We keep evaluating hydrogen purely by $/kg — but forklift TCO might tell a different story...
I've been following Plug's numbers closely (and the hydrogen sector more broadly), and there's something I think gets underdiscussed: we always compare green hydrogen cost ($5-8/kg) to grey hydrogen or to battery electricity, and conclude "it's expensive." But that only looks at fuel cost — not the TCO (total cost of ownership) of running a forklift fleet.
Companies running hydrogen forklifts (Amazon, Walmart, BMW, etc.) gain in places that rarely make it into the comparison:
- Refueling in minutes vs. hours to recharge a battery — less machine downtime
- No dedicated charging rooms (warehouse space saved, no special ventilation needed for battery off-gassing)
- Consistent performance throughout the shift (batteries lose power as they discharge; fuel cells don't)
- Less labor dedicated to swapping/charging batteries
- Possibly fewer total fleet units needed, since there's no "dead" recharge time to plan around
Put this together, and it makes sense that anchor customers would pay a price for hydrogen that covers Plug's actual production cost (that $5-8/kg range), even knowing it's more expensive than the "simple" alternative — because they're not comparing hydrogen vs. electricity, they're comparing total hydrogen operation vs. total battery operation.
Question for the community: does anyone have real comparative TCO data (case study, whitepaper, customer figures) that confirms or debunks this? Or is it just a good sales pitch that hasn't translated into published numbers yet?
And another open question tied to this: if the hydrogen sold to anchor customers (Amazon, Walmart) might be near cost, or even covered once you factor in total operational value (TCO), then where is the cash burn actually concentrated? Plug has never published margin broken out by segment — it could be sitting in electrolyzers.
I've seen in Plug's own filings that most electrolyzer contracts are direct sales (the customer takes ownership of the equipment), with leasing described as "a limited number of arrangements" — not the dominant pattern. But there's a related mechanism that could have a similar cash effect: Plug's cash conversion cycle runs around 233 days — meaning even when selling (not leasing), they may be financing production and installation on large projects (FEED/pre-FID) for many months before receiving final payment. Does anyone know if the big contracts (Hy2gen, Uzbekistan, etc.) have payment staged throughout construction, or only at the end? That would change the read on where the burn is concentrated quite a bit.
Not financial advice, just trying to figure out if we're evaluating this by the right metric.