Chapter 8: The Flashback — Aurel
2:47 AM
September 6th, 2017
Aurel, Texas
I arrived in Aurel with one suitcase, five thousand dollars in relocation money, and the kind of certainty that only exists at twenty-seven—the belief that if you worked hard enough, long enough, smart enough, the system would finally recognize you.
The breakup with Liv Hendrix was three weeks old. Clean break, we’d said. Mutual, we’d said. The kind of lie you tell because the truth—that she’d looked at me one morning and realized I was optimizing her like a conversion funnel—was too precise to say out loud.
“You’re always measuring,” she’d said. “Even when you think you’re not.”
I’d wanted to argue. Wanted to say that paying attention to patterns wasn’t the same as being cold. But she was already packing, and I was already calculating how long it would take to feel normal again, and proving her point in the process.
Three weeks later, Elliot Cade sent the offer.
Subject: Growth Architect — DriveEd
Body: One line.
“We need someone who sees systems, not users. Salary negotiable. Equity standard. Relocation covered. Start Monday if you can.”
I could.
The flight to Aurel was two hours. I spent it reading DriveEd’s public-facing materials—blog posts about “democratizing driver education,” a TechCrunch article that called them “Uber meets Khan Academy,” investor decks that promised seven-figure revenue by year two.
The company was ten people. Elliot Cade was the CEO—twenty-nine, second exit already behind him, the kind of founder who spoke in aphorisms and made you feel like cynicism was a form of cowardice.
The office was a converted house near the university. Exposed brick. Hardwood floors that creaked. A kitchen with a French press and a whiteboard covered in projections that climbed like religious prophecy.
REVENUE TARGETS — 2017
Q3: $50K
Q4: $250K
Q1 2018: $1M
Elliot met me at the door. Three-second handshake. Eye contact that felt like a test.
“You’ve got three months to prove The Mesh wasn’t a fluke,” he said. “After that, we’re either scaling you or replacing you.”
I nodded. Said thank you like this was reasonable. Like I wasn’t carrying everything I owned in a duffel bag and sleeping on an air mattress in a sublet that smelled like someone else’s failed optimism.
He walked me through the space. Twelve desks. A projector someone had salvaged from a high school and turned into ambient light. A filing cabinet full of energy drinks and protein bars.
“We’re pre-product-market fit,” Elliot said. “But the bones are good. Lead gen works. Conversion’s shit. Attribution’s nonexistent. That’s where you come in.”
He pulled up a laptop. Showed me the dashboard—Google Analytics, barely configured. Traffic flatlined at two hundred visits a day. Revenue: zero.
“Fix it,” he said.
Week One
I spent the first three days just watching.
DriveEd’s model was simple on the surface: connect student drivers with local instructors through a marketplace. Instructors paid a monthly subscription to list their services. Students paid per lesson booked. The platform took fifteen percent.
But the execution was chaos.
The signup flow had eleven fields. Half the instructors who started an account never finished. The ones who did finish rarely got matched with students because the search algorithm prioritized alphabetical order, not location or rating.
Students arrived through Facebook ads that said things like “Learn to Drive Today!” and linked to a homepage with no clear call to action. Bounce rate: eighty-seven percent.
I sat in the corner by the window and mapped it all out in a notebook. Not digitally—pen and paper, the way I used to sketch affiliate networks when I was building The Mesh in my apartment.
The Mesh had been duct tape and caffeine. A content network stitched together with referral links, commission splits, tracking pixels I’d coded myself because the third-party tools were too expensive. It worked because I’d stopped thinking about “websites” and started thinking about paths—the route a user took from curiosity to click to purchase.
Every step was a valve. Every valve had friction. Reduce the friction, increase the flow.
DriveEd had friction everywhere.
On day four, I showed Elliot my notes.
