuse empathy with defiance.

Looking back, the two moments were the same test.

First they removed the person who led with trust. Then they watched who would still act like her.

Every organization has its rituals for enforcing obedience. Some use meetings. Others use silence. Ours used boxes.

“Stepping on toes” wasn’t rebellion. It was the quiet act of trying to keep things working after the person who taught us how had already been erased.

I stayed long enough to see the pattern repeat — good people leaving quietly, the loudest ones failing upward, the rest of us pretending it all made sense. The message was always the same: stay small, stay useful, stay out of the way.

When I eventually led my own team, I kept thinking back to her.

How she wasn’t above getting her hands dirty. How she’d plop down at an empty workstation to help if something needed to get done. She’d earned her title, but she never used it as distance.

That kind of leadership doesn’t need to announce itself. You just feel safer doing good work around it.

She gave me a chance. They made me pack up her office.

And somewhere between those two acts, I learned exactly what kind of work is worth protecting — and what kind never deserves to be protected again.

I wasn’t looking for a business lesson. I just wanted my cat to feel better. What I got instead was a masterclass from the most unlikely of places.

I like to think of myself as patient, but I’m really just optimized. I choose restaurants that let me book online. I’ll skip a doctor if their scheduling portal looks like it was built before touchscreens. I’ll pay extra for a product that ships same-day, not because I need it faster, but because the wait feels like uncertainty. Efficiency feels like control, until something small reminds you that you don’t actually have any.

When I adopted my cat, Marble, he was already a full-grown gentleman, two years old and solid. A soft, lumbering, fifteen-pound feline with the kind of presence that makes furniture look small. I was still getting to know him, trying to decode the little signals that tell you when a creature trusts you.

A few weeks in, he started showing hints that his digestion wasn’t keeping up with his appetite. Nothing serious, just enough to throw off his rhythm, and mine with it. The internet said to try a fiber supplement — dried pumpkin. Simple enough.

A local pet store had recently opened nearby. Small, independently owned, the kind of place you want to root for. Their website looked polished, their product listings detailed. They even offered online ordering for pickup. Perfect.

I ordered the dried pumpkin late that night, paid online, and went to bed feeling like I’d solved a small problem responsibly.

The next morning, I kept a close watch on my inbox. By noon, still nothing. I refreshed a few times, checked my spam, then my bank app. The payment had gone through. Each refresh felt a little more pointless. It was such a small errand, the kind you plan between other things, and somehow it was already becoming work.

Four hours into the business day, the email arrived:

“Your order is ready for pickup.”

When I walked in, the place was smaller than I expected — about the size of a one-bedroom apartment. Brightly lit, spotless, completely still. One employee behind the counter, scrolling his phone. I’d expected some bustle after waiting that long, but it was just quiet.

On the counter next to him sat a single bag of dried pumpkin. No tag, no bag, no receipt. Just sitting there, like it had been waiting for anyone who looked remotely interested.

When I said I was there for a pickup, he smiled, reached for the bag, and — as if to fill the silence — started talking about the other flavors they carried. Pumpkin with cranberry, pumpkin with sweet potato, all stocked and ready to go. I nodded politely, still expecting him to hand me my order. Instead, he rang it up and asked me to pay.

And then I felt that small internal sigh you get when the easy thing becomes the hard thing. I’d done everything right: ordered ahead, paid online, waited patiently. And here I was, still explaining myself over a bag of pumpkin. The kind of moment that makes you think, why is nothing ever just easy?

“I already did,” I said. He looked unsure.

I pulled up the confirmation email on my phone and turned it toward him. “Online order, last night.”

He squinted, nodded, then apologized softly, but still uncertain. I could tell he was new, that the system was new. But there was no reassurance, no “you’re all set.” He just handed me the bag, like it was easier to let me leave than to figure out what went wrong.

I thanked him, because that’s what I do, and walked out. The door shut with that dull retail click — the sound of a place returning to its own silence.

By the time I reached the sidewalk, I knew I wouldn’t go back.

Not out of spite — just instinct. My first impression was sealed, a quiet “no” written in permanent ink.

That store is the one that made me wait for nothing. The one that made me explain a purchase I’d already completed. The one that turned an easy errand into effort.

I never meant to memorize that, but I did. And I’ve told that story more times than I ever expected — to my partner, to friends, now here.

The whole thing lasted five minutes, but the impression will last forever.

It didn’t hit me until later — how fast I’d written them off. How frictionless it is to abandon something you almost believed in.

We all do it. We call it convenience, but it’s really trust.

The less a system makes you think, the more you believe in it. The more you believe in it, the more you spend, return, engage.

That’s the physics of friction — it compounds invisibly. Every extra click, every delay, every awkward handoff creates drag. The system doesn’t need to fail dramatically; it just needs to remind you that you’re working harder than it is.

When I look at software now — my own included — I see what I’ve started calling The Pet Store Principle: you only get one first impression — don’t make the customer work for it.

A signup form that doesn’t autofill. A password reset that takes too long. An onboarding flow that assumes enthusiasm instead of anxiety.

Every founder says they’re building trust, but most are just building features. They add when they should remove. They assume the customer will understand the logic instead of feeling the friction.

And just like that store, they lose people not with a blow, but a shrug.

No one writes in to complain about a small delay or a confusing interface. They just never come back.

What stays with me about that afternoon isn’t the mistake — it’s how forgettable it was. That’s the danger. Bad experiences don’t always explode. Sometimes they dissolve quietly, like trust evaporating in slow motion.

The cashier probably forgot about me ten minutes after I left. But I still remember him — the quiet store, the bag of pumpkin sitting there like a metaphor that hadn’t revealed itself yet.

That bag was the whole business model in miniature: good intentions, poor execution, no second chance.

I wanted to believe in that place. But every time I pass by, I think of the delay and the confusion at the counter.

That’s what friction does — it rewrites the story for you.

It turns a small act of care into a quiet lesson about systems: you only get one first impression.

You won’t just lose a future sale. You’ll become the story your customer tells instead.

Once that happens, you never get them back.

I have an odd relationship with Lost in Yonkers.
Every few years I put it on again for reasons I can’t quite explain. It’s Neil Simon—you already know what you’re getting. A little too neat, a little too clever, but when you’re in the right mood, it lands perfectly. The rhythm, the timing, that faint ache of honesty that hides inside all his comedy.

And of course, there’s that line—the one that escaped the film and wandered into popular culture:

“You got moxie, kid.”

Everyone’s heard it, even if they couldn’t tell you where it came from.
For years, I never really understood it. It always sounded like something your uncle would say after you changed a tire. “You got moxie, kid.” Sure. Okay.

But there’s this moment in the movie—set in the dining room. Artie, the nephew, says to his uncle, “What’s moxie?”
And Dreyfuss puts down his bag, walks over, rolls up his sleeves, shifts his stance, gives a little head tilt and shoulder pop—almost a come and get it kind of look—and says,

“That’s moxie.”

And at the end, after everything’s gone sideways and somehow right again, he’s driving off into the night, calling back to that same boy who’s just stood his ground against the mobsters:

“You got moxie, kid!”

That’s when it hits you.
Moxie isn’t swagger.
It’s the decision to face what might eat you alive—because standing still would kill you slower.

Early in my career, I was working at a production company that, for reasons above my pay grade, was already half-doomed.
It was the end of the line for me there. The air had gone sour with one of the executive producers—whom I may or may not have written about before—and I was quietly plotting my exit.

My direct boss, though, was the opposite. The kind of man you never forget. A little bouncy, a little old-school, the kind who printed out every email he ever got. I think he just liked something he could hold. It made sense to him.

I was already the kind of person who gave up weekends for “special projects.”
Young, eager to prove myself, eager to please—that was just who I was back then. I was probably twenty-five, maybe twenty-six, trying to get my first real post-supervisor job.

So I was in on most weekends, tinkering, trying to get noticed.
That’s when I saw it: an email printed out and left on his desk.

