How to Choose a PR agency for Nuclear energy companies

By Aidan Sowa, CEO and Founder, Frontier Industries | October 2026

Nuclear power runs on the strongest evidence in energy: decades of operating data, a regulator that publishes everything, and reactors that deliver around the clock. It also carries the deepest suspicion in energy, suspicion installed by one badly explained week in 1979 and maintained by every evasive sentence since. This guide reads the published record on costs, polling and restarts, and turns it into criteria for choosing an agency. The short version: in nuclear, the engineering case is closed, and the communications case is the only one still being tried.

94 reactors: the US commercial fleet, the largest in the world (EIA 2025).

19 percent: nuclear's share of US electricity generation in 2024 (EIA 2025).

55 percent: Americans who told Gallup they favour nuclear energy, the highest share since 2012 (Gallup 2023).

I Discrimen

Operating safely is not the same as being believed to operate safely

The record is auditable. The belief is not automatic, and the gap between them is your communications department.

The US fleet operated 94 reactors in 2024 with a combined capacity near 97 gigawatts, supplying 19 percent of the country's electricity (EIA 2025). By the evidentiary standard applied to any other technology, the argument is over. Yet every nuclear project still begins its public life as a defendant. The reason is historical: the industry taught the public, at Three Mile Island and after, that its statements arrive late and hedged while its instruments tell a different story.

An agency that proposes to fix this with enthusiasm will fail and invoice you. The deficit is not affection; it is credibility of process. The right agency works like the regulator does: publish routinely, quantify precisely, admit deviations the day they are logged, and let the accumulation do the persuading. Nuclear communication that works is indistinguishable from good operations made visible.

Ask candidates what they would publish about your plant on a completely ordinary Tuesday. If the answer is nothing, they plan to meet the public only on the bad days, which is the exact habit that built the suspicion.

Trust in this industry is not a feeling to be created. It is a filing habit to be observed, over years, by sceptics.

II Machina

The regulator publishes your life daily, so your agency's job is translation

Nuclear is the one energy sector where the primary record is already public, complete and free.

The Nuclear Regulatory Commission posts event notifications, inspection findings and licensing documents as a matter of routine. Your performance indicators are public. Your corrective actions are public. Journalists and opponents read this record directly, which means the traditional agency function of controlling information flow is not merely useless here; attempting it is itself a story. What remains is translation: converting regulatory language into public meaning before opponents convert it into public alarm.

Consider what an unplanned shutdown looks like in the two dialects. In the docket it is a manual reactor trip with systems performing as designed. In an opponent's leaflet it is an emergency at the nuclear plant. Both cite the same event. The company that has spent years explaining its systems to its host community hears the first version repeated back. The company that communicated only through press releases hears the second.

The industry PR resources gathered for this programme show the pattern across regulated industries: whoever explains the record owns the record. Build your agency brief around explanation rights, docket fluency and community education, not around coverage volume.

You cannot spin a public document. You can only be the most reliable reader of it in the room.

III Exemplum

Three Mile Island was a communications accident with a reactor attached

The canonical case is forty six years old and still the syllabus.

In March 1979, a partial meltdown at Three Mile Island Unit 2 injured no one in the public and released radiation doses officials later measured as small. The Kemeny Commission, appointed to investigate, reserved some of its sharpest findings not for the reactor but for the information: statements from the utility and the regulator conflicted, quantities were given in units nobody could interpret, and the public reasonably concluded it was being managed rather than informed. The industry's technical inheritance from TMI was better control rooms. Its reputational inheritance was a presumption of concealment that two generations of communicators have been paying interest on.

Now the counterpoint, from the same site. In September 2024, Constellation announced a twenty year power purchase agreement with Microsoft and the restart of Three Mile Island Unit 1, undamaged in 1979 and shut in 2019 for economics, under a new name, the Crane Clean Energy Center. The announcement led with the buyer, the term, the jobs and the grid need, in numbers, in one release. Whatever one thinks of the project, its communications assumed an informed reader.

Ask agencies to brief both episodes. You are listening for whether they understand that the second announcement was possible only because it behaved as if the first one had happened on its watch too.

The accident ended in weeks. The explanation deficit has run for decades, and it compounds.

IV Damnum

In nuclear, distrust is priced into the cost of capital

The most expensive sentence in this industry is the one that makes a lender add a point.

Plant Vogtle's two new reactors in Georgia entered service in 2023 and 2024, seven years late, at about 35 billion dollars against an original estimate near 14 billion (E and E News reporting of Georgia Power figures 2024). South Carolina's V.C. Summer project was abandoned in 2017 with roughly nine billion dollars spent and nothing to show. Neither project failed on physics. They failed on the compound of construction risk, regulatory friction and political durability, the three things a financing actually prices. Every nuclear proposal since has carried those two postcodes in its interest rate.

