What running a business for 7 years taught me about product

Before I was a PM at other companies, I ran my own business for seven years. It wasn't a startup that was sold. It was a freelance design studio that I closed in 2018.

That matters because the lessons are different. When you sell, you're optimizing for an exit. When you close a business, you're optimizing for what actually worked. The second teaches you more about product than the first.

Lesson 1: Shipping is the only thing that matters

Selfrules started as a design studio in 2012. I had ideas about design systems, process, methodology. I was going to revolutionize web design through structure and rigor.

Nobody cared about the methodology. They cared that I shipped a website in three weeks.

I learned this the hard way. I'd spend hours explaining my design thinking, talking about atomic design principles, user research, information architecture. The conversation would end with "that sounds great, how much and when can you deliver?" The methodology wasn't the product. The delivered website was.

By year three, I stopped selling the methodology and started selling the ship date. "I can do this in three weeks, and here's what you get." The business grew.

This taught me something important about product that applies even now. The strategy, the roadmap, the vision — those matter. But they matter only to the extent that they enable shipping. A team with a clear vision but can't ship is a team with a beautiful strategy and zero results. A team that ships without strategy might stumble, but at least they're learning.

Lesson 2: Clients know what they don't want

Here's a conversation that happened a thousand times:

Client: "I want a beautiful website."

Me: "What does beautiful mean to you?"

Client: "I don't know, but when I see it, I'll know."

This drove me crazy for years. Until I realized the client was right. They didn't know what they wanted. But they absolutely knew when they didn't want something.

I learned to reverse the conversation. Instead of "what do you want?" I'd say "what don't you want?" I'd show five different directions. "Which one feels wrong?" They'd eliminate two immediately. "Which of the remaining three feels closest?" We'd iterate from there.

The revelation: the customer's job isn't to articulate what they want. That's my job. Their job is to reject what they don't want. And they're very good at it.

This applies perfectly to product. User interviews that ask "what features do you want?" get vague answers. Interviews that ask "which of these flows frustrated you most?" get clear answers. Users don't know what they want to build. They know what broke.

Lesson 3: The best marketing is visible work

I didn't have a marketing budget. I had twelve websites shipped and a portfolio.

In 2014, I was struggling. Not broke, but looking for the next project. I decided to redesign my own website. Made it something I was proud of. Put up a case study of one of my recent projects, explained the decision, the problem, the solution. Nothing fancy.

Within a month, three people had found the site and asked if I did work for hire. One of those became a €60K project that sustained the business for six months.

The website didn't have a marketing message. It had proof. Here's a problem I solved. Here's how I solved it. Here's the result.

I stopped trying to explain why I was good and started showing the work.

This is why "build in public" became a thing. It's not new. It's the oldest lesson there is. When you show work, people who need that work find you. When you write about how you solved a problem, people with the same problem listen.

For a PM, this means: the best way to build trust with users, with your team, with leadership, is to solve visible problems. Not to talk about the roadmap. To ship something that matters and explain why.

Lesson 4: Closing a business is an ultimate prioritization decision

In 2018, I closed Selfrules. The business was stable. I was making decent money. I had recurring projects. But I'd been doing the same thing for six years and the craft wasn't challenging anymore.

The decision wasn't failure. It was the ultimate product prioritization call. "Given my finite time and energy, where do I create the most value?"

The answer wasn't running a freelance shop. It was learning development, building products at scale, understanding how teams work together. Closing Selfrules was saying "this isn't the best use of my resources anymore."

I notice a lot of PMs treat sunsetting features the same way they treat business failure. But sunsetting is just the same logic: we built this, we learned from it, it served its purpose, now it's not worth the maintenance cost. Let it go.

The hardest sunsets I've seen come from PMs who get attached to features they built. That's backwards. The feature isn't the success. The business outcome is. If the feature no longer drives business outcome, the question isn't "how do we save it?" It's "what's the most graceful way to transition away?"

Back to product

Running a business taught me that process is beautiful but shipping is required. That users are excellent at rejection and terrible at specification. That visible work opens doors. And that knowing when to close something is as important as knowing when to build it.

It's why CasaHunter exists — it's a side project that ships. Why I focus on solving visible problems. Why I listen for what frustrates users more than what they think they want. And why, when something stops mattering, I'm comfortable saying "let's stop doing this."

The business is closed. The lessons stuck.