Asset Tokenization: Three Structural Models and Why Most Projects Fail at the Wrong Point

The market for tokenized real-world assets has crossed a critical threshold. Institutional investors are actively asking questions. MiCA has established a binding framework. And platforms are emerging at a pace that far outstrips the regulatory maturity of most of the projects behind them.

The result is a quiet epidemic of failed ventures, not because the idea was wrong, but because the structure was.

We see it regularly. And we see exactly where it happens.

What Is the Regulatory Perimeter and Why Does It Determine Everything?

The regulatory perimeter defines which activities within a tokenization project fall under which regulatory regime and which party bears responsibility for each. That may sound technical. But it is the only question that ultimately matters.

A tokenization project is never composed of a single regulated activity. It issues tokens, potentially securities. It holds assets, potentially a licensed crypto custody business. It communicates with investors, potentially investment advice. It onboards clients, KYC, AML, accountability.

Who in your model bears regulatory responsibility for KYC, custody, and distribution? If that question cannot be answered with a document, the perimeter has not been drawn, it has merely been deferred.

What matters here is this: the perimeter is never held by a single party. It is allocated across the platform, the licensed banking partner, the custodian, and the AIFM or tied agent. That allocation (i.e., who bears what, who delegates what, who is liable for what) is the real structuring task. Not the choice of token standard.

Projects that do not answer this question before they build are building on the wrong foundation. Without exception.

The Three Structural Models and Their Typical Breaking Points

Model 1 — SPV: Deceptively Simple, Structurally Complex

A special purpose vehicle holds the asset, e.g., a solar installation, a receivables portfolio, an equity stake, a machinery pool, and issues tokens representing economic interests in it. Tried and tested, scalable, familiar to investors.

The breaking point: What exactly sits in the SPV — direct ownership of the asset, or only operational cash flows and usage rights? Failing to document this clearly creates a liability trap. The second classic mistake: the token is structured as a retail investment product under German capital investment law (VermAnlG) but is economically equivalent to a security. Regulatory inquiries follow and with them, the loss of confidence among early investors.

Regulatory classification follows economic substance, not the technical label.

Regulatory perimeter question: Who actually holds the asset? What rights does the token confer and against whom?

Regulatory framework: Token as security → MiFID II / KWG · crypto custody license may be required · prospectus obligation for public offerings · Alternative: VermAnlG / subordinated loan

Model 2 — AIF: Structurally Sound, but at a Price

The token represents fund units. This is the institutionally cleanest route, i.e., AIFM license, depositary, risk management. For professional investors, it is often the only acceptable structure.

The breaking point: The model is chosen too early, or for the wrong reasons. A full alternative investment fund manager (AIFM) license typically costs €500,000 or more to establish, on an ongoing basis. Sub-threshold variants are possible but carry clear limitations. And the deeper problem: in this model, tokenization is not the innovation. The fund is. Those who fail to understand that from the outset will find themselves managing governance overhead that paralyzes the actual business purpose.

Regulatory perimeter question: Is the AIF the vehicle, or a distraction from the actual product?

Regulatory framework: AIFM license (KVG) · depositary mandatory · sub-threshold AIFM possible · economically viable only above sufficient AuM

Model 3 — Platform: The Most Ambitious and the Most Dangerous

A marketplace on which multiple asset owners offer their projects and investors allocate capital selectively. The most strategically compelling model and the one with the largest perimeter problem.

The breaking point: The platform issues tokens — is it an investment firm? It holds assets — does it need a CASP license? It communicates with investors — is that investment advice? Failing to answer these questions in writing before the first line of code is written is the single most common reason projects of this kind are forced into restructuring six to twelve months after launch. Under time pressure, in front of live investors and partners.

Regulatory perimeter question: Which licensed parties bear which responsibilities and is the overall structure documented thoroughly enough to withstand partner due diligence?

Regulatory framework: CASP license under MiCA · investment firm (KWG / MiFID II) · tied agent as an alternative · allocation of responsibilities must be fixed in writing

Partner Due Diligence: The Underestimated Bottleneck

When a tokenization project approaches its first serious banking partner, custodian, or institutional investor, a review process begins that most teams are not prepared for.

Partner due diligence is not a formality. Banks and custodians will examine: How are transaction flows documented? Is there a coherent AML framework with risk classification, monitoring logic, and clear escalation paths? Is the governance structure designed so that accountability can be unambiguously assigned in the event of a dispute? Is the regulatory perimeter documented, not as a presentation, but as an operational document?

A project that cannot answer these questions at the first meeting will not lose the partner because of the business model. It will lose the partner because of the preparation. This is the most common and most costly mistake we observe. And it is entirely avoidable.

In our experience, well-prepared partner documentation has made the difference between a bank account application that was approved on the first submission and one that required three rounds of remediation. Between a custodian onboarding that took four weeks and one that stalled for six months. That is not a convenience. That is competitive advantage.

Where We Come In

Tokenization projects that engage us typically find themselves in one of three situations.

Partner Readiness: The technology is in place, but the first bank or custodian conversation has revealed gaps. We structure the AML framework, governance documentation, and onboarding dossier so that external partners become enablers rather than blockers. The goal: first attempt is the last attempt.

Execution Support: The strategic decisions have been made, but operational implementation exceeds internal capacity. We support execution, regulatory authority communications, and role definition for teams that would rather build than document.

Licensing Support: The next growth step requires an EU license. From regulatory classification through CASP and WpIG preparation to full application submission, across the EU and EEA. We know what national supervisory authorities expect from direct experience.

The question is rarely whether a tokenization project is regulatorily achievable. The right question is whether the regulatory perimeter was drawn correctly before the first partner draws it for you.

Those who define it themselves are in control of the conversation. Those who have to explain it have already lost it.