“We need someone young and eager. The pay’s not great.”

I recognized the name of the person who’d written it—a producer who was known for working on weekends. He’d usually ask how to get into a project, or how to get access to some drive. And I’d always say, Don’t worry about it, I’ll be there, I’ll take care of it.

So when I saw that email, I waited to hear which weekend he’d be in again. When he mentioned it, I said, “No problem, I’ll cover.”
But I wasn’t covering. I was planning.

I structured the whole day around that meeting. I didn’t want to look like I’d been waiting.
I let him come in, get comfortable, settle into his rhythm.
Then I established mine—back and forth, in and out of the server room, down the hall, back again. Just a regular pattern, nothing sudden.
I wanted him to see me as part of the environment. Background. Expected.

And right around lunchtime, the bay door of his edit room was open. I walked by on one of my “rounds,” did a little swing-by, stopped, turned, leaned into the doorway, and said,

“Hey.”

He looked up.
That’s when I asked,

“How do you think someone gets their first post-supervisor job when they think they’re ready?”

That was it. No pitch. No résumé. Just the right question, dropped in the right ear.

He paused, thought for a second, and said, “You know, I might know someone.”

A two weeks later, I was gone—off to my first big supervising job.
Just before I left—maybe the day after I resigned—my boss came by my desk, smiling.

“So,” he said, “you read the email on my desk.”

That’s when it hit me.
He’d known the entire time. He’d left it there for me.
He gave me the permission I didn’t know I was waiting for.

That’s moxie.
Not arrogance. Not manipulation. Just the nerve to act when the universe leaves the door cracked open—and the grace to thank the person who left it that way.
At the time I didn’t think of it as moxie. I just thought it was guts—balls, really.
But maybe that’s the same thing, just wearing a younger man’s face.

Lost in Yonkers ends the way all Neil Simon stories do—half-hope, half-ache.
That’s what moxie really feels like. Not triumph, just survival with some style.

When I think back, I realize I didn’t have experience or capital or certainty. I had moxie—the calm conviction that I could figure it out, even if I had to invent the way as I went.

So maybe moxie isn’t attitude after all.
Maybe it’s moral physics—momentum with a soul.

“You got moxie, kid.”

Yeah. Maybe I finally do understand what that means.

The first thing he ever said to me was, “I can’t help you.”
The last thing he ever said was, “What about me?”

Everything in between was proof that some people don’t build teams—
they drain them.

By the time I left television in early 2018, I was done.
I was exhausted.

The industry was showing early signs of collapse, even if not everyone could feel it yet.
I did. I saw what was coming—shrinking budgets, longer hours, pay that never moved.
And that’s why I started to leave.

Real estate looked like a way out.
A broker who’d sold apartments to people in my family invited me to join his team at a new type of brokerage.
He gave the perfect pitch: Join my team. I’ll show you everything. You’ll learn fast. I’ll be your mentor.

That last word—mentor—was the hook.
It’s what I’d been chasing my whole career: someone to see the raw parts of me and help shape them.

So I walked away from a steady six-figure job and started over.
I showed up on my first day eager, pressed suit, polished shoes, ready to begin.
I stood there waiting for him to extend his hand and show me around—
not just the office, but the life I’d been promised.

He didn’t.
He barely glanced up.
Then he said, “I can’t help you,” and turned back to his screen.

That was it.
No welcome, no guidance, no map.
Just a line that landed like a door slamming in my chest.

A few minutes later, one of the younger agents — younger than me but the closest in age — walked over.
He looked me up and down, not unkindly, just… puzzled.
“You left a six-figure job for this?” he asked. Then, after a pause:

“I need to take you out for a drink and tell you what you just signed up for.”

It wasn’t a congratulations. It was a warning.
A prophecy, really — one I didn’t understand yet.

I went home that night and sobbed.
The sound startled even me.
Out loud, to no one, I said, “What did I just do?”

That was the moment the floor dropped out.
Every promise of mentorship vanished the second he’d gotten what he wanted—my signature, my slot on his team.

Weeks passed, and the picture filled in.
His splits were brutal—far above the norm.
Every commission bled upward; every bit of credit belonged to him.
He sold opportunity, but the product was extraction.

The team itself was a revolving door.
Faces appeared, smiled tightly, and disappeared within months.
The rule was simple: feed his ego or get out of the way.

The only real contact any of us had with him was the weekly “sales meeting.”
He demanded everyone’s presence like it was church.
We’d file in early, coffee in hand, waiting for the sermon to begin.
He’d stride in with that clipped, overconfident cadence, prepared with his “lesson of the day.”

But it was never about teaching us anything.
It was about proving how right he always was.
Every meeting was a performance: Hear my voice. Watch my success. Bask in it. Now get the fuck out.
He’d talk at us for half an hour, maybe forty minutes, then dismiss the room like he was doing us a favor.
It wasn’t mentorship—it was maintenance of control.

And the people sitting in those chairs made the dysfunction visible.
There was one woman on the team who was sweet, polite, and a great cook.
She was also so catastrophically bad at selling that she couldn’t have moved an apartment if people were paying her to take it.

Then there was the youngest guy—thin, blond, conventionally attractive, always performing confidence like a costume.
He carried himself like success was a birthright, chest out, all surface and zero substance.
Every conversation punctuated with lines about his “FAs”—that’s what he called his financial advisor buddies—

“They’re just crushing it, man. They’re all absolutely crushing it.”
Eventually, he went back to join them. Just another bro in a different uniform.

And then there was the guy who thought he was the star: maybe closed one or two deals a year, but he ran a local real-estate Facebook page with a few thousand followers and believed that made him untouchable.
He especially hated that I lived in his neighborhood—like I was trespassing on his little digital fiefdom.

After a few months, when a handful of those early team members had finally had enough and left, I watched him bring in two new agents. They were sharp, confident, already managing large rental portfolios of their own.
I saw him through the glass office doors, doing the same rehearsed handshake—the firm nod, the practiced smile—and I thought, They’ll be gone in weeks.
They were already making too much money to tolerate him.
And sure enough, they vanished even before I did.

The first real crack in his façade came in an ordinary moment.
I walked into his office—his private glass box where only his administrative staff usually sat—and looked him dead in the eye.
A client’s mother had died.
I told him directly, expecting a flicker of recognition, a brief pause, something human.

One of his staff members nearby gave a soft, instinctive sigh—the kind of sound people make when they hear loss.
He just stared at me.
Blank.
No movement. No change in his face.
Nothing.

I held his gaze, waiting for a beat that never came.
Then I looked at the person who’d reacted, as if to confirm that yes, we were both standing in this strange vacuum together.
It was shocking—but somehow not surprising.
By then, I think I already knew.

That was when I finally understood what those sales meetings were really about.
You can’t teach connection if you don’t have any.

That realization sat in my chest for weeks like a bruise.
I stopped asking him questions.
I stopped expecting guidance.
I started counting the exits.

A few weeks later, after the illusion had completely fallen, I was giving him an update on a deal I was working on.
It wasn’t about the client—it was about the other agent, the opposing broker on the deal.
I was explaining how I’d had trouble trusting them, how they’d said one thing and done another.

I looked at him and said it out loud, like a small piece of borrowed wisdom:

“People tell you exactly who they are. Believe them.”

He nodded, pleased, like he agreed completely.
“Of course they do,” he said, and reached out his hand.

I didn’t take it.

Instead, I told him we needed to talk.
I pulled him into another office, closed the door, and told him I was leaving.

He didn’t ask why.
He didn’t ask what I was doing next.
He didn’t even pretend to care.

He just looked at me and said, “What about me?”

That was the bookend.
Everything between those two sentences was just proof.

I walked out shaking—part fury, part relief—but steady for the first time in months.

A few months later, word spread through the office.
The brokerage—still in a feverish hiring spree—had asked him to leave.

