Engineering Article
How Many Stars Are in the Solar System? And Other Questions Energy Storage Buyers Get Wrong
-
The question that gives away everything
-
Argument 1: Replacement cost is a probability, not a price tag
-
Argument 2: The warranty covers the part. It doesn't cover the money.
-
Argument 3: The revenue line nobody budgets for
-
Argument 4: The chicken coop that taught me more than any RFP
-
The pushback I expect
-
Where this leaves you
The question that gives away everything
Ask ten people "how many stars are in the solar system" and you'll get one of two answers: a confident "billions" (that's the galaxy, not the solar system) or a sheepish "a lot?" The literal answer is one. Just the Sun. Everything else you see at night is a planet, moon, asteroid, or comet bouncing sunlight back at us.
Why does this matter for B2B energy storage? Because I'm seeing the exact same pattern in procurement meetings, and it's quietly costing companies six figures.
Here's my position after four years of managing a $2.3M annual storage and charging budget: most energy buyers ask the wrong first question. They want to know the cost to replace a Tesla battery, or how good the Tesla battery warranty really is. Both matter. Neither is the number that decides whether a project pencils out.
I'm a procurement manager at a 180-person energy integration firm. Over the past four years I've negotiated with 12 vendors, audited every invoice in our cost tracking system, and built a TCO spreadsheet that I now refuse to bid without. I've made the expensive mistakes so your team doesn't have to. This is what I'd tell a peer over coffee before their first storage RFP goes out.
Argument 1: Replacement cost is a probability, not a price tag
The most-Googled question in this space, by a landslide, is the cost to replace a Tesla battery. And I get it—the number feels enormous. Depending on system and installer, a Powerwall replacement runs roughly $7,000–$15,000 all-in (based on installer quotes circulated in early 2025; verify current pricing with your integrator). Commercial Megapack-class swaps can hit six figures depending on enclosure and site logistics.
But here's what the number doesn't tell you: it's not a bill you're likely to pay during the warranty window. In our fleet of 24 commercial-grade units deployed between 2020 and 2023, we've logged zero out-of-warranty replacements through Q4 2024. Zero. The batteries are behaving exactly as the datasheets said they would.
So when a CFO sees "$12,000 per replacement" and budgets for it as if it's a scheduled expense, they're over-reserving—and potentially killing projects that would've cleared hurdle rate. The honest way to model it is as a low-probability, post-warranty event, amortized over the back half of the system's life, adjusted for inflation. That's a very different number on the spreadsheet. Usually 15–25% lower than the sticker-price approach, in my experience.
Was I always this clear-headed about it? No. In 2022 I built a renewal reserve that was, in hindsight, comically conservative. It made one project look unviable on paper. We passed. A competitor took the site and, two years later, hasn't had a single swap.
Argument 2: The warranty covers the part. It doesn't cover the money.
Everyone asks about the Tesla battery warranty. Good. The published terms—generally 10 years or a defined cycle count for residential and commercial systems, with a stated capacity retention floor—are real and enforceable. I've read them. Our legal team has read them twice.
What the warranty doesn't cover is the part where your site stops producing. Downtime has a number attached to it, and that number never appears on the warranty PDF.
In Q2 2023, one of our units threw a fault that was, without argument, a warrantable issue. Tesla shipped the replacement module at no charge. Total direct cost to us for the incident? Around $4,200—crane time, two electricians, a site visit, and roughly 40 hours of lost load-shifting revenue while the unit was offline.
That's where the process gap bit us. We had no formal checklist for "warranty repair logistics" before that. The third time something like this happened, I finally wrote one: who signs off on crane rental, who tracks lost revenue, who negotiates the service-credit schedule. Should've done it after the first time. A lesson learned the hard way.
An informed buyer asks "what's covered?" A smart buyer asks "what's covered, and what does it cost me while it's being covered?"
Argument 3: The revenue line nobody budgets for
This is the part I wish someone had walked me through on day one. Energy storage for grid support rental—leasing your battery's capacity back to the grid operator during peak events—is one of the most under-modeled line items in commercial storage.
Not every site qualifies. Not every market pays well. And the program mechanics vary wildly by region. So treat this as a directional observation, not a guarantee. But in our portfolio, grid-support rental revenue recovered roughly 18% of initial capex over the first 24 months of participation—on assets we already owned and were already cycling.
Why does this connect to the replacement-cost conversation? Because it changes the denominator. If you're weighing a $12,000 potential swap against the purchase price alone, it looks like a cliff. Weigh it against net cost after lease revenue, and the cliff flattens into a slope. The risk is real—grid programs can change rules, and you may not always get the dispatch hours you modeled. But the expected value swung positive for us, and the downside—modest revenue shortfall—was survivable. That's the trade most teams never even put on the table.
Argument 4: The chicken coop that taught me more than any RFP
Hang with me here, because this sounds off-topic. A friend of mine runs a small farm and wanted power out at his chicken coop—200 feet from the house, no trenching budget, no grid tie. He bought a solar generator for the chicken coop setup: about $3,200 all-in for panels, battery, and a timer for the coop door. It saves him around $600 a year in what he'd otherwise spend running conduit and paying a minimum meter charge.
Payback: roughly five years. Same logic as a Megapack.
The lesson isn't that a chicken coop and a grid-scale battery are the same project. They're not. The lesson is that cost logic doesn't change with scale—only the numbers get bigger. My friend didn't ask "what does replacement cost?" He asked "what's the total spend over ten years, compared to the alternative?" Same framework. Same discipline. He happened to apply it to a coop door; I applied it to a $900k storage deployment. The spreadsheet looks identical in structure.
Honestly, that conversation did more for my procurement thinking than half the vendor decks in my inbox.
The pushback I expect
I can hear the objections, so let me address the three I get most.
"You're comparing a chicken coop to a commercial battery. That's not serious." Fair, on the surface. But the entire point of cost control as a discipline is that the framework is scale-invariant. If it doesn't work at $3,200, it's not working at $3.2M either—it's just hiding behind more decimal places.
"Not every site can do grid support rental." Correct, and I said so. This isn't a universal lever. It's a check-the-box item on a workbook that most teams never open. If you're in a market where it works, great. If you're not, at least you'll know why the number isn't there.
"Replacement cost still matters—you said so yourself." Yes, and I meant it. My argument isn't that replacement cost is irrelevant. It's that it's the second question, dressed up as the first. The first question is: what does my model of the total cost look like, and what's missing from it?
Where this leaves you
So back to the solar system. When someone asks how many stars are in it, the useful answer isn't "one" or "billions"—it's "depends what you're counting, and why you're counting it." The same is true of your storage budget. The cost to replace a Tesla battery is one answer. The Tesla battery warranty is another. Grid support rental is a third. The real number you need is the one that shows up when you stop counting line items and start counting outcomes.
I'd rather spend an extra hour in a planning meeting explaining this framework than three months explaining to a CFO why the project we greenlit doesn't pay back on schedule. An informed buyer makes faster decisions and fewer of them regretted. That's the whole job.
Pricing and program figures are for directional reference only, drawn from early-2025 installer quotes and our internal project data. Verify current rates and eligibility with your integrator and local grid operator before you build a business case.
Ask a technical follow-up