“Signup flow: cut it to three fields,” I said. “Name, email, zip code. Everything else comes later. Search algorithm: prioritize proximity and reviews, not alphabet. Facebook ads: stop selling the dream, sell the next step. ‘Find an instructor near you in sixty seconds.’ Link directly to search, not homepage.”
He stared at the notebook. Then at me.
“How fast can you build it?”
“Two weeks if I do it myself. Four if we involve engineers.”
“Do it yourself.”
Week Two
I stopped sleeping more than four hours a night.
The office had a couch in the back room. I kept a blanket there. Worked until 2 AM, slept until 6, kept going. The rhythm felt like prayer—repetitive, numbing, clarifying.
I rebuilt the signup flow first. Stripped it down to name, email, zip code. Added a progress bar so users could see they were almost done. Reduced cognitive load from “this will take ten minutes” to “this will take thirty seconds.”
Instructors who’d abandoned the old flow started finishing the new one. Completion rate went from forty percent to seventy-six percent in three days.
Next: the search algorithm. I rewrote it to prioritize proximity, then user ratings, then availability. Added filters for transmission type, language, price range. Made the results page feel like browsing Airbnb—photos, bios, star ratings.
Then: the ads. I killed everything that said “Learn to Drive Today” and replaced it with surgical specificity.
New Ad Copy:
“Find a driving instructor within 5 miles. Rated 4.5+ stars. Book your first lesson in 60 seconds.”
I pointed the ads directly at the search page. No homepage. No explanations. Just: here’s what you want, here’s how to get it, go.
On day thirteen, Elliot called me into his office.
“Revenue’s up,” he said.
I pulled up the dashboard. Seven thousand dollars that week. Previous week: zero.
“Don’t fuck it up,” he said.
Week Four
The dashboard became the only thing that mattered.
I kept the same playlist—Bon Iver, The National, Explosions in the Sky, the songs people used for time-lapses of cities waking up. I wore the same hoodie. Sat in the same spot by the window. Ordered the same breakfast tacos from the same food truck.
It was superstition dressed as routine. I was afraid that if I changed anything, the algorithm would notice and punish me for arrogance.
Revenue hit fifteen thousand. Then twenty-two. Then thirty-eight.
Elliot started calling me “the growth guy.” Then just “Growth.”
I didn’t correct him. I liked being a function instead of a name.
The rest of the team started asking questions. How did I know which ads to run? How did I decide what to cut from the signup flow? How did I predict what users would do?
I didn’t have good answers. The truth was intuition sharpened by repetition—watching numbers long enough that you started seeing patterns the way some people see faces in clouds.
But “intuition” sounded mystical, and I wanted to sound scientific.
So I said things like: “Reduce friction at every valve. Optimize for the next click, not the final conversion. Measure everything, trust nothing.”
They wrote it down like gospel.
Week Eight
One hundred thousand dollars in a month.
Elliot bought everyone drinks at a bar on Sixth Street. Gave a speech about how we were going to change driver education. How we were going to scale to every state, every country, every market where people needed to learn something and someone needed to teach it.
“This is just the beginning,” he said.
Everyone cheered.
I stood in the back and watched the dashboard on my phone. Green arrows. Conversion rates climbing. New instructor signups. New student bookings. The numbers moved like a heartbeat, steady and relentless.
I felt something I couldn’t name. Not pride, exactly. Not satisfaction.
Relief.
Relief that the system had recognized me. Relief that effort had translated to outcome. Relief that I’d proven Liv wrong—I wasn’t just measuring, I was building.
I didn’t call her. Didn’t text. Didn’t post anything publicly.
But I wanted her to know I was worth the work.
Month Four
Liv came to Aurel in October.
I didn’t know she was there until three days after she left.
A mutual friend posted a photo: Liv at a coffee shop on South Congress, sun in her hair, laughing at something off-camera. Tagged with a location and a date that proved she’d been twelve blocks away and hadn’t called.
I stared at the photo until the screen went dark. Then I opened it again. Read the timestamp. Checked the geotag. Did the math on when she must have arrived, how long she must have stayed, whether she’d thought about reaching out.
