Being exploited by a startup is almost a rite of passage. I don’t think I can even call this a cautionary tale, because of what I took out of the experience.
Subscribe to Jacob’s Tech Tavern for free to get ludicrously in-depth articles on iOS, Swift, tech, & indie projects in your inbox every week.
Gather ‘round, friends. This one has been a long time coming.
Grab a beer, we’ve got time 🍺.
This is a cautionary tale—one which left me scarred, jaded, and wiser. I hope that by reading this story, I can protect some of you from 11 months of pain.
The year was 2019.
My migration from academic overachiever to professional underachiever was an overwhelming success, as a slightly under-levelled junior at a big consultancy shop. But I couldn’t shake the feeling I was destined for great things.
I was engulfed by the daydreams many 24-year-olds experience when they stop being described as “precocious” and start spending their commutes listening to billionaires on podcasts. As my parasocial relationships to Peter Thiel and Reid Hoffman took hold, I was pretty sure I could become the next Jeff Bezos, or at least the next Drew Houston—I just had to execute well.
Enter: Fixr
The chance of a lifetime fell into my inbox through a friend of a friend of a friend. I was put in touch with Jimmy, the cofounder & Chief Financial Officer of a startup. His team was on the lookout for someone that knew mobile apps, to advise their startup and put out a few fires.
Involvement with a Startup.
In an Advisory Capacity.
If my weeks of tech consultancy training, and several years in mobile development, had prepared me for anything: it was this.
The venture was operating in stealth mode (if you don’t count the website), so Jimmy was cagey with the details until our initial call. I understood—as (probably) the next Zuckerberg, you need to be careful with your ideas around me.
We got on great. I briefed him on my credentials and showcased my passion for startups, and he told me I’d be a great fit for the team. It was a done deal. I’d taken the first step to fulfilling my destiny as the next great startup (advisor).
Fixr was the operating system for your car. It was the one-stop shop connecting you to qualified, vetted mechanics in your local area for anything your car needs—annual MOT, on-demand repair, and even roadside recovery.
Fixr had been operating for close to 3 years under a crack team of 3 part-time visionaries:
Jimmy (CFO) — an innovation manager at a consultancy
Kim (CMO) — a legal associate at an accounting firm
Mike (COO) — a mechanic
*I changed some names and obfuscated a few details outside the public domain; please simply imagine the cast of Better Call Saul. While you’re at it, feel free to picture me as Tony Dalton.
In a few short years, the team had an impressive collection of achievements:
Securing some startup capital to fund development, from a personal bank loan.
Winning a pitch contest at a local university.
Potential interest from VCs and accelerators, subject to getting traction.
Registering with the UK SEIS scheme; giving a tax break on seed investment.
Arranging a referral partnership, where we could get paid for convincing mechanics to switch banks.
Extensive market research, speaking to hundreds of mechanics, with nearly all of them expressing interest in joining the platform when launched.
Contriving a financial model demonstrating how £250k of seed funding will be transformed into £3M in revenue, with operations across Europe, by year 3.
A live static landing page, hosted on an AWS EC2 Medium server.
4 mostly-built apps for customers & mechanics, on iOS & Android.
Con-sultant
That final point was where they needed some expertise. Most of their bank loan runway had been burned on an overseas contractor they fired after 2 years, for incompetence. They switched to a Hyderabad-based agency, who were sanding off the rough edges and getting the 4 apps ready for launch day.
I met up with Jimmy for some casual in-person end-to-end testing of the near-finished product and… Oh boy.
One could say the app wasn’t quite production-ready.
In my advisory capacity, I was in charge of speaking to the agency and communicating the many, many, many bugs that remained. This is where I got my first taste of gaslighting from a client relationship manager whose job is to keep suckers on the hook.
“We agreed to the previous list of fixations only and never agreed anywhere as the part of the agreement. Please refer the email snapshot above.”
“Look at the specification. It does not specify anywhere we were asked to support screen dimensions larger than iPhone 4s sized.”