People couldn’t believe it.

They fired him?
During expansion?

He must have been unbearable.

The gossip didn’t sting; it validated everything I’d already lived.
He didn’t build teams; he consumed them.
He’d even recruited people like me just to trigger sign-on bonuses the brokerage was offering to team leads.
Once the bonus cleared, our usefulness was over.

Hearing he was gone didn’t make me angry.
It just made sense.

When I left television, I thought I was escaping a collapsing system.
Instead, I ended up watching a smaller one implode in real time—
one man, perfectly built for the old order, finally running out of people to drain.

I used to believe success came from finding the right mentor.
Now I know it comes from becoming the person you needed back then.

He taught me nothing about real estate.
But he taught me the cost of selfishness.
And from that, I learned the only rule worth keeping:

The best way to find the help you need is to become the person who would have given it.

Each week in , we show how operators design, build, and run systems — from architecture and workflows to the decisions behind them.

Subscribe now

Using AI to Compete in a Crowded Marketplace

If you’re reading this, you likely already use AI in your work. You’re probably above average compared to the general public. But that pool is getting crowded, and the advantage now comes from understanding how to use the tools at a higher level than the people around you.

We’re going to walk through the actual method we use to build productive, stable AI projects. Not hacks, not prompt packs, not hype. Just the mechanics that let you get more out of the tools than most people ever will.

This piece teaches the most important mental shift you can make — something you can apply immediately, today, before you learn anything else.

The First Win: Stop Approaching AI Like Deterministic Software

Most people assume AI behaves like traditional computing:
input → output → repeat.

That assumption is wrong.

Language models are probabilistic. They adapt to context, tone, structure, and intent. They infer far more than they’re explicitly told. And the place where they make the biggest inference — the place where they decide what the entire session is — is the first message you send.

If you start casually, the model becomes casual.
If you start vaguely, the model becomes vague.
If you start with unstructured language, the model will invent structure — usually the generic kind you’ve seen everywhere and dislike immediately.

It’s the result of the environment you unknowingly created.

AI Is Not Just for Coding

A lot of people limit themselves because they assume AI is mainly for code generation or quick rewriting. That belief shrinks what they attempt. But AI is useful in any domain where structured thinking, explanation, summarization, comparison, reframing, or diagnostic reasoning are helpful.

• Operations

• Marketing

• Brand development

• Market research

• Documentation

• Strategy

• Product articulation

• User research

• Internal communication

• Conflict de-escalation

• Workflow design

The list isn’t the point.
The point is that the model’s usefulness extends to anywhere you think, not just where you type syntax.

The reach of the model is large.
How much of it you access depends entirely on how you begin.

Once you understand how that framing works, AI moves from being a novelty to being a genuine part of your workflow.

Why the First Message Is the Environment

The model does not interpret your first message as a question. It interprets it as orientation. It uses the opening to determine:

• the type of work being done

• the tone and role it should assume

• the expected depth

• the level of formality

• the acceptable deviation from user phrasing

• the structure (or lack of structure) it should use going forward

This is why the early versions of this essay sounded generic.
We opened a fresh project with no system instructions, no voice print, and no constraints. The model filled the gap with broad, average English. Once we corrected the environment, the writing shifted.

How ChatGPT 5 / 5.1 Interpret Input

With Chat 5 and 5.1 specifically, these models weight the start and the end of your message most heavily.

• The opening defines the kind of work being done.

• The closing defines the constraints that override everything else.This is why we end messages with structured context blocks wrapped in pseudo-XML:

<context>
  [instructions, constraints, references]
</context>

It cleanly separates the operational content from the conversational content. The model respects what’s inside the block because structurally it cannot mistake it for prose.

This isn’t a trick. It fits how these models actually process input.

Why Prompt Packs Don’t Solve This

Static prompts degrade with each model update.
They work until the next release and then become misaligned.

More importantly, they don’t teach you how to shape the environment.
Without that skill, you’ll always get the model’s defaults — the very patterns you’re trying to avoid.

Everything in these tools depends on one thing you control:

the room you put the model in.

The Mental Model Shift: Your First Message Is the Project

Don’t start a session by asking AI for output.
Start by telling it what kind of work you are doing and what environment you expect it to operate in.

That single adjustment will improve your results immediately, even before you know anything about system instructions, voice prints, or structured reasoning.

The opening message is the part most people overlook.
It’s also the part that determines the entire trajectory of the work.

A Practical Example: Removing “AI-Generic” Through Environment Fixes

Here’s a real example pulled from the earlier drafts of this essay.

What the model originally produced:

“This is not a failure of the model. It’s the result of the environment you unknowingly created.”

Structurally, this is accurate.
But the phrasing uses an extremely common AI pattern:
X is not the problem — Y is.

You’ve seen this everywhere. It’s overused, generic, and easy for a human reader to detect as AI-written. Even though the underlying idea is correct, the construction makes the writing feel automated.

The corrected version:

“It’s the result of the environment you unknowingly created.”

One line removed.
Nothing lost.
Everything reads cleaner.

This is how you challenge the model’s defaults:
you locate the pattern you don’t want, you tell the model what specifically violated your standard, and you ask it to rewrite within the constraints of your voice.

We did this again with another pair of lines:

Original:

“The limit is not the model’s capability. The limit is the initial framing you give it.”

Again: correct idea, generic structure.

Corrected:

“The model is capable of far more than most people attempt. What you get depends on the framing you start with.”

Same meaning.
No X/Y contrarian scaffold.
Human, not synthetic.

The difference isn’t rhetorical.
The early drafts lacked constraints and voice definition, so the model reached for statistically common constructions. Once we defined the boundaries, the model stopped reflexively using those patterns.

This is your immediate win — right now, today.
Start training yourself to identify these generic constructions, and you will be able to shape the model’s output far more precisely.

“AI Slop” and Why This Work Matters

You’ve probably seen the term “AI slop.” The criticism isn’t wrong. You can identify the patterns once you’ve worked with these models long enough: the flattening, the repetition, the softened language, the indistinct phrasing. Most people are producing it because they never pressure the tool and they never pressure themselves. They copy a block of output, paste it into their workflow, and call it done.

That isn’t how this works.

Using AI does not remove the need for effort. You still have to read what it produces. You still have to look for drift. You still have to identify the artifacts that make the writing feel automated. The work is in those small corrections. That’s what gives the output a human edge. You can’t skip that step and expect to stand out.

If you want to differentiate yourself in a market that is now full of people using AI, this is how you do it:
you build the environment, you challenge the model, you check the output, and you refine. That is the gap between people who plateau with the tool and people who move ahead with it.

Whether you’re trying to grow a business, compete in a tighter talent market, expand into new roles, or increase your capacity, the advantage comes from how well you use the tool — not from the fact that you use it.

Start here.
Build the right mental model.
Put in the work that everyone else avoids.
The people who don’t will fall behind.

Share

Leave a comment

Each week in , we show how operators design, build, and run systems — from architecture and workflows to the decisions behind them.

Subscribe now

This Is the Highest-ROI Move You Can Make Right Now

The fastest way to get real operational lift from AI—lift you can feel in the very next message—is to write System Instructions. The actual behavioral framework your model will think inside.

You can draft the first version in under fifteen minutes and immediately see a different class of output: sharper, denser, more disciplined, and less improvisational.
It’s simply the difference between a model trying to please the median user and a model forced to operate under the rules you define.

Most people never feel the absence of System Instructions because the defaults are polite enough to pass as “intelligent.”
But the moment you ask the model to execute a multi-step build, maintain internal logic across sections, or work across days without losing context, the cracks appear.
It guesses.
It fills gaps that needed escalation.
It smooths your work instead of sharpening it.
It forgets the shape of your project.

System Instructions fix all of that.
They teach the model how to think for a specific project, which in turn forces you to articulate what you’re actually doing—with a precision that makes your work more legible to both humans and machines.