I wanted to text her. Ask why she’d come. Ask if she’d wanted to see me but changed her mind. Ask if she thought about me the way I still thought about her—late at night when the numbers stopped climbing and there was nothing left but the hum of the refrigerator and the too-bright light of the dashboard.
Instead, I opened the analytics. Watched the conversion funnel. Tracked the users moving from ad click to signup to booking.
Every metric was up except the one that mattered.
Month Six
DriveEd expanded to eleven states.
Cease-and-desist letters arrived from two competitors and one regulatory board that didn’t understand how referral bonuses worked. Elliot pinned them to the wall next to the revenue graph.
“If lawyers notice you,” he said, grinning, “you’re winning.”
We hired. Twelve people became thirty. The house got too small so we moved to an office park with glass walls and a courtyard where people took calls in the sun.
My salary hit one hundred sixty thousand. I signed the paperwork in Elliot’s office and felt nothing. The number was abstract. I still ate the same breakfast tacos. Still wore the same hoodie. Still slept four hours a night on the couch when I stayed too late to justify going home.
Elliot started sending me to conferences.
Times Square first. A growth marketing panel where I sat next to a VP from HubSpot and a founder who’d sold his company to Salesforce for eight figures. The moderator asked how I thought about attribution.
I said: “Attribution is just accountability for machines. You’re teaching the system to remember what worked.”
People wrote it down.
Caesars Palace next. A convention hall full of affiliate marketers and SEO consultants. I gave a talk called “Systems That Scale” and showed a slide with DriveEd’s revenue graph—zero to seven figures in eight months.
Someone asked: “What’s your secret?”
I said: “There’s no secret. You reduce friction, you remove objections, you make the next step so obvious that not taking it feels like effort.”
They applauded.
Amsterdam. SXSW. A panel in Brooklyn where everyone was half-drunk and I talked about growth loops like they were poetry.
The after-parties blurred. Open bars. Founders swapping war stories. Investors asking if we were raising. Someone always asking: “What’s your next move?”
I didn’t have a next move. I had this move, and I was riding it until the wheels came off.
But beneath the gloss, beneath the applause, a small truth I couldn’t quite say out loud:
I felt instrumental, not essential.
The system worked. It would work without me. I was just the one who’d figured out how to turn the key. And once you turn a key, once you prove it can be done, you become replaceable.
I started staying later. Testing things no one asked me to test. Building features no one needed yet. I wanted to be indispensable, but the harder I worked, the more automated the system became.
Month Nine
The office installed a cold-brew tap. Someone started a Slack channel called #thingstobragabout and everyone pretended it was ironic.
We had slogans on the walls now:
DRIVE FAST. LIVE SMART.
MOVE FAST. LEARN FASTER.
SCALE OR FAIL.
I stood at the window one evening after everyone left. The Aurel skyline reflected in the glass—cranes and high-rises climbing like the revenue graph. My face superimposed over the slogan.
The city looked like a promise.
For nine months, I’d believed I was building something that mattered. I’d believed the system rewarded people who understood it. I’d believed merit was a ladder you could climb if you were willing to sacrifice sleep, relationships, rest, the idea that being human was anything but a drag coefficient.
I’d believed I was the architect.
I’d believed I was teaching the machine.
The Turn
It was Elliot who told me.
Not cruelly. Not even directly. Just a conversation in his office three weeks before Christmas. Revenue was up. Hiring was up. We were closing a Series A.
“We’re bringing in a VP of Growth,” he said. “Someone with experience scaling to nine figures. You’ll report to them.”
I nodded. Said congratulations. Said it made sense.
He said: “You built something incredible. But we need someone who’s done this before.”
I walked back to my desk. Opened the dashboard. Watched the numbers climb.
For a moment—nine months, really—I’d thought I was the system.
Until I wasn’t.
