As tokenisation use cases mature and connectivity across counterparties and infrastructure providers improves, institutions will increasingly find themselves occupying positions across issuance, distribution, custody, settlement, infrastructure, and servicing. Too often, those positions are shaped by a sequence of tactical decisions rather than a clear strategic choice. A business line finds a use case, a pilot launches, a vendor gets picked because its platform makes the pilot work, with custody and settlement following shortly after. Nobody stops to ask where the institution wants to sit in five years. They only ask whether the pilot works today.
The problem lies in the long-term vision, or lack thereof. While one institution treats its pilot as a one-off, a competitor treats the same use case as the opening move in a deliberate position, building the moat alongside the infrastructure rather than after it. The top positions in a value chain are scarce, and they don't stay open. By the time the first institution finally asks what role it wants, that position may already belong to whoever built toward it on purpose.
The problem with scaling these initiatives is structural, not necessarily technical. Most tokenisation programs are still run by digital assets and innovation teams, with business line leadership consulted but rarely accountable. No institution would let its technology team unilaterally decide who its counterparties are, yet that is effectively what happens when platform selection, not the business case, defines the organisation’s tokenisation role. These are decisions with potentially decade-long implications for custody, interoperability, and competitive position. Technology selection has become procurement, when the real questions are strategic:
In our conversations we have seen that most institutions are either: not answering these questions with intention or answering them in the wrong order.
The sequence most institutions follow starts with technology, then evolves into whatever the vendor's platform makes convenient, and treats client demand as something to validate afterward. One large financial institution spent 12 months building out a tokenised product that the digital asset team loved only to discover at go-live that the sales desk couldn’t sell it because no one had asked what clients would be willing to invest in or pay for. Another moved straight to vendor selection for a retail stablecoin and crypto trading product without first inventorying where else in the business the technology might create more value, some of it possibly internal.
When digital asset and innovation teams are the ultimate owners of the program, they naturally start where their expertise sits, with the technology, rather than where the commercial case sits, with the client.
Successful adoption begins with demand. What are the institution's actual clients, the counterparties, the funds, the corporates, asking for or would pay a premium for:
Those answers, not the value chain role, should determine the role an institution plays, and which use cases get resourced first. Only once demand has identified the use cases does the capability stack become concrete. Wallet infrastructure, smart contract development, custody architecture, collateral optimisation, transfer agency, and settlement orchestration are not something that institutions should build out all at once. They are a set of choices sized and sequenced against the specific client demand identified above. For example, an institution chasing collateral mobility needs a different capability stack, and a different urgency, than one responding to early interest in tokenised fund access.
That capability stack is what should shape the value chain decision, the build-versus-partner choice, and the vendor evaluations that follow it.
For strategy purposes, institutions can think about four primary roles: issuer, infrastructure provider, custodian and distributor. Each role is a different business introducing different capabilities, revenue economics, operating burdens, and control requirements.
Issuers own the entirety of asset’s onchain lifecycle: creation, corporate actions, redemption, and legal representation. This requires enhancing traditional securities infrastructure for onchain enforceability while retaining the same compliance and communication responsibilities required today.
Infrastructure providers operate the rails, the ledger, the smart contracts, and any interoperable layers built on top. This is the role with the steepest near-term operating cost: a real surge in technical and operations staffing to run and continuously maintain the rails.
Custodians hold the keys in the literal sense, in the form of private key management and figuratively in the sense of being the party clients and regulators hold accountable when an asset moves or doesn't.
Distributors own the client relationship and the point of sale without necessarily owning the rails or custody underneath. It is the lowest capital intensity role, and the one most institutions default into without deciding to, simply because it requires the least new infrastructure.
The strategic question institutions should ask is: Where can we capture differentiated value while operating capabilities efficiently and managing risk?
Not every capability needs to be built internally. The bigger question is which ones are strategic, differentiating, or risk-sensitive enough that the institution cannot afford to relinquish control.
Four questions can help determine this:
Does this materially affect why a client would choose us?
Are we still accountable if a third party operates it?
Does owning it improve economics at the scale we expect?
Would outsourcing it constrain our ability to change platforms or counterparties later?
This is also where strategy begins to shape the vendor selection process.
Vendor selection and lock-in risk (what it costs in time and money to leave) are one decision viewed from two angles, and both belong at the governance level and not in procurement. Choosing your provider will have numerous downstream effects including custody, interoperability, and competitive position long after the contract is signed. Who owns the ultimate outcome once the pilot ends is an extremely important question to answer up front.
A rigorous evaluation looks at five general categories: the vendor’s technical maturity and production track record, not just pilot performance; its regulatory posture and how it holds up across the jurisdictions the institution actually operates in; the strength of its balance sheet, since a platform failure is a worse outcome than a platform delay; how genuinely interoperable it is with other chains, custody providers and internal systems, versus interoperable in the pitch deck; and exit terms, what it costs in time and money to leave.