This is why the ROI is immediate.
The clearer the framework, the better the model performs, and the more clearly you understand what you’re building.

What System Instructions Actually Do

System Instructions define the behavioral posture of the model: how it escalates ambiguity, how it interprets constraints, where it stops, what it refuses to assume, how it guards density, and how it detects drift.
They turn the model from an improviser into an operator.

The reason most people don’t use them is simple: the interface never demands them.
You can chat with a model for a year without realizing the entire interaction is happening on the default operating system—the same one used for poems, recipes, tech support, college essays, and corporate emails.
That framework is fine for one-offs.
It collapses under real work.

Once you define your own, the model no longer optimizes for friendliness, readability, or “engagement.”
It optimizes for the constraints you specify.
And it finally treats gaps in your articulation as structural blockers rather than creative invitations.

This is the part that surprises most builders: writing System Instructions doesn’t just change how the model operates—it changes how you think.
You’re forced to describe your audience, your density expectations, your escalation rules, your drift boundaries, and your referential logic.
This articulation clarifies the project for everyone involved.

Get It Wrong on Purpose, First

The first real step is not “write perfect System Instructions.”

The first step is forcing the model to show you what its defaults look like when pointed at your project—so you can see exactly what you’re fighting against.

System Instructions are always attached to a specific context. They’re not generic rules for “being helpful.” They are rules for this project, this series, this audience.

For this article, we started by opening a fresh Project in ChatGPT and giving it a short description of what the Nitty Gritty series is supposed to be. Nothing mystical. Just a clear paragraph a human operator would understand:

“The Nitty Gritty is for builders who want to see the wiring — the schemas, formulas, and blueprints pulled straight from production. Each piece shows the real logic behind the decisions we make, the systems we design, and the work that actually ships. This isn’t theory or performance; it’s the fingerprints, trade-offs, corrections, and reasoning pulled from live builds, exposed without filters.”

That’s the first move you should make in your own project:
open a new Project, write a single paragraph that describes what you’re doing, who it’s for, and what the work is supposed to expose.

Once that context was in place, we gave the model one simple instruction:

“Draft system instructions for writing Nitty Gritty articles.”

We knew the first draft would be wrong. We needed it to be wrong.

Because that first failure tells you exactly how the model tries to solve your problem before you constrain it.

The Iteration Arc

From that one sentence, we had the model produce three distinct failures before it gave us something structurally usable.

The point of walking through them isn’t to admire the prose. It’s to show you:

• how the defaults behave,

• how we pushed back, and

• how you can run the same loop on your own work.

Iteration 1 — The “Helpful Internet Writer” Default

The first draft looked like this:

“Write in a friendly, supportive tone to keep the reader engaged. Use simple language and analogies. Summarize key points to reinforce understanding…”

Exactly what you’d expect from a model trained to keep the median user comfortable.

The assumptions baked into this draft were obvious:

• The reader needs reassurance.

• Accessibility matters more than precision.

• Repetition and summaries are a default good.

• “Engagement” is a higher priority than exposing the wiring.

For this series, all of that is wrong. Our readers are operators. They don’t need uplift, hand-holding, or performance. They want density and reasoning.

Our response back to the model was not polite nudging. It was a direct correction of the failure mode:

• We removed friendliness, simplification, analogies, summaries, and engagement tactics.

• We told it the audience does not need to be protected or inspired.

• We told it to focus on ambiguity handling, drift control, escalation behavior, and reasoning discipline.

In your own project, this is where you start doing real work:

1. Look at the first draft and name the failure mode in plain language.

2. List what is unacceptable given your audience and use case.

3. Tell the model exactly what to stop doing, and what behavior it should prioritize instead.

You are not editing sentences. You are correcting the model’s picture of the job.

Iteration 2 — The “Corporate Policy Manual” Swing

Once we stripped out “friendly internet writer,” the model overcorrected in the other direction:

“Maintain neutral, readable communication. Ensure clarity and accuracy. Maintain flow between sections…”

This draft looked like a corporate comms style guide. It was still thinking like a writer trying not to get in trouble, not like an operator bound by constraints.

The failure mode shifted, but it was still a failure:

• Neutrality instead of posture

• “Readable communication” instead of behavioral rules

• Flow and aesthetics instead of constraint enforcement

Our response this round:

• We removed every stylistic preference.

• We told the model that System Instructions must govern behavior, not presentation.

• We made it explicit that tone is a side effect of constraints, not a goal.

Again, your move in your own project is the same:

• Strip out anything that sounds like “be pleasant,” “be neutral,” or “maintain flow.”

• Replace it with rules about what the model should do when it hits ambiguity, conflict, or missing information.

Iteration 3 — The First Functional Skeleton

On the third draft, the shape finally started to match the job:

“Treat ambiguity as a blocker and request clarification before proceeding. Identify drift and escalate when constraints conflict. Avoid assumptions and anchor reasoning to user framing…”

This was the first time the model was talking in terms of behavior instead of vibes. It still wasn’t complete:

• No density rules

• No explicit paragraph discipline

• No clear audience definition

• No wiring ethos

• No safeguards against sliding into PMC / thought-leadership voice

But for the first time, the skeleton was right. The draft described how the model should operate inside the project, not how it should sound.

From here, we did one more pass:

• We pulled in missing constraints (density, voice, escalation, failure handling).

• We forced it to collapse everything into a single block of prose instead of numbered rules.

• We removed anything that smelled like “content marketing,” summaries, or key takeaways.

The next output was good enough to ship into this project: a clearly structured System Instruction block we could paste into the Project’s System Prompt and use to help write the article you’re reading now.

It is not “final.” System Instructions are never final. They are stable enough to test until your next round of corrections.

That’s the standard you should aim for: stable enough to test, not perfect.

Once Your System Instructions Exist, the Real Work Starts

Most people stop here. They have a block of text that looks serious, they paste it somewhere, and they move on.

That’s where most of the value is lost.

System Instructions only pay off if:

Read more

Each week in , we show how operators design, build, and run systems — from architecture and workflows to the decisions behind them.

Subscribe now

For no/low-code and self-taught builders, file hosting is always the hidden weak link. URLs expire. Image hosts throttle. Platforms rename your assets or delete them when you hit a plan limit. And the moment you switch tools, half your automations break.

This guide solves that permanently.

Amazon S3 (Simple Storage Service) is the internet’s universal file cabinet — cheap, stable, and fast. Don’t worry if AWS feels intimidating. You don’t need to “learn cloud infrastructure.” You only need one service and six steps.

Let’s get you set up.

What Is an AWS S3 Bucket?

Amazon S3 is an object storage1 system — meaning it stores files exactly as they are, hands-off and uncompressed, until you decide to change or remove them.

If you upload an image, logo, JSON config, video, or tiny HTML file, S3 keeps it:

• permanent

• globally available

• predictable

• and served from a stable URL that never changes

It’s not a web server. It’s not a database. It’s not “file hosting” in the cutesy startup sense. It’s industrial-grade storage you’ll never have to think about again.

Why We Use It

Most no-code platforms eventually hit the same limits:

• File URLs break

• File hosts rate-limit you

• Assets get renamed

• Storage gets expensive

• Switching platforms creates chaos

S3 avoids all of that. It’s stable, boring, and reliable — exactly what you want from the thing holding all your assets.

It scales down to pennies and behaves the same whether you’re storing five files or five million. And when you update a file, every automation, email, or app using that URL updates instantly.

Why You Need It

S3 gives you a universal home for anything your workflows depend on — separate from whatever app or tool you’re using. That separation is the whole point.

The short version:

1. Your URLs never break.
Files don’t expire, change names, or disappear when a platform updates something.

2. Your stack gets faster.
S3 serves assets at global speed.

3. Your costs drop to almost nothing.
Hosting common assets usually costs under $1/month.

4. Everything in your stack can use it.
Make, Zapier, Bubble, Webflow, n8n, Retool, Framer — all of them accept a public S3 URL.