Artifact: Internal Memo
From: Elliot Cade
To: All Team
Date: December 18, 2017
Subject: Leadership Changes
Team,
As we close out an incredible year, I’m excited to announce that we’re bringing on Marcus Vahn as VP of Growth. Marcus comes from [REDACTED], where he scaled their platform from $10M to $100M ARR in under two years.
Bran will continue leading optimization and attribution, reporting to Marcus.
This is a huge win for the team. Onward.
— Elliot
I didn’t read the memo until two days after it went out.
By then, everyone had already congratulated me. Told me how great it was that we were leveling up. Told me Marcus was a legend, and I’d learn so much.
I smiled. Said thank you. Said I was excited.
That night, I stayed until 4 AM. Rebuilt the Facebook ads. Rewrote the email drip campaign. Optimized the checkout flow even though it was already converting at ninety-two percent.
I told myself I was being thorough.
But the truth was simpler:
I was trying to prove I was irreplaceable.
I was trying to build something so essential, so foundational, that firing me would feel like tearing out the walls.
And when that didn’t work—when Marcus arrived in January and started “auditing my frameworks” and “pressure-testing my assumptions”—I started looking for the next place that would let me build from scratch.
Because starting over was easier than staying.
Because being the architect was easier than being the blueprint.
Because I’d learned, in Aurel, that the system doesn’t reward you for understanding it.
It rewards you for being useful.
And the moment you stop being useful, you disappear.
Month Eleven
Marcus Vahn arrived on a Monday in February.
He wore expensive jeans and a blazer that looked casual but cost more than my monthly rent. He had a way of shaking hands that felt like he was assessing your net worth. His first meeting was with Elliot. His second was with the engineers. His third was with me.
“Walk me through your attribution model,” he said.
I did. Showed him the tracking infrastructure. The conversion funnels. The cohort analysis. The LTV calculations. Everything I’d built from scratch.
He nodded. Took notes. Asked clarifying questions that felt like traps.
“And this is all proprietary?” he asked. “You built this yourself?”
“Yes.”
“No vendor solutions? No third-party platforms?”
“I looked at them. They were too expensive and didn’t do what we needed.”
He smiled. Not warmly.
“That’s the problem with building custom,” he said. “It works until the person who built it leaves. Then you’re fucked.”
I didn’t say anything.
“I’m going to bring in some tools,” he said. “Standardize everything. Make it scalable.”
Scalable meant replaceable. I knew that. He knew I knew that.
But we both pretended the conversation was about efficiency.
Month Twelve
The tools Marcus brought in cost sixty thousand dollars a year.
They did exactly what my custom system did, but slower, with worse UI, and with vendor lock-in that meant we’d be paying forever.
But they were “industry standard.” They had “support.” They came with “documentation.”
Which meant anyone could use them.
Which meant I wasn’t essential anymore.
Marcus started holding meetings without me. Started CC’ing me on emails instead of addressing them to me. Started referring to my frameworks as “legacy infrastructure” that needed to be “migrated.”
I saw it happening. Saw the pattern forming. The same pattern I’d seen before, at other companies, with other people. The slow erasure. The polite marginalization. The careful construction of a narrative where your departure becomes inevitable.
I could have fought it. Could have gone to Elliot. Could have made the case that replacing me was a mistake, that the custom system was better, that Marcus was optimizing for his resume instead of the company’s needs.
But I didn’t.
Because I’d learned something in those twelve months:
Fighting to stay somewhere you’re not wanted is just another form of optimization.
And I was tired of optimizing.
The Meeting
Elliot called me into his office on a Thursday in March.
Marcus was already there. So was the VP of People—a woman named Keisha who I’d helped hire six months ago and who now couldn’t look me in the eye.
“We’re restructuring,” Elliot said. “Marcus is consolidating the growth team. We’re moving to a more vendor-led approach. It’s not personal. It’s just—we’re at a different stage now.”
I nodded.