Communications sits inside that price whether it wants to or not. Political durability, the belief of host states and federal policymakers that a project will not become a scandal, is an input to every loan committee and every legislature that votes support. A company with a disciplined public record borrows against a different future than a company whose name arrives with a warning label.

So evaluate the agency as you would evaluate insurance. Its fee is trivially small against a quarter point on a ten billion dollar borrowing. Its absence is correspondingly expensive, and the bill arrives denominated in basis points, not headlines.

Cheap communication is the most expensive line in a nuclear budget. It is just invoiced by the bond market.

V Genus

The controlling variable is political durability, not public opinion

Polls say the argument is being won. Projects die anyway, because durability is a different asset from approval.

Gallup measured 55 percent of Americans favouring nuclear energy in 2023, the highest reading since 2012 (Gallup 2023). A national majority, however, has never permitted a reactor. Permits come from statehouses, utility commissions and host communities, institutions that ask a narrower question: will supporting this plant still look wise in ten years, after the next incident anywhere in the world and the next election at home? Fukushima moved German policy in a week and Japanese restarts for a decade, regardless of what German or Japanese voters told pollsters about energy in the abstract.

The agency skill that matters is therefore coalition maintenance across cycles: host community relationships that survive management changes, statehouse education that survives elections, and a waste and safety narrative honest enough to survive the next foreign incident's news cycle. That is slow, unglamorous, document heavy work, and it is mostly invisible in agency portfolios, which prefer launches.

Interrogate the quiet years in any candidate's history. Ask what they did for a client between announcements, across one election and one piece of bad industry news. The answer is the product you are buying.

Opinion is weather in this sector. Durability is climate, and projects are built for climate.

VI Regio

An operating utility, an SMR startup and a decommissioning firm sell three different futures

The word nuclear covers companies whose publics barely overlap.

Each row has a distinct temptation. Operators are tempted to say nothing in quiet years. Developers are tempted to announce commercial dates as if licensing were a marketing channel; the regulator's schedule is not. Fuel companies are tempted by obscurity, until a geopolitical shock makes their obscurity look like unpreparedness. Decommissioning firms hold the industry's most underrated asset: a finished, visible promise, a site returned as agreed.

An agency must know which temptation it is hired to resist. The oil and gas agency guide in this series describes the cousin problem, licence maintenance under scrutiny, and the overlap is real: both industries are believed in proportion to their filing habits.

One element symbolises the whole sector. It should not make the four business models sound alike.

VII Occasio

The opening arrived as a purchase order from the data centre industry

For thirty years nuclear advocates waited for public opinion to save the industry. What saved it was load growth and a signature.

The hyperscale data centre buildout has made technology companies into nuclear's most credible advocates, because they arrive as buyers with twenty year contracts, not as campaigners. The Three Mile Island Unit 1 restart agreement, and the return to service of Michigan's Palisades plant after shutdown, the first such restart attempted in the United States, share a structure: a creditworthy buyer, a grid that needs firm clean power, and a host community that mostly wants the plant back. Palisades' neighbours, like most plant neighbours in survey research, were never the opposition of national imagination.

This changes the agency brief fundamentally. The audience that matters most now reads procurement documents: technology companies selecting power partners, utilities weighing life extensions, states competing for restarts and new builds. Communication aimed at them is evidentiary and commercial, heavy on delivery record and light on advocacy.

The public relations services approach for this sector has been rebuilt around that buyer. Ask candidates how they would write for a chief energy buyer at a hyperscaler. If they reach for the climate speech, they are fighting the last war, eloquently.

Nuclear did not need better arguments. It needed better customers, and they have arrived with contracts.

VIII Structura

Host communities are the industry's proof, and they are routinely ignored

The equaliser available to every nuclear company is standing in its own fence line towns.

Survey research among plant neighbours, conducted over decades, finds support for the local plant far above national averages, because neighbours have what national audiences lack: daily evidence. They know the plant's employers, its tax contribution, its outage etiquette and its sirens tests. That gap between experienced and imagined nuclear power is the industry's central communications fact, and its most underused asset. The voices that close the gap are not executives. They are the plant's neighbours, workers and local officials, speaking in their own register about their own town.

Agencies that understand structure build that bench deliberately: community advisory panels with real documents, local spokespeople briefed like experts because they are, and plant access programmes that let sceptics see the ordinary competence of the place. It is slow and it does not scale nationally, which is precisely why it works. Belief formed at fence line distance does not blow away in a news cycle.

Ask to meet a candidate's community work, literally: the towns, the panels, the years. A firm that answers with audience numbers has brought a media plan to an evidence fight.

The plant's best argument lives within ten miles of the plant and has a first name.

IX Consilium

Score for docket fluency, candour under boredom and community tenure

The order of hiring criteria that this record supports.

Docket fluency first: the agency must read NRC documents the way journalists and opponents do, and translate them before breakfast. Candour under boredom second: the willingness to publish routine data in routine weeks, which is where credibility is actually manufactured. Community tenure third: named towns, named years, relationships that predate your hiring decision.

