The Nuclear Power Market in 2026

The global nuclear power plant market was valued at roughly $33.53 billion in 2025 and is projected to reach $38.77 billion by 2030, according to The Business Research Company's 2026 market report. Sitting inside that figure is a much faster-growing segment: small modular reactors (SMRs), valued at $6.54 billion in 2025 and forecast to reach $10.69 billion by 2033 at a 6.8% CAGR, per Grand View Research.

Context matters here: for roughly four decades after Three Mile Island's 1979 partial meltdown, almost no new large reactors broke ground in the US or UK, and the handful that did (Vogtle, Hinkley Point C) became case studies in cost overrun rather than a template anyone wanted to repeat. What changed in 2024-2026 wasn't the underlying technology so much as the buyer: instead of a regulated utility slowly recovering costs from ratepayers over decades, a small number of investment-grade hyperscale technology companies started signing direct, multi-decade contracts to guarantee firm power for AI data centres, which is a fundamentally more bankable revenue structure than the merchant power market nuclear plants used to compete in.

Industry analysts have started calling 2025-2026 the start of a "Golden Age of Nuclear," and the numbers back that framing up. A US executive order targeting 10 new large reactors under construction by 2030 sits alongside a wave of corporate demand nobody was forecasting five years ago: hyperscale data-center operators signing direct power purchase agreements with nuclear plants to guarantee firm, carbon-free electricity for AI infrastructure.

Two things make this market unusual to plan around. First, revenue is almost never sold on the spot market: it is locked in years ahead through a power purchase agreement (PPA) or, in the UK, a regulated tariff. Second, the buyer list has changed. Microsoft signed a 20-year, 835MW PPA with Constellation Energy to restart Three Mile Island Unit 1 (targeted for 2028), described by Constellation as the largest PPA it has ever signed; Amazon Web Services followed with a 17-year, 1.92GW PPA for output from Talen Energy's Susquehanna plant; and Google signed the first corporate SMR PPA with Kairos Power in August 2025 for its Hermes 2 project in Oak Ridge, Tennessee.

Who this guide is actually for: almost nobody funds a full-scale reactor with a $5 template, and this page doesn't pretend otherwise. It's written for two overlapping audiences: (1) founders building a nuclear-adjacent services, staffing, inspection or supply-chain business who need a properly researched, lender-ready plan, and (2) analysts, students and small development teams who need an accurate picture of how reactor-scale projects are actually licensed and financed, so the assumptions in a bespoke plan hold up under investor or DOE scrutiny.

The policy backdrop matters for both audiences. A May 2025 US executive order, "Reinvigorating the Nuclear Industrial Base," set a target of 10 new large reactors with complete designs under construction by 2030, and the DOE has since attached real financing to that target rather than leaving it as a slogan. On the workforce side, the US Bureau of Labor Statistics puts the median annual wage for nuclear power reactor operators at $122,610 (range $99,300-$152,690), nuclear engineers at $127,520, and nuclear technicians at $104,240 across roughly 6,000 US jobs: figures worth anchoring a staffing budget to, whether you're planning crew costs for a services business or a management-team section for a developer (BLS). That's a meaningfully different labour market from conventional (non-nuclear) power plant operators, whose median wage sits lower at $103,600 and whose employment is projected to decline through 2034 as older fossil and legacy plants retire, a contrast worth naming explicitly if your plan is trying to convince a lender that nuclear-sector staffing costs are structurally different from the wider utilities sector.

None of this changes the basic economics for a small business entering the sector: whether you're supplying inspection services, dosimetry, specialty welding, or project-management support, the demand pool is driven by the same reactor pipeline described above, and it is growing for the first time in a generation.

People Also Ask

Questions that come up repeatedly once someone starts researching a nuclear-sector business plan:

What is a small modular reactor (SMR) and how is it different from a traditional plant?

An SMR is a nuclear reactor design generating up to roughly 300MW (versus 1,000MW+ for a traditional large reactor like Vogtle or Hinkley Point C), built largely from factory-fabricated modules shipped to site rather than poured and welded in place over a decade. The pitch is lower absolute capital cost per project, shorter build times, and a licensing pathway (NRC Part 53, in the US) designed to be faster and cheaper than the traditional route. NuScale, X-energy, Oklo, Kairos Power, TerraPower, GE Hitachi and Rolls-Royce SMR are the most active developers as of 2026.