5. Your system becomes cleaner and more modular.
Files live where files belong. Everything else becomes easier.

Why not store files in your app or hosting provider?

Because those systems are built to run code, not hold assets.

They’re slower, more fragile, and way more expensive for the same job.

S3 gives you speed, stability, and a permanent centralized place for the parts of your system that shouldn’t ever move.

What you can actually do with S3

• Host images for emails, automations, or landing pages

• Store JSON configs for workflows

• Serve icons, brand assets, or videos

• Keep a permanent, versioned library of files

• Use one stable URL everywhere

You’re not becoming an AWS engineer.
You’re simply using the one AWS service that makes everything else in your stack easier.

Creating an AWS Account and S3 Bucket

Before You Start
The AWS Console looks messy the first time you see it. That’s normal. AWS hides simple things behind complicated menus.

Ignore 95% of what you see — you don’t need it.

Just follow these steps exactly. We’ll tell you what matters and what you can safely disregard.

Step 1: Create Your AWS Account

1. Go to: https://portal.aws.amazon.com/billing/signup

2. Verify your email, payment method, and phone

3. You’ll be signed in as the root user

You only use the root user for billing and top-level account operations. Everything else happens from a safer admin account.

Step 2: Create a Secure Admin User (IAM)

1. Open IAM

Search for IAM in the console → open Identity and Access Management.

2. Add MFA to the Root Account

• Click your name (top-right) → My Security Credentials

• Under MFA, choose Activate MFA

• Select Virtual MFA device

• Scan with Authy/Google Authenticator

• Enter both codes

Your root account is now secure.

3. Create a New Admin User

Go to: IAM → Users → Add users

• Username: admin or your name

• Enable: Provide user access to the AWS Management Console

• Use Autogenerated password

• Require a password reset

Click Next.

4. Create an Admin Group

• Add user to group → Create group

• Name: Admin

• Policy: AdministratorAccess

• Confirm and create

5. Save Your Sign-In URL

AWS gives you a login link like:

https://your-account-alias.signin.aws.amazon.com/console

Bookmark it.

6. Log In as the Admin User + Add MFA

Log out of root → log in via your new URL → create a new password → activate MFA again.

Your daily-use admin identity is ready.

Step 3: Create Your S3 Bucket

1. Search for S3 in the AWS Console

2. Open it → click Create bucket

3. Name your bucket (lowercase, hyphens allowed)

4. Choose your nearest region

5. Leave ACLs disabled

6. Keep Block Public Access ON for now

7. Click Create bucket

This is your permanent asset home.

Step 4: Make the Bucket Public (For Non-Sensitive Files)

If you’re storing images, logos, or static assets, making the bucket public is safe and gives you universal, easy URLs.

1. Disable Block Public Access

Bucket → Permissions → Edit
Turn off all four toggles → Confirm.

2. Add This Bucket Policy

Replace YOUR-BUCKET-NAME:

{
  “Version”: “2012-10-17”,
  “Statement”: [
    {
      “Sid”: “PublicReadGetObject”,
      “Effect”: “Allow”,
      “Principal”: “*”,
      “Action”: “s3:GetObject”,
      “Resource”: “arn:aws:s3:::YOUR-BUCKET-NAME/*”
    }
  ]
}

Save.

Note: Do not change the Version date to match the date you paste it. Leave it as-is.

3. Test It

Upload a test image → open the Object URL → try it in Incognito.

If it loads instantly, you’re set.

Step 5: Add Cache Headers (Optional but Recommended)

Caching makes your files load faster and reduces costs.

1. Open your file → Properties

2. Edit Metadata

3. Add:

   • Key: Cache-Control

   • Value: max-age=86400

Save.

Step 6: Verify Access and Understand Cost

Test your file’s URL in a private window.

For typical no-code/automation usage, your monthly cost is usually under $12 — often covered entirely by the AWS Free Tier.3

Just keep public assets separate from private data.

You’re Done

You now have:

• A permanent home for your assets

• Stable URLs that never break

• A faster, cleaner, more reliable automation stack

• A future-proof system that won’t collapse when you switch platforms

Top Object Storage Alternatives to AWS S3

If you want the deeper layers of this work — the wiring, configs, edge-case failures, and the full iterations — you can unlock all of it by becoming a member.

And if you’re at the point where you’d rather have this built inside your business instead of trying to replicate it yourself, start with our $5K Starter System:
10 days. One workflow. No contracts. Full ownership.

Subscribe now

Share

Leave a comment

Each week in , we show how operators design, build, and run systems — from architecture and workflows to the decisions behind them.

Subscribe now

We came into a full-scale rebuild — a messy enterprise overhaul touching operations, logistics, data handling, scheduling, task tracking, ticketing, automation, Slack, on-prem security, and multiple IT environments.

It was the kind of infrastructure project where everyone knows something’s broken but nobody can quite define what.

That’s why real discovery isn’t a kickoff meeting. It’s listening for what’s missing.

One of the most useful things anyone said came from HR in passing, during a conversation about payroll.

“Oh honey, I sent my last timecard reminder already.”

That was it — not a complaint, but fatigue. She was done sending them manually. But the company still needed them.

That line became the foundation for a fully automated, dynamically scheduled, biweekly reminder system that’s still running three years later.

Before You Can Send an Email, You Need a Contact Sheet

Before there were reminder emails, there had to be someone to remind.

That sounds obvious — it wasn’t.

When we started the enterprise rebuild, we assumed the company already had a centralized contact sheet somewhere.
It didn’t.

There were fragments: a locked HR system we couldn’t touch, a production spreadsheet that was months out of date, and a string of emails across 20 people used as an address book.

No one could answer a basic question:

Who’s here right now, and how do we reach them?

That’s where we started.

From Contact Sheet to System Backbone

We needed something fast, accessible, and flexible enough to scale into other automations. So we built it in Airtable1. It wasn’t glamorous, but it was the only tool that gave us a database with an API, relational fields, and permissions — out of the box.

Why We Chose Airtable

We couldn’t afford to manage poorly-designed stock forms or build another front-end just to capture data. We needed a platform that could start collecting, connecting, and automating fast. Airtable gave us that. It’s a real database with structure, permissions, and a front end our teams could actually use.

We didn’t skip other tools because we were avoiding complexity — we skipped them to protect momentum. Every hour spent building a custom data layer would’ve delayed the automations that actually mattered. Airtable made data instantly available for downstream workflows while keeping it digestible for the people using it. That speed wasn’t just technical — it was psychological. When users see a live system responding right away, they trust the change faster.

Starting directly in Airtable meant we could evolve the automations in parallel without wasting time on infrastructure. It was the fastest path to something real — a backend with a digestible front end, and a foundation solid enough to grow from.

The first version was bare-bones:

Employee Airtable Schema

Everything started from there.

Preferred Email Logic

We needed a way to automatically choose the right email address for each person — their company email if they were still active, or their personal one if they’d finished out their contract and their company email had been deactivated. The goal was simple: make sure every message always reached the right inbox without anyone manually switching fields.

So, we wrote a simple Airtable formula that became the foundation of every workflow that followed:

IF( 
   {Status} = “Active”,
   {Company Email},
   {Personal Email}
  )

That single field meant every downstream automation — reminders, reports, project notifications — always had the right email address.

It also respected security rules: messages to active staff stayed on company domains; notices to inactive staff went to personal ones once, then stopped.

Every send checks the employee’s status. If someone wrapped a show, it sends to their personal email once and thanks them. If they’re still active, it uses their business address.

Airtable handles it with a simple conditional:

It’s simple, but it guarantees delivery continuity — even through offboarding.

What started as a contact sheet turned into a lightweight personnel database. It powered every other system: timecard reminders, production scheduling, and even access control for on-prem assets.

Because it lived in Airtable, we could:

• Filter by department, title, or show.

• Pull live data via the API into Make.com.

