A venture studio doesn't buy the building. It funds the layer above it (app, brand, data) while the asset stays ring-fenced. How the split works.
If a studio backs a real-asset venture, what exactly is it funding? Not the asset. The layer above it.
Disclosure: the venture used as the worked example here is in our own portfolio. The structural points are general; the figures we cite about the market are third-party and sourced.
In the last field note we said a software studio can back a real-asset venture as long as it builds the layer above the asset, not the asset itself. That raises the next question, and it's the one that actually matters to anyone evaluating the model. Where does the money go? What is the studio's capital buying?
This is the mechanism. We call it the Layer-1 model.
A real-asset venture, done this way, splits cleanly in two.
Layer 1 is the platform and brand. The app. The member and investor portals. The broker workflow. The data and intelligence the venture runs on. This layer is asset-light: software, design, and process. It scales the way software scales, and it's where a studio earns its keep.
Layer 2 is the real estate. The homes, the developments, the long-duration assets. This layer is asset-heavy by definition. It needs property capital, lender relationships, and patience measured in years, not sprints.
The studio funds and builds Layer 1. Layer 2 is financed separately and held in its own structures. One round doesn't pay for both, and one balance sheet doesn't carry both. That separation is the design, not an accident of timing.
The studio funds the layer that scales. The asset stays where the asset belongs: ring-fenced, on its own capital.
Because mixing them is how platforms break.
Put the real estate and the operating company on the same balance sheet and you get a familiar failure: a venture with fixed, long-term obligations against the asset, funded by the variable, early revenue of the platform. When the platform is young, that mismatch is fatal. The most cautionary cases in this space, the ones that filed for bankruptcy at high occupancy, generally didn't die of bad operations. They died of a capital structure that funded fixed liabilities with variable revenue.
The Layer-1 split is the structural answer to that. The platform's capital builds the platform. The asset's capital carries the asset. Neither subsidizes the other, so a soft quarter on one side can't pull down the other.
Concretely, Layer-1 capital funds the things that scale:
The product. The app, the portals, the workflow: the software a member, owner, or broker touches.
The data layer. The intelligence the venture builds on top of its own activity: pricing, matching, occupancy, absorption, service quality. This is a second product in its own right, and it compounds.
The brand and distribution rails. The broker environment, partner onboarding, the connective tissue that turns the product into a channel.
What it does not fund is bricks. No villa sits on the studio's books. The asset is acquired, leased, or developed inside Layer 2, in vehicles built for exactly that.
The asset side has its own discipline, and it's worth being specific, because "real estate venture" sets off alarms for good reason.
The durable approach is one vehicle per asset: a separate entity per property, ring-fenced, so a problem in one can't contaminate the others. The economics favor earning fees and carry on other people's capital and other people's assets rather than warehousing property on your own books. And in development, the strongest model is to build against capital that's held in trust and released only against verified construction milestones, so deposits can't be spent ahead of the work they're owed against.
There are bright lines, too. No fixed-rent master leases re-let at variable rates. That arbitrage has sunk more than one accommodation platform. No cross-collateralizing the vehicles, which would turn one default into a portfolio-wide one. No promising liquidity the structure can't deliver. The point of the model is that the asset layer is conservative by construction, which is what lets the platform layer move fast.
Emerging-market real estate adds a trust problem on top of the capital problem, and the answer is structural rather than promotional. In jurisdictions like the Dominican Republic, capital and title can be held inside a bank-administered trust (a fideicomiso) that segregates the assets from the developer's balance sheet and releases funds against milestones, while tourism-incentive regimes can improve the after-tax math on a qualifying project. We cover the mechanics, and the important line between "bank-administered" and "bank-guaranteed," in a dedicated piece. The relevant point here is that the protection comes from how the vehicle is built, not from anyone's say-so.
Read the two layers back to back and the division of labor is obvious. Layer 2 is a developer's and an allocator's world: capital, assets, duration. Layer 1 is a builder's world: product, data, rails, shipped in cycles.
A studio that tried to be the developer would be playing a game it can't win without a balance sheet it doesn't have. A studio that builds Layer 1 is playing exactly to type: financing and shipping the software-and-data layer that a real asset increasingly can't operate without, while the asset itself stays conservative and ring-fenced one layer down.
That's the model. The studio funds the layer that scales. The asset stays where the asset belongs.
Nothing here is an offer to sell a security or investment advice. Structural descriptions are general and illustrative; they are not legal, tax, or investment advice, and they are not representations about any specific company's structure or performance. Statements about market trends and legal regimes are drawn from third-party sources.
Field notes on venture building, AI, and capital. No spam, unsubscribe anytime.
By subscribing you agree to receive AOS Insights e-mails. We use your address only for this newsletter - see our Privacy Policy.
