Founder reflecting on startup lessons and personal growth

What I Learned From My First Startup

My first company didn't turn out the way I hoped. Here's what those years taught me about building a startup in a healthier, more sustainable way - and how it shapes the way I'm building Glitter.

Yuval Karmi
Yuval KarmiNovember 23, 2025

My first company, Simpo, didn't work out.

I raised $15.5 million and spent five years of my life on it. I hired a team I genuinely cared about. And then I had to wind it all down.

By venture capital standards, it was a failure.

I've written elsewhere about what went wrong. But here I want to dig into something different: what those years actually taught me, and how that's shaping the way I'm approaching my second startup, Glitter AI.

It wasn't the company. It was me.

Looking back, the biggest problem at Simpo wasn't product-market fit.

It was me.

I treated every single decision like it was a referendum on my worth as a person. Every setback felt like proof that I wasn't good enough.

Unfortunately, even when we did win, it didn't feel like redemption. Just a brief pause before the next thing went wrong.

That's an exhausting way to operate. It made the work harder, the team dynamics messier, and the decisions way more fraught than they ever needed to be.

I'm done with the pain olympics

Founders in Silicon Valley love talking about how hard they grind. The all-nighters. The "I haven't taken a vacation in three years." They say that "sleep is for the weak."

I bought into a lot of that at Simpo. I thought being a serious founder meant never stopping. Never slowing down. Never giving yourself a break.

I don't believe that anymore.

These days I try to work smarter, not just harder. I automate what I can. I bring in people who are better than me at the things I'm bad at.

When something feels overwhelming, I don't just power through. I stop and ask myself why it feels that way. Is there a better approach? Usually there is.

Here's the funny thing: Glitter is moving at roughly the same pace Simpo did. But I'm way less stressed.

Sanity over hype

Simpo was built in a typical VC (venture capital) environment. Aggressive growth targets, a constant sense that we were "behind," and metrics that looked great in pitch decks but meant almost nothing in practice.

I wanted Glitter to be different. Bootstrapped. Calm. Real customers paying real money. No fantasies about becoming the next Salesforce.

I want to be able to do this for ten years and actually enjoy the process. That's my bar now.

This is a do-over, and I'm not pretending otherwise

I'm not building Glitter to prove anyone wrong. That would just be falling into the same validation trap I got caught in at Simpo.

I'm building it because I want to know what it feels like to do this without the panic. Without wrapping my entire identity up in whether the company succeeds or fails.

Turns out that's harder than it sounds. But I'm having fun with it. I enjoy the product work, talking to customers, watching things grow at their own pace.

I even like most of the annoying parts (like writing comparison pages).

I really couldn't say that about much of my time at Simpo.

The difference this time

Glitter doesn't define me. It's something I'm making. Not proof that I deserve to exist.

And that mental shift, from "this company is me" to "this company is something I'm building," makes a huge difference.

The work feels lighter. I don't feel as anxious about the future.

And strangely enough, I think I'm actually better at the job now :)


If I could go back and tell my younger self one thing:

Build something because you care about it. Not because you're trying to prove your self-worth.

Your company can succeed or fail. That's completely separate from whether you're a worthwhile person.

Took me way too long to learn that one.

founder lessons
startup mistakes
motivation
Glitter AI
personal growth
Turn any process into a step-by-step guideGet Started

Turn any process into a step-by-step guide

Create SOPs and training guides in minutes
Glitter AI captures your screen and voice as you work, then turns it into step-by-step documentation with screenshots. No writing required.
Try Glitter AI Free