• Validate changes instantly without IT tickets.

That table became the backbone for the entire automation suite.

See our separate Contact Sheet Build Log_ entry for the deep dive on schema design and field relationships. Releasing soon.

Why This Step Matters

Everyone wants to jump straight to automation. But if your contact layer is wrong, every downstream system inherits that will also be wrong. Before you build any workflow, you need a single source of truth for who’s involved.

Once we had that, the rest — including the email reminders — could finally start.

HTML, Not Plain Text

We could’ve sent simple text emails. We didn’t. Because UI and UX design matter, even in internal systems.

UI and UX design aren’t optional — they determine comprehension, timing, and accuracy. Properly structured HTML emails reduce friction and error.

So we built HTML emails from day one.

The goal: make every reminder feel clean, readable, and impossible to miss. Every visual element served a function — structure, hierarchy, or accessibility.

The Template and the Design System

We started with a free GitHub email template2 — a simple, table-based layout. We stripped everything except what we needed, rebuilt spacing, typography, and structure.

We used Postdrop.io3 for render previews and early layout testing. We stored all graphics in Amazon S3, exported at 2× resolution for crisp Retina rendering. That means each image was rendered at twice its displayed size, but defined in HTML at the normal (1×) width and height. It’s an old-school approach — not the modern srcset way — but it works. And in internal systems, reliability beats elegance every time.

Every visual decision had a purpose — readability, not flash.

Color choices followed basic brand guidelines and accessibility contrast standards4; banners were bold, typography minimal.The design could survive desktop and mobile with no modification (for now).

Every layout decision had a reason:

Read more

Each week in , we show how operators design, build, and run systems — from architecture and workflows to the decisions behind them.”

Subscribe now

“Sameness—done right—moves faster than balance ever could.”

I used to think founders needed opposites.
The visionary and the executor. The dreamer and the realist.
Every book, podcast, and accelerator pitch said the same thing: find your complement.

Then I met my now business partner, Craig—and realized sameness, done right, moves faster than balance ever could.

We think at the same speed. We obsess over the same details. We catch each other’s typos mid-sentence and finish design notes before the other person hits send.

On paper, that sounds dangerous: two people with identical blind spots steering the same ship.

But sameness only kills you if you stop evaluating.
We never do.

The Myth of “Balance”

Balance photographs well on founder decks.
But most balance is translation.

“Balance” sounds responsible. Investors love it.
But most so-called balance is translation. One person speaks in instinct, the other in spreadsheets—and both burn half their energy converting ideas back and forth.

That gap costs time—momentum’s most fragile resource.

Sameness compresses that gap.
When both people process decisions through the same rhythm, you don’t spend energy convincing each other why; you spend it improving the how.

You start the iteration loop faster.
That’s the real advantage. Not yin-yang mythology—velocity.

Our Operating Loop

We run a simple loop:

• Decide. Move while the signal’s fresh.

• Plan. Name the risk, name the resource.

• Execute. Build, test, publish, ship.

• Evaluate. Ask, What did we learn?

• Apply. Adjust the system, not the person.

Then repeat.

“Every launch, every misread, every near-miss becomes data—not drama.”

That feedback loop is our partnership’s real infrastructure.
It’s what keeps sameness from collapsing into echo.

Shared Pace ≠ Groupthink

People assume two similar founders will reinforce bad habits.
Sometimes that’s true. But similarity only creates danger when ego outruns structure.

Our guardrail is ruthless reflection.
We say the quiet parts out loud:
Did we rush that? Did we over-design it? Did we miss the human part?

The conversation that would be awkward in most teams is routine for us.
We built the habit early, when it was just survival—two people figuring out how to ship before the budget ran out.
We couldn’t afford politeness.

Now it’s the operating culture.
Honesty before harmony.
That’s how sameness stays sharp.

Why “Balance” Can Slow You Down

“You can test ideas without being opposites. You just need equal stakes and equal speed.”

Opposite-type partnerships often mistake tension for productivity.
They believe friction keeps ideas tested.
Sometimes it does.
But often it just keeps them tired.

You can test ideas without being opposites.
You just need equal stakes and equal speed.

Craig and I both treat momentum as sacred.
When one of us hesitates, the other fills the space—not to dominate, but to protect the flow.
It’s like editing in real time: cut the pause, keep the rhythm, stay in motion.

That rhythm compounds.
Two aligned founders aren’t twice as fast; they’re exponential.
Each iteration multiplies instead of adds.
The work starts to accelerate on its own.

The Cost of Speed

Of course, there’s a downside: sameness can blur boundaries.
When you both care about everything, you risk caring about nothing in particular.

The fix isn’t more structure—it’s sharper structure.

We divide by responsibility, not hierarchy.
I lead narrative and architecture; Craig leads operations and validation.
But every decision crosses both lanes once before it locks.

“If we can hand it to each other at any stage and it still makes sense, the system’s working.”

That single crossover keeps quality consistent without dragging decisions into committee.

Details as Culture

We don’t celebrate speed for its own sake.
We celebrate precision at speed.

Every deck, doc, and deliverable has to look professional because it signals care.
The details are the culture. They’re how the team knows what “done” means.

People sometimes read that as perfectionism. It isn’t.
It’s stewardship.
The moment you treat the details as optional, you start teaching mediocrity.
The company remembers.

So we protect the details the way we protect each other’s time:
fiercely, quietly, every day.

The Line

Partnership isn’t romance and it isn’t balance.
It’s shared standards held at the same pace.
It’s the relief of not having to translate your ambition.
It’s knowing that when you say good enough isn’t, someone else already agrees.

The hardest part of solo work isn’t isolation—it’s momentum decay.
A good partner keeps the loop alive.

If you want the deeper layers of this work — the wiring, configs, edge-case failures, and the full iterations — you can unlock all of it by becoming a member.

And if you’re at the point where you’d rather have this built inside your business instead of trying to replicate it yourself, start with our $5K Starter System:
10 days. One workflow. No contracts. Full ownership.

Subscribe now

Share

Leave a comment

I read Erin Green’s post in the quiet of the morning — the kind of scroll that stops you mid-swipe.
It was about the psychology of buying — how almost every decision happens before logic even joins the conversation.

“Professionals make safety feel effortless.”

That line stayed with me. Because a few days earlier, I’d sat through a sales call that embodied the opposite — not maliciously, just structurally. A small, ordinary collapse of detail that made everything feel unsafe before a word had been sold.

The Setup

It started with a message from a director of sales. He had the kind of pedigree that usually signals competence — big blue-chip tech background, years of experience in enterprise sales. The note was simple: business development opportunity.

The site he linked to didn’t help. It looked like every template-built service company trying to be everything at once: “LinkedIn experts,” “growth partners,” “revenue generators.” A dozen phrases that said nothing. But curiosity is part of the job. I take first meetings because, every so often, something real hides behind bad design.

He followed up quickly, saying the founder would also join the call. Also was the keyword. I expected two people — the operator and the owner.

Then the calendar invite arrived. The agenda sat neatly in the notes, the kind of boilerplate that already tells you what the experience will be:

• Your business, including its products, services, and goals…

• What we do to win clients new business.

• Pricing.

• Next steps.

And at the top, a line that made me smile for the wrong reason:

When you book a call, you’ll automatically be penciled in with your LinkedIn expert.

Except I hadn’t booked them.
They’d booked us.

I turned to my partner and said, “Watch this. This is going to be a cold-call demo in disguise. He’ll want me to explain what we do so he can pitch it back to me. He’ll call it ‘personalization,’ but it’s really just guessing on camera.”

My partner laughed. We both knew the type.

The Call

Only one square lit up when the meeting started — the founder. No director of sales. No apology. No mention. The camera was tilted too high, the energy already off.

He opened with the line I’d predicted to the syllable: “So tell me a little about what you do.”

