Nuclear Consulting Limited

Routes to design approval: GDA or site licence?

15 September 2026

A vendor or developer bringing a reactor design to the UK faces an early choice that shapes everything after it: pursue Generic Design Assessment, or go directly to site-specific licensing and permitting. The choice is often presented as a regulatory question. It is better understood as a commercial one, because the two routes distribute cost, risk and optionality quite differently.

Temelín nuclear power station, Czechia

What each route is for

Generic Design Assessment is a voluntary process in which the Office for Nuclear Regulation and the Environment Agency, together with Natural Resources Wales where relevant, assess a reactor design in the abstract, without reference to a particular site. Completion of a full three-step GDA produces a Design Acceptance Confirmation and a Statement of Design Acceptability; a two-step GDA does not. It does not authorise construction anywhere; it establishes that the design, as submitted, is capable of being built and operated acceptably in Great Britain.

The site licensing route runs the design assessment as part of the site-specific case. The licensee applies for a nuclear site licence and the necessary environmental permits, and the design is assessed in the context of that site, that operator and that safety case. There is no generic output; the permissions attach to the project. A nuclear site licence alone does not authorise nuclear safety-related construction or operation; further regulatory permissions and hold points apply.

The structural differences that matter

Who carries it. GDA is run by the requesting party, typically the vendor or a vendor-led entity, and does not require a licensee. Site licensing requires an applicant capable of holding a nuclear site licence, which is a substantial corporate and organisational test in its own right. For a vendor without a UK licensee partner, GDA is the available route.

What it produces. GDA produces a transferable asset. A design that has completed assessment can be offered to multiple prospective customers at multiple sites, with the generic regulatory position already established. The site route produces an approval for one project. For a design intended to be deployed repeatedly, that difference is the heart of the matter.

When the money is spent. GDA front-loads significant cost before any site, customer or financing is secured. That is the price of the transferable asset, and it is a genuine barrier for developers who cannot fund a multi-year regulatory programme on the prospect of future deployment. The site route defers regulatory spend until a project exists, at the cost of carrying design risk into the project.

Where the uncertainty lands. GDA resolves generic design questions early, which removes a large category of risk from every subsequent project. It cannot resolve site-specific questions, so the site stage still happens. The site route resolves everything at once, which is efficient for a single project and means that design issues surface when the project is already committed.

Two-step and three-step

GDA is organised in three steps, of which Step 1 is initiation, in which the requesting party and the regulators prepare for assessment; Step 2 is a fundamental assessment of the design and its safety, security, safeguards and environmental case; and Step 3 is a detailed assessment. A requesting party can choose a two-step GDA, which stops once the fundamental assessment is complete and ends with the regulators’ statement of their findings, or a three-step GDA, which continues into detailed assessment. Completing Step 3 is the only route to a Design Acceptance Confirmation and a Statement of Design Acceptability.

A two-step GDA provides findings following fundamental assessment without the design acceptance outputs associated with completion of Step 3. Detailed assessment must still be completed through the applicable site-specific processes and regulatory hold points. The choice should reflect design maturity and the evidence investors require before committing capital.

In its May 2026 guidance, ONR explains that Step 1 now runs as a distinct preparatory activity outside the regulators’ target of completing design assessment within two years, and that the expectations for a Step 2 fundamental assessment are being aligned with what ONR needs to see before granting a nuclear site licence. Information equivalent to a Step 3 detailed assessment is still required before ONR releases the hold point for nuclear safety related construction, and currently for operational environmental permits. A single integrated framework for early engagement and GDA is due by spring 2027.

Designs that have been licensed in another jurisdiction are in a particular position. The evidence is real and valuable, but the UK’s goal-setting regulatory approach asks different questions from prescriptive regimes, and ALARP demonstration in particular has no direct equivalent in most other systems. Entering GDA with a mature prescriptive case and expecting it to satisfy a goal-setting regulator is a well-documented way to lose time.

Which is better?

Neither, as a general proposition. The sensible test is a small number of questions.

Is the design intended for repeat deployment in Great Britain? If yes, GDA’s transferable output is likely to justify the front-loaded cost. If a single project is the realistic ambition, it probably does not.

Is there a credible UK licensee today? If not, GDA is the practical route, because the site process needs an applicant that can pass the licensing test.

Is there a committed site and customer? A project with both may find the site route faster, because it avoids assessing generically what will have to be assessed specifically in any case.

Can the regulatory programme be funded without a project? This is where several designs have come unstuck. GDA is a substantial sustained commitment, and a programme that stalls mid-assessment is worse than one that never started.

The common thread

Whichever route is chosen, the determining factor in how it goes is rarely the route itself. It is whether the submitting organisation understands what a goal-setting regulator is asking for, has built its safety case around ALARP and BAT demonstration from the start rather than retrofitting it, and has the specialist capability to engage substantively rather than defensively. Designs that have those things progress on either route. Designs that do not struggle on both.

NCL advises vendors, developers and investors on regulatory strategy, GDA readiness and the licensing pathway. Get in touch to discuss your route.

Contact

Speak to our team about your project.

Contact NCL