How long does it take to build a nuclear power plant?

A traditional large reactor takes 10-15 years from application to commercial operation once delays are accounted for; Vogtle Units 3 and 4 came online roughly seven years late. Advanced reactors are targeting much shorter construction windows: Kairos Power's Hermes 1 demonstration reactor completed its formal NRC review in 18 months, and the Natrium project in Kemmerer, Wyoming had its Construction Permit review completed in roughly 22 months. Add pre-application engineering, site permitting and long-lead-item procurement, and a realistic timeline from first capital raised to commercial operation still runs 5-8 years even for a first-of-a-kind SMR.

How much does it cost to build a nuclear power plant?

Vogtle Units 3 and 4 cost roughly $35 billion against an original $14 billion estimate. Hinkley Point C is now projected at £48-49 billion (2025 prices), about 25% higher per MW than Vogtle. Sizewell C, financed under the UK's Regulated Asset Base model, is budgeted at approximately £38 billion for two EPR reactors totalling 3.2GW. SMR developers are targeting a fraction of that per project: TerraPower has publicly discussed a roughly $1 billion first-of-a-kind cost for its Natrium plant, though first-of-a-kind projects in this sector have a well-documented history of cost growth.

Startup Costs & How Nuclear Projects Are Financed

Two very different scales apply here, and a credible business plan needs to be honest about which one it's targeting. A nuclear-adjacent services or supply-chain business (inspection, radiological safety, staffing, decommissioning support, specialty fabrication) typically launches for $85,000 to $450,000 (£60,000-£350,000). A reactor developer's pre-construction phase alone (engineering, licensing, site work, before any long-lead items are ordered) runs into the tens of millions; the long-lead items themselves (reactor pressure vessels, steam generators) commonly run $300 million-$800 million per unit, and a full EPC build of a single first-of-a-kind SMR is in the $1-2.5 billion range.

The single biggest driver of that spread is equipment ownership versus leasing. A one-crew radiography or dosimetry startup can launch on leased gamma/neutron detection instruments and a single certified technician; a multi-crew operation that owns its own equipment, carries its own transport isotopes under license, and can staff simultaneous outages at two or three sites needs the full $450K to cover redundant instrumentation, a second vehicle, and 6 months of working capital rather than 3. Either way, insurance and bonding is consistently the second-largest line item, because general liability carriers price radiological work as a distinct, higher-risk category from ordinary industrial inspection.

Funding Routes at Small-Business Scale

If you're building the services or supply-chain business rather than the reactor itself, conventional small-business finance still applies. In the US, SBA 7(a) loans (up to $5M) and equipment financing are the standard routes; in the UK, the Start Up Loans scheme (up to £25,000 at 6% fixed) with free mentoring covers early-stage capital. A written business plan with financial projections is a hard requirement for almost every application at this scale, and lenders reviewing a nuclear-adjacent applicant will specifically ask to see the certification and insurance costs broken out as their own line items, since these are unusually large relative to a similarly-sized business in an uncontrolled industry. Equipment financing and leasing lines are also worth building into the plan explicitly: dosimetry and radiography instrumentation depreciates quickly and is commonly refreshed on a 3-5 year cycle, which lenders will want to see reflected in the cash flow forecast rather than treated as a one-time Year 1 cost.

Funding Routes at Reactor-Project Scale

SBA's $5 million cap explains why it never appears in reactor-scale financing: the numbers are two to three orders of magnitude larger. In June 2026, the US Department of Energy's Office of Energy Dominance Financing issued a $17.5 billion conditional loan commitment to finance long-lead items across five projects sponsored by utilities partnering with Westinghouse, targeting 10 new large reactors under construction by 2030 (Utility Dive). The UK's Regulated Asset Base (RAB) model financed Sizewell C to financial close on 22 July 2025, combining a National Wealth Fund debt facility with £5 billion of Bpifrance export credit; a RAB levy of £3.455/MWh has applied to UK electricity bills since 1 November 2025 to fund construction-phase returns (GOV.UK, World Nuclear News). Private capital markets have also opened up: X-energy completed a $1.02 billion IPO in April 2026, Oklo now trades at roughly a $12.9 billion market cap as the largest SMR pure-play, and NuScale holds a $1.35 billion DOE grant on top of roughly $1 billion in liquidity.