I smiled, quietly, inwardly. The small grin of confirmation when a hypothesis lands. I caught my partner’s eyes on the other screen — the silent told you so that only two people can share mid-meeting.

I let him run for a bit. Then I stopped him.
“You reached out to us,” I said. “Why don’t you start?”

He stumbled, rewound, restarted. The sentences came out like cards dealt from an old deck — the same pitch he’d probably used a hundred times. No evidence of research, no curiosity, just the cadence of a man following a script he no longer believes in.

That’s when my role shifted. He thought he was selling me; I decided to use him as a live lesson. My partner is younger, and sometimes the best way to teach what not to do is to let someone else demonstrate it perfectly. So I stayed on the call. Every minute of it.

He wasted my time, but he didn’t waste my partner’s. Because now he knows exactly what bad looks like — what unprepared feels like, what fear smells like, how quickly confidence drains out of a room when someone arrives with nothing but a pitch deck and hope.

The Aftermath

The follow-up email landed within the hour. A “proposal,” if you could call it that — a single page with logos splashed like stickers, each one demanding attention, none earning it. A layout that shouted look at me instead of trust me.

That was the moment the pattern clicked. This wasn’t a one-off bad call. It was a system failure disguised as hustle.

Because here’s the truth — there’s nothing wrong with a founder joining a sales call. But if you’ve hired a director of sales, and you’re still taking every call yourself, you don’t have a sales process. You have a bottleneck. You’ve built roles you don’t trust to do their job. That’s not leadership. That’s fear.

And fear is the enemy of safety.

I wrote back. Politely. A clean no, thank you. I could have ghosted. I didn’t. He asked for feedback.

So I gave him a gift — founder to founder, line by line. Where it went wrong, why it felt unsafe, and what preparation would have looked like if he wanted another outcome.

He didn’t take it well. Defensiveness is the instinct of the unready. But that’s fine. I wasn’t writing for him. I was writing for everyone still learning how to carry care forward — for the people who will one day be in his seat and need to know what safety actually feels like.

The Lesson

Erin’s post came a few days later. Her line echoed the whole thing back to me:

“Professionals make safety feel effortless.”

That’s the word — feel. Because safety isn’t a bullet point on a slide. It’s a felt sense that someone has already done the thinking, that they’ve built a structure around you so you don’t have to brace.

It’s in the first thirty seconds of a meeting, when the agenda proves they know who they’re talking to.
It’s in the design of the follow-up, where the typography and tone align with what they said on the call.
It’s in the discipline of showing up prepared enough that the other person can relax.

Safety is proof of care, made visible through detail. And when it’s absent, fear leaks through every pixel.

This founder wasn’t a villain. He was just unprotected by his own system — trying to improvise competence instead of codifying it.
That’s how businesses quietly corrode from the inside: presence replacing preparation, hierarchy standing in for trust.

The irony is that I didn’t leave the call angry. I left amused. Grateful, even. Because it reminded me how fragile professionalism becomes when it’s built on personality instead of process.

Nothing Lost

That call didn’t cost me anything but forty minutes. It bought my partner a lesson worth years.

He saw what happens when someone reaches out promising expertise but delivers uncertainty.
He saw that a company claiming personalization without a single personal gesture has already told you everything you need to know.
He saw that fear is loud — but safety is quiet.

The man on that call was selling shortcuts.
What he gave me instead was proof that the long way still wins.

Professionals make safety feel effortless.
Not because it is —
but because they’ve built a system strong enough to carry the effort for everyone else.

The details aren’t decoration. They’re everything.

Fifteen years ago, I was newly promoted—my first real supervisory job.
We were juggling multiple shows at once, each with its own specs, codecs, and delivery quirks.

So I drafted a simple note for everyone’s desks:

“Please check with the post supervisor on your show before making changes.”

It wasn’t a power move. It was a safeguard—clarity that would keep people from wasting hours or breaking a pipeline.

The executive producer walked in, tore every copy off the desks, and told me to send him the file.
When it came back, he hadn’t changed a single instruction—just run Find and Replace on the title.
Every line that once said the post supervisor now said the Daniel.

That’s when I realized how fragile some hierarchies really are.

The Mentor I Thought I Had

This wasn’t just any executive producer.
He was someone I admired—someone who had once taken me under his wing.
When I first got the job, I didn’t even call him by his first name. I called him Sir.

Not out of fear, but respect. Genuine respect.
I sandwiched every sentence like a soldier. “Yes, Sir.” “No, Sir.” “Absolutely, Sir.”

I didn’t grow up that way. I wasn’t raised in a “Yes, Sir” household.
But he carried himself like a mentor, and I followed his lead because I wanted to earn it.
He was sharp, decisive, magnetic—the kind of person who made you believe that excellence required obedience.
For a while, I believed it.

One afternoon, the owner of the company walked through the bullpen and stopped beside us.
People were everywhere—editors, assistants, producers.
She smiled at me and said to him,

“I can’t believe he calls you Sir.”

And then it happened.

It wasn’t just a grin—it was a glow.
He leaned back in that cheap bullpen TV chair everybody knows too well—the one from the back of the catalog, the one an intern can assemble with an Allen wrench in twenty minutes.
He rocked into it like it was a tufted leather executive throne, the kind where a fat CEO leans back with a cigar, half-laughing at his own invincibility.
That was the look—I have more money than you’ll ever have, kid.

He said, almost purring,

“I love it!”

No one laughed.
No one even looked.
They were too afraid.

And I saw it then: the whole performance of mentorship collapsing into something smaller, meaner, and deeply insecure.
He didn’t love being respected.
He loved being obeyed.

The Day Respect Was Deleted

When he hit Find and Replace on my name, it wasn’t about copy or formatting.
It was about power.

He turned a line about teamwork into a memo about control.
He needed every instruction to point to him, not the system.
Not the system we were trying to build.

In one keystroke, he replaced a process with a person—himself.

He printed it out, made copies, and handed them to me—because he didn’t want me changing it back.
I was staring at the paper, holding proof of how much he meant it.
He was a shithead enough to make me put those pages back on everyone’s desk.

It wasn’t a suit.
It was a cheap t-shirt and a baseball cap.
That’s what authority looked like that day—casual, petty, and absolutely certain it was in charge.

It wasn’t rage that hit me. It was recognition.
He wasn’t protecting the work; he was protecting the illusion that the work ran through him.
And for the first time, I saw what hierarchy really looked like up close:
fragility pretending to be control.

When Ego Runs the Room

I stayed long enough to see where that fragility leads.

Assistants worked unpaid hours.
Credits vanished.
People were rewarded for staying quiet.
And when I tried to protect them, it became personal.

Then came the email.
It was long, deliberate, and designed to wound.
The first line read:

“I don’t like what went down here.”

From there, it unraveled into a lecture—paragraph after paragraph of condescension and control.
He didn’t just question what I did; he questioned that I had any right to do it at all.
It was hierarchy defending itself.

It’s not unique to television—it happens in any business where pressure and ego share the same room.
Television just exposes it faster.
When ego takes over, the damage happens in real time.
The deadlines don’t wait for anyone’s pride to cool off.

And the fire doesn’t stay in the room.
It spreads.
What starts as lost hours becomes lost trust, lost talent, and eventually, lost business.
That’s the part people miss: ego doesn’t just damage teams; it dismantles the system that keeps them working.

Learning to Unlearn “Sir”

When I finally resigned, I started calling him by his first name.
It felt strange, unnatural—like trying to speak a new language after years of habit.
I had to force it.

The reflex to call him Sir was so deep that it caught in my throat.
It took months to stamp it out.
Any time I answered the phone or drafted an email—to anyone—I had to remind myself:
you don’t owe that deference anymore.

That’s the strange part about professional betrayal—
you don’t just lose a mentor; you lose the version of yourself that believed in them.

What Real Leadership Looks Like

The best leaders I’ve met since all share one trait:
they hire people who are smarter than they are and give them room to prove it.

