RAMP AND THE ONTOLOGICAL ADVANTAGE
How a more faithful ontology became a durable competitive advantage
Tyler Cowen once posed a question that has stayed with me: "What is it you do to train that is comparable to a pianist practicing scales?" My version of this practice is studying why certain startups break away from crowded markets to become the defining company in their category, understanding what structural insight led the founders to see an opportunity that others missed.
Plato suggested in the
Phaedrus that the art of understanding consists in carving nature at its joints, distinguishing things where they naturally divide rather than where habit has placed them. The metaphor is a butcher's. A skilled butcher does not hack through bone but finds the seams and separates along them. The same principle governs the construction of software systems. The companies that break away tend to be the ones that carved their domain correctly, imposing a structure that corresponds to the reality of the problem rather than to the inherited categories of the industry.
Ramp is one of the most instructive cases I have encountered. It has become one of the fastest-growing SaaS companies in history, reaching billions in annualized transaction volume within a few years of launch. The standard explanations for this kind of trajectory invoke execution speed, timing, or product polish, and while none of these are wrong, they are insufficient. Something deeper is at work, something that has to do with the way Ramp carved its domain.
Inherited Categories
Before Ramp, corporate financial software was organized around a set of categories that no one questioned. Expense management was one product, corporate cards were another, bill pay was a third, procurement was a fourth. Each had its own vendor, its own interface, its own data model, and its own logic. A CFO managing corporate spend would move between Concur for expense reports, a bank for corporate cards, Bill.com for accounts payable, and a procurement platform for purchase orders, each operating on its own ontology, as if these were fundamentally different domains.
What these systems treated as separate categories were in fact different views on a single underlying reality, money leaving the company. The prior products had imposed category boundaries where no natural joints existed, and in doing so, they rendered the most important control surfaces structurally invisible. There was no single representation of total corporate spend that a CFO could query, analyze, or enforce policy against, because the representation was shattered across 4 incompatible data models.
The Re-Carving
Ramp recognized that the right way to model this domain treats the fundamental entity not as an expense report or a card transaction or a bill but as an outflow of company capital. The fundamental process is not expense approval or card issuance or payment scheduling but the lifecycle of corporate spend from commitment to reconciliation. Seen from this vantage point, the old category structure reveals itself as arbitrary, a historical accident that had calcified into infrastructure.
What Ramp accomplished was an act of ontological creation. It brought into existence a unified computational object, corporate spend as a single trackable flow, that could not exist under the prior categorization. When spend was fragmented across 4 products with 4 data models, no one could see it whole. Ramp created the representation that made it visible, and in doing so, created the control surfaces that visibility makes possible. Total spend tracking across every channel became actionable for the first time. Policy enforcement could apply uniformly whether the spend originates from a card swipe, a bill payment, or a purchase order. Real-time budget awareness no longer depended on reconciling 4 separate systems after the fact.
Why the Advantage Holds
Incumbents face a version of the innovator's dilemma, but the dilemma is ontological rather than technological. Their existing customers, integrations, and workflows all depend on the old category structure. Rebuilding the data model means rebuilding everything that rests on top of it, which is why category re-carvings, when they succeed, tend to produce durable advantages. The new entrant builds on the right foundation from the first day. The incumbent must demolish and reconstruct while continuing to serve the customers who depend on the old structure.
Ontology as Moat
Incumbents in such markets are often unaware that their ontology is misaligned, because the categories feel natural. They have been in place for so long that they have acquired the appearance of inevitability. Expense management and bill pay seem like different things because they have been different products for decades. But seeming different and being different are not the same. What Ramp understood was that the deepest source of competitive advantage in software is not better engineering or better design but a more faithful ontology. The companies that win are the ones that find the joints.