You can't scale yourself

I started Devcore after Artemis worked. The path felt obvious — I’d just shipped a real product solo, the client had loved it, and the next ask was can you do this for someone else. So I said yes, and then yes again, and then I decided I had a company.

I didn’t. I had me, with a logo on top.

That’s the lesson I want to write down before it stops feeling fresh. It’s the one I keep meeting other founders who haven’t learned yet, and I’d rather have it on the record than keep saying it in DMs.

The thing nobody was actually buying

I thought clients were buying Devcore. I’d put a brand on the work. I’d built a website. I’d thought about positioning. I had a name with a capital D and lowercase everything else and a domain with two V’s. I was very organized about the appearance of the business.

What clients were actually buying was my time. They were buying the part where Myles, specifically, took a look at their problem and built the thing. The brand was decorative. The website was decorative. The thing they paid for was a guy.

This is fine if you know that’s what you’re selling. It’s not fine if you think you’ve built a company. The first sets a price; the second sets an expectation that breaks the moment you can’t be everywhere at once.

The high-performer trap

The honest version of why this happened: I’m good at what I do, and being good at what you do is its own kind of trap. When you can build the thing yourself, faster and better than the people you’d hire, every delegation feels like a tax. So you stop delegating. You take one more project. You take another. The agency grows by you working more hours, which is the one resource that doesn’t compound.

I tried to bring on a team anyway. I tried to run it the way I’d want to be managed — laissez-faire, just-do-the-thing, no micromanagement. I hired friends and family. I was too nice when I needed to be direct. I didn’t fire when I should have. I gave people projects without giving them the structure to actually own them, because owning them would have meant I didn’t own them, and somehow my brain refused to let go that far.

What this produced: a lot of anxiety, strain on relationships I cared about, and a quiet resentment of work I used to love. I was the bottleneck and the broker and the builder all at once.

What I think is actually true

I’m still learning this one — I’m not writing from the other side of it. But here’s where I am now:

Delegating is a skill, not a personality trait. You don’t get better at it by thinking you should. You get better at it by doing it before you think you’re ready, repeatedly, and being willing to be bad at it for a while. The muscle takes reps.

Build the org earlier than feels reasonable. When the company is just you, putting structure in place feels like LARPing being bigger. It isn’t. It’s how you stop being the bottleneck before the bottleneck breaks you.

Don’t trade time for money any longer than you have to. Hourly is a trap. Project-based is also a trap if the project is “Myles, but billable.” The point of a business is that it earns when you’re not in the chair. If your business stops earning when you stop sitting, you don’t have a business; you have a job with extra paperwork.

The agency’s job is to scale the output, not the founder. I conflated the two for too long. I thought scaling the agency meant doing more work, which meant being in more rooms, which meant me. Wrong. Scaling means more output happens with the same amount of me, and eventually with less.

Where I am now

Devcore still exists. I’ve stepped back from running it day-to-day — it’s mostly managed by my team while I’m full-time at Gravity. That arrangement only works because I finally let go of the part that said it had to keep going through me. The lesson cost me a few relationships and a lot of sleep, and I think it’s the lesson I’d most want to give the version of me who started Devcore.

You can’t scale yourself.

The earlier you accept that, the earlier you start building something that actually scales.