The moat moved
Software was the barrier. Then it wasn't. Here's what replaced it.
Issue #06 · June 13th 2026
I write about the things that don’t change in a world that won’t stop changing.
I believe adding real value is what gives us purpose, keeps us happy and improves our lives.
For two decades, software was the moat. If you could build it and your competitor couldn’t, you won. The ability to ship a working product was itself a competitive advantage. The code was the castle.
That era is over.
What made software a moat
The barriers were real. Engineers were scarce and expensive. Infrastructure had to be built and maintained. Shipping a production-grade product took months, sometimes years, and required a team with specialised skills that most organisations couldn’t assemble quickly. The compounding advantage of a mature codebase, one that had been tested against real users, hardened against edge cases, and integrated into customer workflows, was genuinely difficult to replicate.
These constraints shaped everything. Valuations were built on them. Hiring strategies were built on them. The entire venture capital model of the 2000s and 2010s was built on the assumption that technical execution was the hard part, and that whoever solved it first and fastest would be difficult to dislodge.
They also created a generation of founders who confused the ability to build with the reason to build. If shipping was hard, then shipping was the achievement. The question of whether anyone needed what you shipped was secondary, something to sort out once the technical problem was solved.
“A generation of founders confused the ability to build with the reason to build.”
What collapsed it
Not one thing. A sequence.
Cloud computing commoditised infrastructure. You no longer needed to own servers to operate at scale. The capital expenditure that once separated serious players from aspirants became a monthly subscription anyone could afford.
Open source commoditised the stack. The frameworks, the databases, the tooling — everything that once required years of engineering effort to build from scratch was now available, maintained by communities, and free to use. The technical foundation of most software products stopped being a differentiator the moment it became available to everyone simultaneously.
And then AI commoditised the act of writing code itself. The last remaining barrier, the human expert who could translate a business problem into working software, became assisted, accelerated, and in many cases replaceable. What took a team of ten and six months now takes one person and a weekend. Sometimes less.
Each layer of the barrier dissolved in turn. The moat was drained from the outside in, over twenty years, until there was nothing left to defend.
Where the moat moved
The moat did not disappear. It moved.
Distribution is a moat. The ability to reach the right customer, at the right moment, with enough trust already established to have a real conversation. That does not come from a codebase. It comes from years of showing up, building a reputation, and earning access. It cannot be replicated overnight by someone who just decided to enter your market.
Customer relationships are a moat. Not the CRM record. The actual relationship, the understanding of how a specific organisation makes decisions, who the real stakeholders are, what they have tried before and why it failed, and what success looks like in language that matches their internal vocabulary. That knowledge is not transferable. It lives in the person who built it.
Proprietary data is a moat. Not data you bought. Data you generated by operating, by running transactions, by watching how customers actually behave rather than how they say they behave, by accumulating signal over time that nobody else has access to because nobody else was in the room.
Domain expertise so deep it cannot be faked is a moat. Eight years running a mobility platform teaches you things about enterprise fleet management, about how automotive companies evaluate vendors, about the gap between what procurement says it wants and what operations actually needs, that no amount of research can replicate. That expertise is yours. It is not in any training dataset in a form anyone else can use the way you can.
“The moat did not disappear. It moved.”
What this means for how you build
If the moat is no longer technical, the dangerous thing is spending your time and capital as if it still is.
The founders who lose in this era are not the ones who cannot build. Building is no longer the filter. The founders who lose are the ones who build beautifully and distribute nothing. Perfect product, empty pipeline. Technically impressive, commercially invisible.
The allocation question has inverted. In 2010, the right move was to put most of your energy into the product and trust that distribution would follow from quality. In 2026, quality is assumed. Anyone with a weekend and access to the right tools can produce something technically competent. The scarce resource is the relationship, the channel, the trust that gets your product in front of the right person at the right moment.
This does not mean stop building. It means build fast, build cheap, and spend the time you save on the things that cannot be automated. The conversations, the relationships, the accumulated understanding of a specific problem in a specific industry that makes your solution the obvious one rather than just another option.
The invariant
The moat was never really the software. It was always the insight the software encoded, the understanding of a real problem earned through proximity to real customers, that made the product worth building in the first place. And it was always the relationships that gave you access to that problem before anyone else knew it existed.
The technical layer sat on top of those things and made them look like the point. Now that the technical layer is free, the underlying assets are exposed for what they always were.
The castle was never the code. It was what you knew, and who trusted you enough to let you solve it.
Fernando Martín is Managing Director of NEXMO Movement Data Hub (UC3M), Venture Builder at MOVEN, and founder of Eccocar. He writes here about venture building, AI agent operations, and the European technology landscape.
The Invariance — by Fernando Martín In a constantly evolving world, only value is the invariance that holds everything together.
Thanks for reading. If this resonated, share it with someone who needs to read it.