From pilots to scale: Why "successful" energy programs don't expand
The energy sector doesn't have an innovation problem. It has a deployment problem.
Every year, utilities, manufacturers, technology developers, national laboratories, and public agencies launch pilots to test promising energy technologies, and many of those pilots succeed. Equipment performs. Energy use falls. Emissions decline. Customers learn something valuable. Someone writes it up.
And yet, in most cases, nothing further happens.
The technology doesn't become a standard utility offering. The customer segment doesn't transform. The measure doesn't move into routine procurement. The market doesn't scale. Proving something works turns out to be the easier half of the task. The harder question — what would make this technology routine? — rarely receives equal rigor.
A pilot can prove performance. It rarely proves adoption.
There's already a name for the first kind of proof. Technology Readiness Level (TRL), the nine-point scale the Department of Energy and NASA use to track a technology from basic principle to full-scale demonstration, measures whether something works. It says nothing about whether anyone can buy it, finance it, permit it, or keep it running. DOE's Adoption Readiness Level (ARL) fills that gap, scoring a technology's value proposition, market acceptance, resource maturity, and social license to operate. Transformative Pathways for U.S. Industry, a national industrial decarbonization roadmap developed by DOE with Energetics and national laboratory partners, applies this same distinction directly. A technology can be technically mature and still stall because no one has resolved who bears the adoption risk.
A pilot advances a technology along the TRL scale. It rarely advances the ARL score. That gap, between proof of performance and proof of adoption, is where most promising technologies quietly stop.
Industrial heat pumps
Thermal systems — process heat, combined heat and power — account for roughly half of all energy-related manufacturing emissions, more than 90% of it from burning fossil fuels, according to the same DOE analysis. Industrial heat pumps are among the leading proposed solutions: high-efficiency systems capable of delivering steam, hot water, and hot air.
That's the technical case. The adoption case is harder.
AtmosZero's work with New Belgium Brewing illustrates why. The company installed a 650-kilowatt pilot unit at the brewery capable of providing roughly 30–40% of the brewery's steam needs at full capacity — evidence that industrial steam electrification is technically viable. It is not evidence that every food processor, brewery, chemical plant, or paper mill will adopt it quickly. Scale requires a business case that withstands cheap natural gas, downtime risk, and competition for capital inside the plant. Plant managers don't purchase coefficient of performance; they purchase dependable production. In ARL terms, this is a value-proposition problem. The technology performs, but it hasn't yet outperformed the status quo on price and risk for most buyers.
Sustainable aviation fuel
Sustainable aviation fuel (SAF) presents the same pattern. SAF has demonstrated technical legitimacy — multiple pathways are certified, and aviation has no viable electrification substitute — yet production remains marginal relative to need. IATA estimated that 2024 SAF production reached one million tonnes, approximately 0.3% of global jet fuel production, below earlier projections. ICCT's 2025 analysis identified the principal barriers to SAF scale-up: high capital costs, first-of-a-kind technology risk, and offtake and price uncertainty.
In other words, the pilot question — can low-carbon fuel be produced and used? — differs from the scale question. The scale question is whether developers can finance multi-billion-dollar facilities, secure bankable offtake, and compete in a commodity fuel market not designed for them. SAF's constraint isn't creativity. It's the absence of a financed, de-risked supply chain willing to commit capital at scale. This is a resource-maturity problem. The capital, feedstock, and offtake infrastructure the fuel demands don't yet exist at the scale it requires.
Electric heavy-duty trucks
Demonstrations can show that trucks move freight reliably on defined routes, that drivers adapt to new equipment, and that depot charging functions under controlled duty cycles. The scale barrier isn't the truck itself. The Electrification Coalition's 2026 freight charging report identifies barriers, including high upfront vehicle costs, grid interconnection delays, misaligned electricity rate structures, and fragmented local permitting processes. The charging site, the utility upgrade, the tariff, the freight schedule, and the financing structure are each components of the technology in question.
The "measure," in this instance, isn't the truck. It's a coordinated freight-electrification system; and permitting, interconnection, and rate design were built around vehicles that don't plug in. This is as much a license-to-operate problem as a technology problem.
LEDs: the counterexample that proves the rule
LEDs didn't scale because they were efficient. They scaled because the ecosystem aligned around them. Product availability improved, costs fell, contractors learned the technology, and standards made selection easier. DOE estimated LED adoption saved 1.3 quadrillion Btu in 2018, worth $14.7 billion to consumers, with installations roughly doubling from 2016 to 2018 to 2.325 billion units, or 30% of general illumination lighting.
That's what scale looks like: not a successful demonstration, but a market that already knows what to do next. LEDs didn't wait for a stronger TRL score. They waited for the ARL score to catch up, across every dimension at once.
The technologies that prevail are not always the most technically advanced. They are the ones easiest to deploy, finance, permit, install, operate, maintain, explain, and repeat.
Redefining a successful pilot
This should change how the sector defines success. A pilot should ask not only whether a technology works, but who owns the risk, who pays the premium, who installs and maintains it, what infrastructure must already exist, what rate design makes it viable, what customer decision point triggers adoption, and what would have to be true for the next 100 projects to be easier than the first five. These are ARL questions, and the sector generally asks them after the ribbon-cutting rather than before it. A technology shouldn't graduate from pilot to program until the ARL questions can be answered as confidently as the TRL ones.
There’s a further barrier that rarely appears in a pilot report. The same analysis names inefficient information flows as one of five common barriers across the industrial ecosystem. Companies withhold performance and cost data over competitive and intellectual property concerns, so each new adopter must relearn the same lessons independently. Closing that gap doesn't require better technology. It requires a party with genuine technical depth and no stake in the sale.
That's the model behind Energetics' INTERConnect — an evaluator with no financial stake in any technology it reviews, backed by more than a decade of experience supporting DOE's Office of Technology Commercialization on portfolio and grant management, technology screening, and commercialization strategy. Paired with CLEAResult's established utility relationships and program infrastructure, INTERConnect screens technologies before they reach a pilot; not just for performance, but for financing fit under existing rate structures, maintenance and staffing realities, and supply chain depth beyond a handful of installations.
The better question
The sector's next frontier isn't inventing more technology. It's building the discipline to advance proven technology through all four ARL dimensions, project after project.
For decades, the energy sector has asked, "Did the pilot work?" That's the wrong question. The better one is: What would it take for this technology to become boring?
Boring means the installer knows what to do. The customer understands why it matters. The utility knows how to support it. The regulator understands its value. The financier can price its risk. The supply chain can deliver it. And the next project is easier than the last.
That's scale.
Until the sector measures success on both scales — TRL and ARL — it will continue to produce successful pilots that lead nowhere.