The Reactor Supply Chain

A supply-chain or services business plan is much stronger when it names exactly where in this chain it sits, rather than describing itself generically as "nuclear industry." Westinghouse is the anchor long-lead-item supplier partnering with utilities under the DOE's 2026 loan programme (reactor pressure vessels, steam generators); GE Hitachi supplies the BWRX-300 SMR design used at Darlington and by several US developers; Rolls-Royce SMR is the leading UK-based SMR vendor; Doosan and BWXT are major heavy-forging and component-fabrication suppliers (reactor vessels, steam generator shells); and Holtec is active in both SMR development and spent-fuel/decommissioning services. A business plan for a fabrication, inspection, or logistics company should identify which of these primes it is (or intends to be) a qualified sub-tier supplier to, since that relationship, not the general market size figures above, is what a lender or investor will actually underwrite.

Where Nuclear Projects Are Being Built

Siting drives nearly every other assumption in a nuclear business plan, from workforce costs to interconnection queue position to which regulator you're dealing with. A handful of locations account for most of the active 2025-2026 pipeline:

For a services business, the practical takeaway is simpler than the table looks: cluster near an operating or under-construction site (Georgia, Pennsylvania, Tennessee, Wyoming, or a UK/Ontario equivalent) and your addressable market of inspection days, staffing contracts, and vendor qualification opportunities is materially larger than trying to build a national footprint from a cold start.

Two siting factors are worth naming explicitly in a plan, because they change every downstream assumption: cooling water access and grid interconnection queue position. Large reactors need substantial cooling water (river, lake, or coastal), which is why the US pipeline clusters around the Southeast and the Great Lakes/St. Lawrence corridor rather than being evenly spread across the country; several advanced-reactor designs are specifically marketed on reduced cooling-water requirements as a siting advantage. Interconnection queues, separately, are a US-specific bottleneck: a project can be fully licensed and still wait 2-4 years for a grid interconnection study and upgrade in a congested regional transmission organisation's queue, which is exactly the kind of delay a lender will expect a plan to acknowledge rather than assume away.

Workforce Availability by Region

Workforce availability tracks the same geography. Existing nuclear-fleet states (Georgia, South Carolina, Pennsylvania, Illinois) already have a trained labour pool of licensed reactor operators, radiation protection technicians and NDT inspectors who can move between plants; new-build states like Wyoming are building that pipeline from a much smaller base, often through direct partnerships between the developer and a local community college or university. For a small services business, an existing-fleet state is usually the easier place to hire experienced staff quickly; a new-build state can offer a first-mover advantage with less established competition, at the cost of a thinner local talent pool to recruit from.

Revenue Model & Profit Margins

Nuclear revenue is contracted, not sold spot. At reactor scale, output is committed years in advance through a PPA, typically 17-20 years, increasingly with a hyperscale data-center operator rather than a traditional utility offtaker. In the UK, the RAB model replaces the PPA with a regulator-set allowed return during construction, funded by the per-MWh consumer levy described above.

Worked example, reactor scale: an 835MW unit selling 100% of output under a 20-year PPA at $110/MWh, running at a 92% capacity factor, generates approximately $739 million in annual contracted revenue (835,000kW × 8,760 hours × 0.92 × $110/MWh ÷ 1,000). Once construction debt is serviced, operating margins of 45-65% are typical, since fuel and O&M costs represent a comparatively small share of revenue at capacity factors above 90%.

Worked example, services scale: a nuclear inspection contractor running 6 crews at a $1,850 day rate and 70% utilisation generates roughly $2.84 million in annual revenue (6 crews × $1,850 × 365 days × 0.70). Net margins for this segment typically run 15-30%, reflecting labour, certification, insurance and equipment-depreciation costs rather than fuel and O&M.

The margin gap between the two segments is the single most important number to get right in a plan: a services business pitched with IPP-style 50%+ margins will not survive lender scrutiny, and a reactor project pitched with services-style 20% margins will look under-capitalised to a project-finance investor.