Trial finalists on the real artefact: one routine NRC filing about your plant, translated for your host town, on a deadline. Then check the people against the firm's account of its team and reference call two host community leaders the agency has worked with. In nuclear, the references are findable and the industry is small. Make the calls.

Hire for the ordinary Tuesday and the worst Wednesday with the same test: does this firm publish early, precisely and without being asked twice?

X Quaesita

Questions nuclear company buyers actually ask

How do we talk about waste without either dismissing it or surrendering to it?

State the facts in order: volumes are small, the material is solid, it is stored on site under licence, and the federal repository programme is a political failure with a date on it, not a technical mystery. Then state your position and your funding. Audiences forgive an unsolved national problem described honestly. They do not forgive being told the problem does not exist.

Should the company comment when there is an incident at another operator's plant?

Almost never on the event, always on your own house if asked. The discipline is to have your house in order publicly before the question arrives: performance indicators current, community panel briefed, statements about your plant factual and short. Commentary on a competitor's accident is read as opportunism even when it is sympathy.

How much of the budget belongs to host community work?

More than feels rational from headquarters, because the host community is the only audience whose belief is based on evidence and the only one whose opposition can actually stop you. National advocacy is cheaper per impression and worth less per unit of risk reduced. Fund the fence line first.

Can an agency help with an SMR licensing narrative?

It can keep the narrative inside the licensing reality: milestones described as the regulator defines them, no commercial dates detached from dockets, investor materials that match public filings word for word on the technical claims. The fastest way to destroy an SMR company is to let its marketing outrun its docket, and agencies are usually the accelerant.

What should we publish routinely?

Performance indicators, outage explanations in plain language, environmental monitoring summaries, drill schedules and community panel minutes. The set matters less than the cadence. A sceptic who finds the same document updated on schedule, in good months and bad, recalibrates what kind of company you are.

How do we prepare for the next foreign incident's news cycle?

Before it happens: pre briefed host community leadership, a standing explainer on your reactor type and its differences, and executives who have rehearsed saying what is known, what is not, and what does not change about your plant. The cycle will come. The only choice is whether your first day sounds like your ordinary Tuesdays.

What metrics matter?

Host community survey trends, correction rates in coverage of your filings, buyer and policymaker meetings earned, and the time between an event entering the docket and your public explanation of it. National sentiment scores are decoration. Measure the audiences that hold pens.

XI Methodus

How this guide was built, and where it is thin

The fleet and generation figures come from the US Energy Information Administration's 2025 account of the nuclear fleet, the favourability figure from Gallup's March 2023 environment survey, and the Vogtle cost account from contemporaneous reporting of Georgia Power's figures, each cited in place and listed below. The Three Mile Island account reflects the Kemeny Commission's published findings and Constellation's September 2024 restart announcement.

The caveats deserve their space. Gallup's figure is a 2023 reading of a question asked since 1994, and nuclear opinion is historically event sensitive. The Vogtle figure is a reported total that varies by what is counted, financing included or not. Neighbour support research is largely industry commissioned. And no public data ranks agencies by licensing or community outcomes, so the hiring order here is disciplined inference from the failure record, labelled as inference.

XII Incipe

Where Frontier Industries fits in this

Frontier Industries treats nuclear communication as an evidentiary practice: docket translation, community tenure and routine publication, built for a sector whose record is strong and whose habits of explanation are its only real deficit.

If you are choosing an agency for a nuclear company, begin with an ordinary document from your own docket and ask the agency to explain it to your host town: request an industry campaign consultation or call 401-219-4207.

Sources

1. Gallup, Americans' Support for Nuclear Energy Highest in a Decade 2023, https://news.gallup.com/poll/474650/americans-support-nuclear-energy-highest-decade.aspx 2. US Energy Information Administration, the United States operates the world's largest nuclear fleet 2025, https://www.eia.gov/Todayinenergy/detail.php?id=65104 3. US Energy Information Administration, US nuclear industry explained, https://www.eia.gov/energyexplained/nuclear/us-nuclear-industry.php 4. Institute of Nuclear Power Operations, industry performance organisation, https://www.inpo.info/ 5. World Nuclear Association, nuclear power in the USA country profile, https://world-nuclear.org/information-library/country-profiles/countries-t-z/usa-nuclear-power 6. Nuclear Energy Institute, industry association, https://www.nei.org/ 7. International Energy Agency, Nuclear Power and Secure Energy Transitions, https://www.iea.org/reports/nuclear-power-and-secure-energy-transitions 8. International Atomic Energy Agency, Power Reactor Information System, https://pris.iaea.org/ 9. US Department of Energy, Office of Nuclear Energy, https://www.energy.gov/ne/office-nuclear-energy 10. American Nuclear Society, nuclear news and policy coverage, https://www.ans.org/news/