Founders love to believe that quality wins on its own. Build something good enough and the world will beat a path to your door. It's a comforting belief, partly because it's flattering — it means the hard, interesting work of building is also the only work that matters — and partly because the counterexamples are invisible. The graveyard of superior products that lost to inferior ones with better distribution doesn't put up headstones.
But that graveyard is enormous, and once you start looking for it you see it everywhere. The better product loses to the worse one all the time, and almost always for the same reason: more people could find and adopt the worse one. Distribution beat quality. Not because the market is stupid, but because a product nobody encounters might as well not exist, and encounter is a separate problem from quality that quality does not solve.
The usual mistake is to treat distribution as a phase that comes after building. First you make the thing, then you figure out how to get it to people. This sounds reasonable and is quietly fatal, because it means you designed the product without any thought for how it spreads, and how a thing spreads is partly a function of what it is. By the time you bolt distribution on, the shape of the product has already made it hard.
The better frame is that distribution is a design constraint, in the same way performance or usability is. You ask, from the beginning, how a new person finds this, and why they would tell someone else about it. Those questions change the product. A thing built to be discovered and shared is a different thing than one built in a vacuum and marketed later, and the difference is structural, not cosmetic.
Some of the most durable companies have distribution built so deeply into the product that the two can't be separated. The product gets more useful as more people use it, or using it naturally exposes other people to it, or the free version is itself the advertisement for the paid one. None of that is marketing in the usual sense. It's architecture. The growth is a property of the design, not a campaign bolted onto it.
I think the reason builders resist this is that distribution feels less honorable than building. Making the thing is craft; getting people to use it feels like sales, and sales feels like a tax on the real work. But that's a category error. Getting the right thing to the right people is part of making it good. A solution that can't reach the problem it solves is not, in any sense that matters, a complete solution.
This doesn't mean you should build a mediocre product and out-distribute everyone, though some companies do exactly that and win. It means the two are not separate competitions. Quality and distribution are both required, and the founders who treat them as one integrated problem — what to build and how it spreads, decided together — tend to beat the ones who solve building first and distribution as an afterthought.
The test is simple and worth asking out loud, early, while the product is still soft enough to change. How does a new user find this? Why would they tell someone? If you don't have answers, you don't have a marketing problem to solve later. You have a product problem to solve now, and the longer you wait, the more expensive it gets, because the answer is supposed to be baked into what you're building.