Revenue diversification looks different at each scale, too. A reactor developer's realistic diversification options are limited to the terms of the PPA itself: contracting with more than one offtaker to reduce counterparty concentration, negotiating capacity payments alongside the energy price, or, in a regulated market, layering a RAB-style construction-period return on top of eventual operating revenue. A services business has more conventional levers: adding a second service line (for example, moving from radiography inspection into staff augmentation or outage-support staffing), securing multi-year maintenance framework agreements rather than one-off inspection call-outs, and building a preferred-vendor relationship with a specific plant or developer so revenue is contracted a year ahead rather than won job-by-job. Utilisation is the variable that moves the needle fastest for a services business: the worked example above assumes 70% crew utilisation: moving that to 85% adds roughly $690,000 to annual revenue with almost no added fixed cost, which is exactly the kind of sensitivity a lender or investor will want to see modelled explicitly rather than asserted.

For a reactor-scale plan, the equivalent sensitivity is capacity factor: dropping from a 92% to an 85% capacity factor on the 835MW worked example above reduces annual contracted revenue from roughly $739 million to $683 million, even though the PPA price itself hasn't changed. Both examples make the same underlying point: the headline price or rate is only half the model, and the other half is how reliably the asset or the crew actually shows up to generate against it.

Licensing & Regulatory Requirements

United States

A reactor developer needs a Combined License under the traditional 10 CFR Part 52 pathway or a Construction Permit/Operating Licence under the newer Part 53 advanced-reactor pathway from the Nuclear Regulatory Commission. The NRC bills review time at $337/hour standard, or $154/hour at the reduced rate available to advanced-reactor applicants and pre-applicants. Timelines vary enormously by pathway: Vogtle 3&4's Combined License review took approximately 46.5 months from application to Commission vote; Part 53 is designed to compress that to roughly 18 months, and Kairos Power's Hermes 1 review actually completed in 18 months against a 21-month accelerated schedule (NRC). Services businesses face a much narrower set of requirements instead: 10 CFR Part 34 registration for industrial radiography, Part 21 defect-reporting obligations for suppliers, state radiation-control agency registration, and site-specific background checks/security clearances for plant access.

United Kingdom

New reactor designs go through a Generic Design Assessment (GDA) run jointly by the Office for Nuclear Regulation (ONR) and the Environment Agency, in three steps: Step 1 (scope and schedule, ~12 months), Step 2 (fundamental design assessment, ~12 months), and Step 3 (detailed assessment to construction-readiness level, ~24 months): roughly 4 years end-to-end. A separate nuclear site licence from ONR is required before construction or operation can begin at a specific site; full cost breakdowns are only partially disclosed under FOI (ONR).

Canada (Third Jurisdiction)

The Canadian Nuclear Safety Commission (CNSC) issues a power reactor construction licence followed by a separate operating licence. Ontario Power Generation applied in October 2022 and was granted a construction licence for one GE Hitachi BWRX-300 unit at the Darlington New Nuclear Project in April 2025, valid to March 2035 and subject to 3 regulatory hold points; the first hold point (reactor building foundation) was cleared in March 2026, and OPG filed for its 20-year operating licence the same month, the first grid-scale SMR construction approval in Canada (CNSC / Canada.ca).

Insurance & Liability

Reactor-scale liability is handled differently from almost every other industry a business plan template covers. In the US, the Price-Anderson Act caps and pools operator liability for a nuclear incident across the industry, requiring each licensed plant to carry primary private insurance (currently around $500 million per site) backed by an industry-wide secondary retrospective pool. The UK operates a broadly similar model under its own nuclear third-party liability legislation, with operator liability caps set in line with international conventions. For a services business, the relevant insurance conversation is much narrower and much more familiar to a commercial underwriter: general liability, professional indemnity, and (for anyone handling sealed radioactive sources) a radiation-specific rider, none of which require the plan to engage with Price-Anderson-style pooling at all.

Common Planning Mistakes in This Section

• Sizing the funding ask to the reactor's headline cost while ignoring that long-lead items must be ordered years before first concrete, on their own separate payment schedule that a lender or DOE reviewer will ask about directly
• Writing a revenue section that assumes merchant wholesale power prices instead of showing a contracted PPA or a RAB-style regulated return, which is how essentially every recent project has actually been financed
• Treating NRC/ONR/CNSC review as a flat permit fee rather than an hourly-billed, multi-year line item with its own dedicated budget line, distinct from general legal and compliance costs
• Omitting a pre-funded decommissioning provision, which both US and UK regulators require to be funded over the life of the asset and which investors will specifically check for in a financial model
• Ignoring grid interconnection queue position, which routinely adds 12-36 months and changes workforce and financing-cost assumptions materially, especially in congested regional grids

