Moving to post-quantum cryptography (PQC) is less like a software upgrade and more like the Y2K programme: it touches almost every system, much of the work sits with vendors, and the deadline is uncertain but real. This roadmap breaks the work into five phases you can start now.
Phase 0: governance (months 0–3)
- Appoint an owner. A named programme lead, usually under the CISO, with support from enterprise architecture, procurement, and application and infrastructure teams.
- Get a mandate. A short board or executive paper explaining the risk (harvest now, decrypt later), the regulatory direction and the multi-year nature of the work.
- Set principles. For example: follow NIST standards (ML-KEM, ML-DSA, SLH-DSA); prefer hybrid modes during transition; use vendor-supported, validated libraries; and require crypto-agility in new systems.
Phase 1: discover (months 3–12)
You can't migrate what you can't find. Build a cryptographic inventory, increasingly called a cryptographic bill of materials (CBOM):
| What to find | Where to look |
|---|---|
| Protocols and versions (TLS, SSH, IPsec, S/MIME) | Network scanning, load balancer and gateway configs |
| Certificates and their algorithms | Certificate management tools, CT logs, scans of internal endpoints |
| Keys and HSMs | KMS and HSM inventories, secrets managers |
| Cryptographic libraries | Software composition analysis, SBOMs, build manifests |
| Algorithms in code | Static analysis for crypto API calls |
| Embedded and OT devices | Asset inventories, vendor documentation |
| Third-party connections | Partner and vendor integration lists |
For each item, record the algorithm, key size, owner, the data it protects and the data's shelf life, the vendor, and whether it can be upgraded.
Expect this to be the hardest phase. Automated discovery tools help, but owners have to fill gaps.
Phase 2: prioritise (months 9–15)
Score each system on:
- Data shelf life: how long the protected data must stay confidential or authentic.
- Exposure: internet-facing, third-party links and high-value targets first.
- Migration effort: quick configuration changes versus hardware replacement.
- Dependency: does it block others? PKI and HSMs often do.
That usually produces an order like:
- Internet-facing TLS and VPNs carrying long-lived sensitive data (hybrid key exchange).
- PKI and HSM upgrades, since many other migrations depend on them.
- Code and firmware signing (long-lived trust).
- Internal service-to-service traffic.
- Long-tail and embedded systems, replaced at end of life.
Phase 3: pilot (months 12–24)
- Enable hybrid key exchange (X25519 + ML-KEM-768) on a few internet-facing services and measure handshake size, latency and compatibility. See hybrid key exchange in TLS.
- Test PQC certificates in a lab PKI: issuance, chain sizes, client support and HSM support.
- Run interoperability tests with key partners.
- Find breakage early: middleboxes, old clients and protocols with message-size limits.
Phase 4: migrate (years 2–8+)
- Roll out by priority, aligned with normal refresh cycles where possible to keep costs down.
- Update standards and baselines so new builds use approved post-quantum or hybrid configurations by default.
- Track progress with metrics: percentage of external TLS endpoints with hybrid key exchange, percentage of critical systems with a PQC plan, and the number of systems that can't be upgraded.
- Decommission or isolate systems that can't be upgraded.
Vendors: the critical path
Much of your cryptography sits inside products you don't control: cloud services, network equipment, HSMs, databases, SaaS and devices.
- Ask now. Send key vendors a short questionnaire: their PQC roadmap, NIST algorithm support, hybrid modes, crypto-agility, and timelines.
- Change procurement. Add PQC requirements to RFPs and contract renewals: support for the NIST standards or a committed roadmap, plus a CBOM.
- Avoid lock-in to dead ends. Don't buy long-lived hardware that can't be upgraded.
Align with external timelines
Regulators are converging on the early-to-mid 2030s for completion. NIST's draft transition guidance proposes deprecating quantum-vulnerable algorithms by 2030 and disallowing them by 2035. The EU roadmap targets high-risk systems by 2030. India's 2026 task force report sets a phased national timeline with critical sectors first. See India's quantum-safe roadmap. Sector regulators may set their own dates; financial services and telecoms are likely to move first.
Common pitfalls
- Waiting for certainty about when quantum computers will arrive. Inventory and agility pay off regardless.
- Treating it as an IT project only. Procurement, legal (contracts) and business owners all have roles.
- Rolling your own. Use mature, validated libraries and vendor implementations.
- Ignoring performance and size. Larger keys and signatures can break constrained devices and protocols. Test.
Learn more
Start with post-quantum cryptography explained and the field guide to the NIST standards. For cryptography and PKI fundamentals, see the free CISSP track.
Written and reviewed by the DigiTrust Asia editorial team. How we research, review and correct articles.
Strengthen your crypto fundamentals
Cryptography lessons and practice questions in the free CISSP track.
Create a free account