The quantum threat to finance is usually framed as a technology choice. Which post-quantum algorithm should an institution adopt, when should it migrate, and how long will the transition take? Standards bodies have answered the first part of that question. Post-quantum encryption and signature schemes have been selected and published, and the migration path is no longer a research problem. Yet most banks, insurers and asset managers still cannot say with confidence where cryptography sits in their own estate.
That gap is the real constraint. An institution cannot migrate cryptography it cannot see, and in most large financial firms encryption is scattered across legacy applications, vendor products, hardware security modules, firmware, certificates and third-party connections that no single team owns. For the Global Finance Awards 2026, the distinction matters: quantum readiness is not a question of which algorithm has been chosen, but of whether discovery and governance have become operating practice.
The exposure is present, not future
The habit of treating quantum computing as a distant problem rests on an assumption that is difficult to defend. Adversaries do not need a working quantum computer today to harm an institution today. Encrypted traffic and stored data can be captured now and decrypted later, once the capability exists. Security specialists describe this as harvest now, decrypt later, and it changes the urgency calculation entirely.
For financial institutions, the data most at risk is the data held longest. Customer records, identity documents, deal pipelines, pension and claims data, and long-dated contractual information may remain sensitive for decades. A pension provider with liabilities stretching over thirty years cannot assume that a protection gap of ten or fifteen years is tolerable. The relevant question is not when quantum computers will arrive, but whether the retention period of a given data set outlasts the expected migration window. Where it does, the exposure already exists.
This is why urgency varies by business line rather than by firm. A trading operation with short-lived market data faces a different profile from an insurer holding medical histories or a trustee holding retirement records. A credible programme starts by identifying which data must remain confidential beyond the migration horizon, then working backwards to the systems and connections that protect it.
Cryptography hides in places no one owns
The second reason institutions struggle is structural. Cryptography is rarely a single function that a single team can locate. It is embedded in application code written years ago by developers who have since left. It sits inside vendor products whose internal algorithms are not disclosed. It lives in hardware security modules and device firmware, in certificates issued by teams that may no longer exist, and in keys stored outside any central inventory.
Third-party connections compound the problem. An institution's data may be protected by cryptography that a supplier controls and will not describe in detail. Contractual arrangements often say little about algorithm dependencies, leaving the institution unable to assess its own exposure in a service it depends on.
A useful way to think about this is that the estate has an invisible layer. Business inventories record applications and services; they rarely record the cryptographic primitives underneath. Without a deliberate effort to map that layer, migration planning proceeds on assumption rather than evidence. Teams discover dependencies during cutover, which is the worst possible moment.
The procurement lever
Because so much cryptography arrives through suppliers, procurement becomes a security function. Contract renewals offer a recurring opportunity to require vendors to disclose their algorithm dependencies and to commit to crypto-agile roadmaps. Institutions that treat those clauses as boilerplate lose the leverage that renewal provides.
Practical terms fall into a few categories:
- Disclosure obligations requiring vendors to identify which cryptographic algorithms and protocols their products use, and to notify customers of changes.
- Migration commitments setting out how and when a vendor will support post-quantum schemes, with a defined roadmap rather than a general statement of intent.
- Configuration and export rights allowing the institution to change algorithms, key lengths and certificate handling without waiting for a product release.
- Evidence rights permitting the institution or its auditors to verify that disclosed dependencies remain accurate.
These clauses are most effective when applied consistently across a vendor estate and tracked centrally, so that the institution can see which suppliers have disclosed, which have committed and which remain opaque. Over time the register of vendor dependencies becomes as important as the register of internal ones.
From register to rehearsed migration
The operating model that follows from this is unglamorous. It begins with a cryptographic register: a maintained inventory of where cryptography is used, which algorithms and key lengths are involved, who owns each dependency and what data it protects. The register has to be kept current, which means it is refreshed through change processes rather than compiled once as a project deliverable.
Ownership must extend beyond the security team. Technology owns applications and infrastructure; operations owns the day-to-day handling of certificates and keys; procurement owns the supplier terms. A register without named owners in each of those areas decays quickly.
Prioritisation should be risk-based, driven by data sensitivity and retention rather than by the convenience of the systems involved. A long-retention customer database protected by an ageing algorithm deserves earlier attention than a short-lived internal feed, even if the latter is technically easier to fix.
Finally, migration should be rehearsed rather than planned as a single replacement. Institutions that have practised swapping algorithms in one part of the estate, measured how long it takes and documented what broke, will approach a wider migration with evidence. Those that have not will discover their dependencies under pressure. Treating migration as a change and resilience discipline, rather than a one-off security upgrade, is what turns crypto-agility into an operating capability.
What the awards should measure
This is where the Global Finance Awards 2026 becomes relevant. Award categories in financial services have long rewarded strategy documents and public commitments. Quantum readiness is poorly served by that approach, because intent is cheap and the hard work is invisible from outside.
Assessors can, however, observe evidence of practice. A cryptographic register that is demonstrably maintained. Named ownership across technology, operations and procurement. Contract clauses actually applied at renewal, with a record of which suppliers have disclosed and which have not. A migration that has been tested in a defined part of the estate, with results and lessons recorded. Risk-based prioritisation that ties data retention periods to migration sequencing. These are observable artefacts, and they distinguish an institution that is preparing from one that has written a policy.
Crypto-agility is best understood as a governance and change-management capability. It is the ability to locate, prioritise and replace cryptographic dependencies on a planned schedule, and it can be evidenced.
The quantum debate will continue to focus on algorithms, because that is the part with a clear technical answer. For financial institutions, the binding constraint lies earlier and less visibly, in discovery and governance. The institutions that can describe where their cryptography sits, who owns it and how quickly it can be changed are the ones that will migrate on schedule. That capability, demonstrated rather than asserted, is the appropriate test for the Global Finance Awards 2026.