Nuclear Business Plan Glossary

• SMR (Small Modular Reactor): a factory-built reactor design generating up to roughly 300MW, designed for shorter construction schedules than traditional large reactors
• PPA (Power Purchase Agreement): a long-term contract (typically 17-20 years) locking in the price a buyer pays for a plant's output
• RAB (Regulated Asset Base): the UK financing model used for Sizewell C, paying a regulator-set return during construction via a per-MWh consumer levy
• GDA (Generic Design Assessment): the UK's 3-step process for assessing a reactor design before it can be built at a specific site
• COL (Combined License): the US NRC's traditional construction-and-operating licence under 10 CFR Part 52
• Capacity factor: the share of a plant's theoretical maximum output it actually generates over a year; nuclear plants typically run 90%+
• Long-lead items: components with multi-year manufacturing lead times (reactor pressure vessels, steam generators) that must be ordered years before construction starts
• Merchant plant: a plant selling power on the open market or via bilateral PPA, rather than under a regulated utility tariff
• N-stamp: the ASME certification mark authorising a manufacturer to fabricate nuclear components (pressure vessels, piping) to code; a common target certification for a supply-chain business scaling up from general industrial fabrication
• Decommissioning fund: a ring-fenced financial provision, funded across the operating life of a plant, required by regulators in both the US and UK to cover eventual safe shutdown and site remediation
• COD (Commercial Operation Date): the date a plant begins generating and selling power under commercial terms, marking the point revenue starts and the construction-financing clock stops
• Regulatory hold point: a mandatory checkpoint (used by the CNSC in Canada, for example) at which a regulator must formally verify a construction commitment before work can proceed to the next stage

Sample Business Plan Preview

Here's an illustrative extract showing the level of detail a reactor-scale plan needs, drawn from the same planning assumptions used throughout this page. This is a composite example, not a real client project.

What's in the Template

Every Avvale business plan template includes these sections, pre-structured for your industry:

• Executive Summary — Your business at a glance, written to hook investors or lenders in 60 seconds
• Company Overview — Legal structure, ownership, location, and founding story
• Industry Analysis — Market size, growth trends, and regulatory requirements specific to this sector
• Customer/Offtaker Analysis — Who buys the output or service, and what triggers the purchase decision
• Competitor Analysis — Positioning against direct competitors and the wider vendor/developer ecosystem you compete inside
• Marketing/Business Development Plan — Channels, messaging, and how deals or contracts actually get won
• Operations Plan — Day-to-day workflows, staffing structure, and key milestones
• Management Team — Founder bios, advisory board, and key hires planned

The optional Financial Forecast add-on (included in our $300/£250 and $1,000/£800 packages) provides a 5-year Excel model with income statement, cash flow, balance sheet, break-even analysis, and startup capital requirements — with licensing spend and, where relevant, decommissioning provisions modelled as their own line items rather than folded into a generic overhead figure.

For readers preparing a reactor-scale plan, it's also worth knowing what tools the engineering and licensing teams on the other side of the table are actually using, since referencing them accurately signals real domain fluency. Safety-case submissions to the NRC and ONR commonly rely on thermal-hydraulic and safety-analysis codes such as GOTHIC and RELAP5; large EPC construction programmes typically run document control and scheduling through platforms like Aconex and Primavera P6. None of these appear in a $5 template, but a bespoke plan aimed at a DOE, ONR, or institutional-investor audience benefits from acknowledging the engineering toolchain a project will actually run on.

Frequently Asked Questions

Related Guides

• Free Business Plan Template: general-purpose template if your business doesn't fit the nuclear-sector assumptions on this page
• Industry-Specific Business Plan Template: the $5 template referenced throughout this guide
• Market Research & Content for Business Plan: for teams that want the research done professionally before a DOE, ONR, or lender submission
• Bespoke Business Plan: full plan and 5-year forecast built by Avvale's team
• Avvale Case Studies: more composite and real client outcomes across sectors, including energy
• Business Plan Writer: for readers who want a human writer rather than a template, for a related industrial or energy-sector plan

Useful Links & Resources

These links were preserved from the live page so important references and partner links are not lost during the page refresh.

• Facebook
• Instagram
• LinkedIn
• YouTube