There’s a line between building what you want and building what people need. I’ve spent years pretending I know where it is. I don’t. Not reliably. Not in advance.

Most of the features I’ve shipped across solo projects were ones I wanted, not ones anyone asked for. Some of them turned out to be the best bits. Most of them were just me building for myself and calling it product work. Hard to tell the difference until someone else uses it.

I think that’s worth being honest about.

The comfortable lie

When you build something alone, you tell yourself a story. You say you’re scratching your own itch, which is the respectable version. You say you’re your own first user, which sounds like strategy. You say you understand the problem deeply because you live it, which feels true and sometimes is.

But the uncomfortable bit is simpler than that. You build the thing you feel like building. Then you reverse-engineer the justification.

I’ve done this more times than I’d like to admit. I’ve sat with a mass of code at midnight, adding a feature nobody asked for, telling myself it was important. Sometimes it was. Sometimes I was just enjoying the building and needed a reason to keep going.

There’s nothing wrong with enjoying it. But calling it user insight when it’s personal preference is where things get slippery.

When instinct works

I built Tonen because I needed it. After my autism diagnosis at 41, I realised how much of my daily conversation was essentially scripted. I’d been building mental frameworks for decades to handle situations that other people seemed to move through without thinking. Phone calls, small talk, those moments where someone tells you bad news and you have to respond in real time.

Tonen was the product version of that. Conversation scripts for neurodivergent people. I built it for myself first, and it turned out other people wanted it too. Not because I’d done clever user research, but because the problem was real and specific and shared.

That’s the best case. You build from genuine need, the need is common enough, and the solution is honest enough that it translates.

But here’s what I’ve learned: that outcome is not the default. It’s the exception you remember because it validates the approach.

When instinct doesn’t

I’ve built features across solo projects that I was convinced were essential. Features I used daily. Features I thought were the whole point.

Then someone else tried the product and ignored them completely. Or worse, they found them confusing. The thing I’d built with total confidence turned out to be a mirror, not a window. It reflected what I wanted, not what anyone else was looking for.

DealSheet AI had this problem early on. I kept adding complexity to the deal analysis because I found the nuance interesting. More data points, more comparisons, more detail. I was building a tool for how I think about deals, which is not how most people think about deals. Most people wanted the answer faster with less noise. I was adding noise and calling it depth.

You can’t always see this from the inside. That’s the whole problem. When you’re the builder and the user, you lose the distance that makes good product decisions possible. Everything feels obvious. Everything feels necessary. You can’t tell what’s signal and what’s just your own fingerprints on the glass.

The gap nobody talks about

There’s a gap in the solo builder conversation that I don’t see many people acknowledge. We talk about shipping fast, about learning by building, about the value of having skin in the game. All of that is real. Building creates a kind of learning that discussion cannot reproduce. I believe that completely.

But we skip the part where your taste and your users’ needs diverge. Where your intuition, sharpened by twenty years of designing products for other people, still gets it wrong when you’re designing for yourself. Because designing for yourself removes the one constraint that makes design useful: someone else’s reality.

I’ve led design teams for years. In that context, you have research, you have feedback loops, you have other people telling you when something doesn’t make sense. When you’re building alone, you have to manufacture that friction yourself. And most of us don’t. We build, we ship, we wait for the numbers to tell us what we already suspect but don’t want to hear.

What I actually do now

I don’t have a framework for this. I don’t think frameworks help much here. But I’ve changed a few things after getting it wrong enough times.

I ship smaller. Not because of some lean methodology religion, but because the longer I build in isolation, the more convinced I get that I’m right. Shipping early is a way of interrupting my own certainty before it hardens into something useless.

I watch what people do, not what I expect them to do. This sounds obvious. It is obvious. I still have to remind myself constantly.

I try to separate “I want this” from “this is needed.” Not always successfully. But the act of asking the question, even privately, even just in my own head, changes how I hold the decision. It goes from obvious to uncertain. Uncertain is more honest.

And I’ve accepted that some of what I build will only ever be for me. That’s fine. Not everything has to be a product. Sometimes it’s just practice. Sometimes it’s just something you needed to make to understand what you actually think.

The honest version

Better UX cannot compensate for weak demand. I’ve written that down and I believe it. But I’d add something to it now: strong personal conviction cannot compensate for it either.

The best things I’ve built started with something I wanted. But they only became useful when I let go of the version that was perfectly shaped to my own preferences and let someone else’s reality reshape it.

That’s not a clean process. There’s no moment where you suddenly see clearly. It’s more like slowly accepting that the thing you built with so much care might need to become something slightly different to matter to anyone other than you.

Some builders never make that shift. They keep refining the thing they love, wondering why nobody else sees it the way they do.

I’ve been that person. More than once.

The trick, if there is one, is knowing that building for yourself is a brilliant starting point. It’s just a terrible finishing line.