“The stripe payment integration is not of the phone application, you do not need this from us”
Things looked a little bleak with the late 2019 launch date, so I suggested to Jimmy perhaps I could help squash some of these bugs. The cofounders didn’t have access to the code repo, so I had to spend more time battling the relationship manager for access.
And eventually, I got it.
“Far, far below the deepest delving of FAANG and big consultancies, the world is gnawed by nameless devs. Even Bill Gates knows them not. They are older than he. Now I have walked there, but I will bring no report to darken the light of day.”
—Gandalf, a famous staff engineer
I saw only one way to bring Fixr to market.
We would need a full rewrite.
Co-flounder
Despite being £20,000 in the hole across 4 apps, the management team didn’t take much convincing to let go of a sunk costs. I whipped up a quick live-demo of a modern iOS map app, with custom photo location pin, plus an input form and photo capabilities.
In one Monster-and-Elvanse-fuelled night of passion, I was light years ahead of a build that had taken 3 years. They were sold.
In order to justify the extra time investment, I was provisionally brought on as cofounder and CTO. But we had a marketplace to build, and two platforms to serve. We needed an “Android guy”. I knew just the person: Gus.
Like me, he was keen on the idea, and in the market for a new side project. At a pre-lockdown party, I sold the dream: this product lives or dies via the traction we get in the next couple of months.
Dragging my good friend into the blender is a regret, but we kept each other sane through the subsequent 9 months.
Despite being late to the party, the team valued our potential contributions at 10% apiece—the level required to get into Y Combinator.
To make things official, and certify our equity stakes, Kim drew up contracts, which were quite obviously rehashed drafts of the contracts they gave the original ill-fated dev agency, plus a very spicy clause pertaining to equity shares.
The Management Team may terminate this agreement and reduce or fully remove the equity held by the Developer where the Development Services are not sufficiently met. This clause is to be carried out at the discretion of the Management Team.
I was about to sign, when Gus cleverly consulted his dad who, in a brilliant blaze of prescience, suggested we run a mile and block our teammates’ phone numbers. Gus and I knew better, and simply pushed the team to change the wording to “gross negligence”.
We were officially cofounders.
It was go-time.
The MVP
The prevailing months flew by in a COVID-tinged haze. While everyone else was baking bread and remunerating nursing staff via saucepans, Gus and I spent every free minute turning Fixr into a serious product.
Our weekly Zoom calls had a regular push-and-pull. Gus and I presented the latest flows. Kim and Jimmy praised our progress, while pushing to expand the scope. We had to add more features to win the market as a one-stop-shop.
Engineering busywork complete, we turned to our operators.
Kim shared her latest dream for our upcoming “viral marketing campaign” that would land thousands of sign-ups on launch day.
Mike spitballed when we might start to acquire mechanics. Perhaps we’d hook into local apprenticeship schemes once we had demand from customers. Achieving liquidity would be straightforward.
Jimmy went through his latest tinkering with the financial model—the cornerstone of our investment thesis—and talked us through the latest couple of emails from VC analysts who asked us to “get in touch when we had traction”.
This early traction would be everything.
After months of neglecting our families, Gus and I had the MVP ready for prime-time, with a single robust user flow:
Users on the Customer apps can post local repair jobs. They can pay via the app and approve each invoice item during a repair job in real-time.
Mechanics on the companion app could bid for jobs, perform vehicle inspections, append work items during a job, and handle invoices automatically.
Fun fact, this project is how I cut my teeth on SwiftUI 1.0. The screenshots are lost to time (and dodgy source control), but the internal architecture was a hot mess.
Early UIKit inter-op was janky and it was difficult to pass state about. This was a problem when two of our major features used the Google Maps SDK (MapView didn’t exist!) and the camera.
A gigantic @EnvironmentObject singleton controlled the entire app state, and because NavigationView was so broken in SwiftUI 1.0, every screen transition was a modal.
This flow worked phenomenally, across our suite of 2 iOS apps and 2 Android apps. But it wasn’t enough for our cofounders. We had to have on-demand roadside recovery. We needed to include annual MOT checks. Flame decals are a must-have. Anything you need for your car, we had to be.