The fifth criterion deserves to sit above the other four rather than beside them. In a space where more than 80% of US spot Bitcoin and Ethereum ETF assets are held at a single institution2 there is a clear concentration that would draw regulatory scrutiny almost anywhere else in financial market infrastructure. It is exactly why those in traditional finance are moving to build or expand their own custody offerings and/or trust charters rather than depending on that concentration indefinitely, sophisticated institutions treating exit optionality as a strategic asset, not a contract afterthought. Exit terms are not one line item among five. They are the line item that boards need to think about and which determines whether the other four still matter in three years.
What governance needs to establish now is simpler: No material platform decision should be approved without evidence against all five criteria, and no platform decision should be approved without the exit terms already negotiated, not deferred to renewal.
Once an institution identifies where it wants to participate, it needs to determine whether it can support that role in production. Before committing, management should ask:
Can our legal function enforce and defend the structure today?
Can our technology environment support this in production, or does it still depend on a vendor's engineers to make the pilot work?
Does our risk function understand the exposure and have an appropriate control framework?
If a client, auditor or regulator tested the model tomorrow, could we explain who is responsible, how the activity is controlled and what happens when something fails?
Identifying capability gaps does not invalidate an enterprise transformation strategy, but rather presents 5 pathways: build internally, acquire capability, partner, narrow the scope of the use case or delay assumption of the role. What institutions should avoid is entering the role first and discovering the gap afterward. The roadmap should therefore distinguish between where the institution ultimately wants to compete and what it is prepared to operate today.
Tokenisation in production can impact various functions across an enterprise, and none of that redesign happens by accident or stays contained to one desk.
Front office moves from selling around tokenised product to quoting and settling it directly
Middle office absorbs a new reconciliation pattern once exception volume stops being special enough to staff separately
Back-office shifts from custodian-based settlement to something closer to onchain settlement operations
Legal moves from drafting one-off jurisdiction memos to running a standing playbook that can be applied to a new market in weeks
Technology moves from supporting isolated pilots to operating a shared production platform
Risk moves from assessing this as a technology question to running a full control framework adapted for the asset class, not relying solely on one built for something else.
What governance needs to own now is the fact that this change is happening across multiple functions all at once, on the same timeline whether it is planned for or not. The one structural change that makes the other six possible is standing governance itself not just a project steering group stood up for the pilot and quietly wound down once it ends but a real committee meeting on a fixed cadence, with authority to sign off before platform selection rather than after. Without that standing authority, technology moves ahead of legal, legal moves ahead of risk, and the institution ends up integrating in the wrong order.
Some responsibilities require new ownership, others simply move. Wallet governance is an example of the former. The technical controls, multi-signature approval, hardware key storage, usually exist. What is missing is a named owner accountable for the policies behind them, such as transaction initiation authority, how that access is granted and revoked, and who is responsible when something goes wrong.
The accountability gap, not a necessarily the lack of controls, is what exists in almost no institution today. Reconciliation is an example of the latter, especially at institutions furthest along where tokenised positions run through the same systems and staff who handle every other reconciliation. This is because the exception volume has become routine enough to absorb rather than staff separately. Naming the shift is one aspect, but what that looks like in detail, how it gets folded in and the actual reporting line changes will be addressed in upcoming Operations article.
Every argument in this article lands on the same point: a deliberate strategy, tested against real capability and backed by a governance body with actual authority, is what separates an institution that shapes its position in this market from one that discovers its position after someone else has already claimed it. That discipline matters most precisely because the ground underneath has not yet solidified.
The SEC’s proposal to redefine who qualifies as a custodian for digital assets, sent to the White House for review in August 2026, is one example of a pattern playing out across every major jurisdiction: a definitional shift that can reorder a build-versus-partner decision made in good faith six months earlier. A strategy built as if jurisdiction and custody definitions were fixed rather than moving is a strategy built on ground that is still being poured, which is where the next article in this series picks up.
1 SODA Information Services for Digital Assets, Tokenization at Investment Banks Survey, July 2026. [https://soda-services.com/tokenization-survey-2026/]
2 Yahoo Finance, Coinbase Locks in Conditional Approval to Become a Trust Bank, April 2026. [https://finance.yahoo.com/news/coinbase-locks-in-conditional-approval-to-become-a-trust-bank-153106649.html]
Global and US Digital Assets Lead
PwC United States
Digital Assets EU Lead
PwC Germany
Digital Assets APAC Lead
PwC Hong Kong
Digital Assets UK Lead
PwC United Kingdom