Why I think the small and mighty will win

I've believed this for a while. What's changed is that I'm watching it actually happen.

The cost of building software has collapsed. What used to require a team of five can now be shipped by one person in a few weeks.

AI tools are doing real work, not just automating boilerplate. The cost curve is still moving. This doesn't just help small operators; it fundamentally changes who can compete.

At the same time, distribution is fragmenting. The era of winner-takes-all platforms owning all the attention is cracking at the edges. Audiences are scattering into newsletters, communities, podcasts, niche social spaces. For large players trying to reach everyone, this is a problem. For small operators who already have a specific audience's trust, it's an enormous advantage.

Fast-moving, low-overhead operators who own their audience directly are no longer at a disadvantage against well-funded competitors. In many niches, they're ahead. Large organisations are slow by design. They have stakeholders to satisfy, processes to follow, brand risks to manage. A solo operator or a small team can move, test, and adapt in days. When the environment changes quickly — and it does, constantly — that speed compounds.

None of this means that large businesses stop existing. It means the conditions that made them dominant are weakening, and the conditions that favour small, focused, operator-owned businesses are strengthening. The niches that were too small to be worth serving are now commercially viable. The tools to serve them exist. The audiences are findable.

A business that serves 500 people extremely well, with low overheads and a direct relationship with its customers, is a genuinely strong business.

The fact that it's small is not a weakness — it's often the point.

I'm building for this. Indiemaker is predicated on it. The future of digital business is not one big thing — it's thousands of small, well-run assets, built by people who understand what they're building and why. That world is already here. It's just unevenly distributed.