“We’re offering two months severance,” Keisha said. “And we’ll vest your equity through the end of the quarter.”
Two months. Four thousand shares that would never be worth anything because the company would pivot twice more and shut down inside three years.
I did the math automatically. Couldn’t help it. Even then.
“Okay,” I said.
Elliot looked surprised. Like he’d expected me to argue. To negotiate. To perform some ritual of resistance that would let everyone feel like they’d been fair.
“Okay?” he said.
“Yeah.”
“You can take the rest of the day,” Keisha said. “We’ll have your laptop and access revoked by end of business.”
I stood. Shook Elliot’s hand. Shook Marcus’s hand. Shook Keisha’s hand.
Walked back to my desk. Packed the blanket from the couch. Left the hoodie. Left the notebook. Left everything I’d built sitting on servers I’d never access again.
Drove home and sat in the parking lot of my apartment complex for forty minutes, staring at the dashboard on my phone.
The numbers were still climbing.
The Choice
I could have sued.
Should have sued, probably. The timeline was clear. The replacement was obvious. Marcus had been hired to fire me and take credit for my work. Any employment lawyer in Aurel would have taken the case on contingency.
But I didn’t.
Not because I didn’t have a case.
Because I loved the people I’d worked with.
The engineers who’d stayed late helping me debug tracking pixels. The designer who’d made the dashboard beautiful. The customer support team who’d sent me thank-you notes when the new signup flow made their jobs easier.
They’d be fired too. I knew that.
Marcus would bring in his own people. Elliot would rubber-stamp it. The company would burn through the Series A, pivot to enterprise, lay off half the team, pivot again, then shut down quietly in a blog post that thanked everyone for their contributions.
I could see the pattern. I’d seen it before.
And suing DriveEd—dragging Elliot into depositions, forcing discovery, making Keisha explain in court why they’d fired the person who’d built the entire revenue engine—would hurt them more than it would hurt the company.
So I didn’t.
I signed the severance agreement. Took the two months pay. Let my equity vest into worthlessness.
And I learned something I would never unlearn:
I was a cheap prostitute in a prostitution system disguised as life betterment.
They’d used me to build something they couldn’t build themselves. They’d extracted value. They’d discarded me. And they’d done it all while saying words like “restructuring” and “different stage” and “it’s not personal.”
And I’d let them.
Because I believed—still believed, even after everything—that there was something noble in building systems. That making strangers click was a form of architecture. That optimization was a kind of art.
But it wasn’t.
It was just commerce dressed as progress.
And I was just labor dressed as talent.
Six Months Later
I heard through a mutual connection that Marcus had been fired.
Then the engineers. Then the designer. Then Keisha.
Elliot brought in a new team. Pivoted to B2B. Burned through the Series A. Started fundraising for a Series B that never closed.
Two years later, DriveEd shut down.
The blog post thanked everyone for their contributions.
The Scar
What happened in Aurel—what I learned there, what I became there—changed the trajectory of every choice and mistake I would make over the next ten years of my life.
I learned that loyalty was a liability.
I learned that building something beautiful didn’t protect you from being discarded.
I learned that the system didn’t care about craft or effort or devotion. It cared about narratives. And if your narrative became inconvenient, you were rewritten or erased.
So I stopped being loyal.
I stopped building things to last.
I stopped believing that the people I worked with were anything more than variables in an optimization function.
I became the thing Marcus had accused my code of being: efficient, effective, and completely replaceable.
And when I met Jonas Hale five years later, when he pitched me on building something that mattered, something ethical, something that would learn accountability instead of exploitation—
I said yes.
Not because I believed him.
But because I wanted to prove that what happened in Aurel could be inverted. That a system could be built to remember consequences instead of erasing them. That machines could be taught to carry weight instead of extracting value.
I wanted to build something that wouldn’t fire me.
Even if I had to teach it grief to do it.
For a moment, I thought I was the system.
Until I wasn’t.