But more than that—they give them the space to speak,
and the confidence that nothing bad will come from telling the truth.

When people know you’re aligned on the same mission,
you don’t need control—only trust and shared purpose.

Because when people know you’re aligned on the same mission—
making the work better, faster, easier for everyone—
you don’t need control.
You just need trust and shared purpose.

That’s what Craig and I built instead—
systems that remove bottlenecks, reward judgment,
and let good ideas move without permission.

The Lesson That Stayed

If your business depends on your ego, it’s already dying.
The only systems that last are the ones built on trust, not dominance.

He hit Find and Replace to make himself feel indispensable.
But what he really did was prove how replaceable that kind of authority always is.

Respect can’t be edited in.
You earn it—or you delete it by accident.

I was in the office this week for one of my longest-running clients.

We’d been on-site with them for years — through every production cycle, every platform change, every reinvention of how the work got done. Together we built systems that made things faster, clearer, calmer. It was a partnership that worked.

Now they’re closing their doors.
A series of mergers, layoffs, and restructures finally caught up.

It’s not sudden, but it still lands heavy when you’re standing in the middle of it.
There’s a procedure to endings in this business — finishing deliveries, archiving the drives, signing the final paperwork, clearing folder permissions. It’s half ritual, half checklist.

You look around the office — empty edit bays, nameplates half peeled off — and realize it’s all winding down.

Being one of the last people left is a strange kind of responsibility.
You remember where everything lives, what every note meant.
You’ve already done your grieving; you made peace with it months ago.
But when it finally hits paper — the signatures, the empty desks — it still feels like a loss.

The Chalkboard

I was in the kitchen grabbing a water when I noticed the chalkboard — still hanging there, mostly blank.

A few faint outlines from old designs were left behind, and it made me think of one of the younger team members who’d started there not that long ago.

He was an intern who’d worked his way up to PA.
Smart, steady, reliable. The kind of person who noticed details most people miss.

He was the one who used to take that chalkboard every month and turn it into something great — birthdays, holidays, little seasonal jokes.
Fun, simple designs that gave the office some personality.

He didn’t do it because anyone asked him to.
He just had the natural impulse to make things a little better — to bring some order and life into the room.
That’s what always stood out to me. You can’t teach that.

It’s the kind of instinct that, if you nurture it, can become the thing you build a career around.

He happened to be there that day, picking up a few last boxes.
We caught up for a bit.

He told me he was moving back home, taking shifts at the diner for a while.
He smiled when he said it — that polite, practiced kind of smile people use when they’re trying to sound okay.

He’ll be fine. I don’t doubt that.
But I could see the hesitation — that mix of fear and logic you only get when you’re young and standing on the edge of something uncertain.

I told him again what I’d said before: that he had a gift, and that people would pay for that kind of work.
Chalkboards, weddings, signage, murals — simple ideas, small ways to start.

He said he’d think about it.

And I get it. I really do.
Stability is a powerful thing.
It’s hard to walk away from the safety of what you know, even when what you know is ending.

Sometimes the best advice you can give someone is the simplest one: just start.

It’s easy to say and hard to live by.

Planning feels responsible. Reflection feels smart.
But if it never turns into movement, it’s just a loop.

The Rhythm That Works

I’ve seen it play out in productions, in business, in myself.
You can’t plan your way to motion.
You have to move your way to better planning.

That’s how I run my business.
It’s how I like to lead my teams.

Plan. Move. Adjust.
Over and over.

I learned that rhythm years ago with a counterpart at the client — a VP who ran the production schedule with me.
Every time we sat down to map it out, we’d only plan two weeks ahead.
Anything beyond that was guesswork.

We’d lock what we could see, then start.
The moment we moved, the plan changed — and that was fine.
That was the point.

You learn while you’re doing.
The plan gets sharper because you’re already in motion.

That’s the same rhythm I keep now with my business partner.
It’s how we run everything — short windows, constant recalibration.

We’ll look two weeks out, maybe three if we have to, tighten what’s in front of us, and keep moving.
That’s what keeps the company alive.

We don’t try to lock the next six months.
We make the best next move, measure it, and make another.

Momentum and Trust

Reflection matters. Thinking matters.
But the work only breathes if you’re doing both at the same time.

That’s what I hope that young guy eventually figures out — that momentum doesn’t mean rushing ahead blindly.
It means keeping the loop alive: plan, act, learn, adjust.

Don’t freeze.
Don’t wait for certainty.

He’s got time. He’ll find it.
But I hope he starts soon.

Because in the end, that’s what keeps any of us working —
not the perfect plan, not the guarantee,
just the willingness to move.

That’s what’s carried me through every closing, every reset, every start.

Plan. Move. Adjust.
That’s the work.

It’s become its own genre:
“100 ChatGPT prompts to change your life.”
Swipe, scroll, bookmark.

Every one of them looks the same:

“Write a business plan for [insert your idea].”
“Generate 10 product names for [insert your product].”
“Draft a pitch email for [insert your audience].”

The trick is hidden in plain sight.
The power isn’t in the prompt.
It’s in the brackets.

The Prompt Hustle

I’ve lost count of how many LinkedIn posts or Twitter threads I’ve read that promise “the ultimate prompt pack.” They rack up likes, comments, and reshares. People really want to believe the magic is in the words you feed the machine.

But look closer. The structure is always the same. The actual prompt is filler. The system is selling you the illusion of leverage.

The real load-bearing part — the thing that makes the AI useful or useless — is what you put in the brackets. Your business idea. Your product. Your story. That’s the part they can’t sell you.

It’s the hardest part. It’s also the only part that matters.

What the Brackets Really Mean

• “Write a sales plan for [my product].”
  → But what is the product? Who buys it? Why would they care?

• “Write a funding pitch for [my startup].”
  → But what problem do you solve? What market are you entering? What proof do you have?

• “Write a marketing strategy for [my brand].”
  → But what’s your positioning? Your wedge? Your voice?

The machine fills space. You still have to fill the brackets with something real.

That’s not a trick. That’s the work.

Why the Brackets Hurt

Most people stumble here.
Because what goes in the brackets isn’t surface-level.
It’s identity.

• Naming the real thing you want to build.

• Accepting the risk that it might fail.

• Distilling the messy human intent into a single sentence that can survive automation.

AI can’t carry that for you. It will carry the repetition, the scaffolding, the formatting. But it will not invent the thing that matters in your brackets.

That’s your job.

The False Shortcut

The prompt packs are popular because they hide that truth. They let you feel like you’re making progress without facing the hard work of defining what belongs inside the brackets.

It’s the same instinct that fills hard drives with half-finished pitch decks, or notebooks with abandoned outlines. Collecting prompts feels like motion. But motion isn’t traction.

At some point, you have to stop scrolling and name your damn product.

What Doing the Work Looks Like

Doing the work isn’t glamorous. It means:

• Interrogating the brackets. Not “a book,” but your book. Not “a candy bar,” but your candy bar, your market, your brand story.

• Breaking the silence. Writing the messy draft, the first outline, the ugly version you’d never show anyone.

• Iterating in public. Testing your idea where it can fail, and learning if anyone cares enough to pay.

AI can draft 10,000 versions. None of them matter until you’ve put something true into the brackets.

A Personal Note

I use AI every day. It carries the load. It structures my thoughts, cleans up my phrasing, and helps me see the edges of my own argument. I’m not anti-prompt.

But I don’t confuse the filler words with the work.
The work is sitting with the bracket until it’s not empty anymore.

That’s why most of the “ultimate prompt packs” feel hollow. They skip past the hard part. They pretend the brackets will magically fill themselves.

They won’t.

The Human Part

Prompts don’t build businesses.
People do.

AI doesn’t save you from naming what you want to make.
It just waits for you to fill the brackets.

And if you can do that — if you can sit with the silence long enough to name your book, your show, your product, your company — then the machine becomes powerful.

Until then, it’s just filler.