JOTO GLOBAL
← Insights

Post-Quantum Migration Starts with a Cryptographic Asset Inventory

Find certificates, keys, protocols and supplier dependencies before choosing replacement algorithms.

Technician inspecting equipment circuitry

Know where public-key cryptography is used

NIST approved FIPS 203, 204 and 205 in 2024 for post-quantum key establishment and digital signatures. Approval of standards does not mean an enterprise can replace every certificate or device at once. First identify internet-facing services, internal TLS connections, VPNs, code signing, identity systems, firmware updates and long-lived protected records. Record the algorithm, protocol, key length, owner, vendor, renewal date and the data lifetime for each use.

Map dependencies and priorities

A certificate may be generated by one team, stored in another platform and consumed by many applications. Link each cryptographic asset to the service it protects and the supplier that must support an upgrade. Prioritize data that must remain confidential for many years, critical communications and hardware with long replacement cycles. Validate whether suppliers offer interoperable implementations and whether current monitoring can identify algorithm use. This is an inventory and planning exercise, not a claim that a quantum computer is breaking current production systems today.

Define the decisions the inventory must support

The aim is to identify services that need migration, who can change them and which dependencies block progress, rather than count algorithms. In a hypothetical supply-chain portal, browser connections, partner interfaces, firmware signatures and historical contract verification may use different components. One public URL does not mean one website certificate captures the scope. This is a planning example, not a completed customer migration.

Business owners should state confidentiality periods, how long signatures must remain verifiable, and expected service life. Engineers then identify mechanisms and upgrade capabilities. Long confidentiality, long operating life and replacement difficulty are separate dimensions; do not label everything highest priority without explanation.

Inventory uses, not just algorithm names

Group transport protection, authentication, software and firmware signing, document signing and storage encryption. Record service, algorithm purpose, protocol, library or device version, certificate chain, owner and supplier support. Public-key algorithms, symmetric encryption and hashes serve different roles and do not all require the same replacement approach.

Store necessary metadata and references to controlled locations, not private keys, passwords or production key material. Cross-check asset records, configuration, dependencies and authorized observations, dating the evidence. Not finding a mechanism does not prove absence. Ask specific questions about managed services and closed devices instead of treating an empty scan as completion.

Ask suppliers verifiable questions

Beyond asking whether a product supports post-quantum cryptography, ask which component, version, purpose and protocol it supports, whether the implementation is experimental or supported for production, and whether adoption requires hardware or software changes. Confirm peer compatibility, certificate and key management, logging, support lifetime and incident responsibility. Require documentation or a test arrangement.

Keep roadmaps separate from delivered capabilities and record confirmation dates and contacts. Procurement can require adaptability, exportable inventory metadata and compatibility testing, but future support is not present acceptance. For critical dependencies without a timeline, retain alternatives and a reassessment deadline.

Combine risk priorities with business windows

Consider confidentiality duration, business importance, external exposure, dependency count and migration lead time. Long-lived confidential data may face a collect-now, decrypt-later risk; that is not evidence that quantum computers have already decrypted current data. Preserve the distinction and avoid unsupported dates for the threat.

Identify the dependency that governs the overall schedule, then coordinate with software updates, support renewal and replacement windows. Systems that cannot upgrade soon need explicit risk owners and transition measures. High, medium and low categories with explanations may be more useful than decimal scores that imply precision without evidence.

Test compatibility, performance and recovery

In a representative non-production environment, use supported implementations and peer combinations to test connections, certificate or signature verification, load, message-size effects and monitoring. Resource overhead varies; base conclusions on actual products and workloads instead of translating one algorithm experiment into procurement numbers for every service.

Preserve configurations and assign rollback conditions and owners before testing. Exercise incompatible clients, failed upgrades and certificate lifecycle operations, checking that support staff can diagnose them. Where fallback exists, verify its behavior and risk against the approved design. A successful connection alone does not establish use of the intended protection. State which paths remain untested.

Report progress through resolved dependencies

Security staff maintain scope and risk; application and platform staff confirm implementation; procurement follows suppliers; and business owners schedule validation. Each item needs current and target states, prerequisites, owner, completion criteria and evidence location. Supplier response, pilot success and production acceptance should remain separate states.

Deliver an inventory and pilot plan that guide the next upgrade, then expand coverage. Changes to certificates, libraries, devices, managed services or confidentiality needs should trigger updates. These are JOTOglobal preparation recommendations, not instructions to invent cryptographic algorithms. Technical choices must follow applicable standards, product support and organizational security review.

Plan migration as a programme

The UK NCSC recommends completing discovery and an initial plan by 2028, the highest-priority activities by 2031 and migration by 2035. These are UK guidance milestones, not universal legal deadlines. An enterprise can set its own timetable after risk assessment, test representative systems, preserve rollback options and update procurement requirements. Revisit the inventory whenever certificates, software or suppliers change.

Sources

Approval of Three FIPS for Post-Quantum Cryptography

https://www.nist.gov/news-events/news/2024/08/announcing-approval-three-federal-information-processing-standards-fips

Timelines for migration to post-quantum cryptography

https://www.ncsc.gov.uk/guidance/pqc-migration-timelines