⚠ Service Margins Have Been UnreliableModerate threat

The service annuity is Bloom's best moat mechanism, but its economics have historically been unreliable, and that undercuts the quality of the whole. The core of a service agreement is Bloom's obligation to keep the systems performing, which means replacing the fuel-cell stacks as they degrade — and if the stacks wear out faster, or cost more to replace, than the contract assumed, the service business loses money on that installation. Bloom has, at various times in its history, struggled with exactly this: stack degradation and replacement costs that ran ahead of expectations, forcing charges and turning the supposedly high-margin annuity into a drag. A service moat is only as good as the margins it earns, and Bloom's have not been dependable.

The risk is structural to the model: Bloom commits to long-term performance at contracted prices, but bears the cost of maintaining hardware whose degradation is not perfectly predictable, so it carries the risk that reality proves worse than the contract assumed — especially as it deploys a new generation of systems at unprecedented scale and speed, where any unexpected reliability problem would show up in service costs across a large fleet. Bloom's technology has improved and its newer systems are more durable, and a well-run service business on a large base would indeed be a genuine, high-quality moat. But an investor should treat the service annuity's profitability as unproven rather than assured: the mechanism is real and the base is growing, but the history of unreliable service margins means the annuity's quality depends on execution Bloom has not consistently delivered — and a reliability problem across a rapidly-scaled fleet would strike at the very part of the business that is supposed to be its sturdiest — and the fleet is being scaled at a +215% product-revenue clip¹.