If you're comparing energy storage systems by price per kilowatt-hour, you're probably making the same mistake I made when we started buying batteries in 2021.
I chased the lowest unit price. The system that looked cheap on paper cost us more in O&M, warranty friction, and lost revenue within three years than the "expensive" option would have cost over its full life. That's not a hunch—I went back and built a TCO model for our next round of BESS procurement, and the numbers were brutally clear.
So let me say this without hedging: comparing BESS manufacturers on $/kWh alone is one of the most expensive mistakes a storage buyer can make. Total cost of ownership (TCO)—and its cousin, levelized cost of storage (LCOS)—is the only comparison that tells you what a system will actually cost over 10 or 15 years. If a supplier can't show you LCOS assumptions, in my opinion, they're not giving you a price. They're giving you the opening bid.
The Quote Is Only the Tip of the Iceberg
When I first started evaluating BESS manufacturers, I built a spreadsheet with the usual columns: capacity, round-trip efficiency, price per kWh. I sorted by unit price and felt pretty good about the top row.
Then our operations team caught something I'd missed. The lowest-quoted system had a warranty that covered only the battery modules—not the thermal management system, not the BMS boards, not the interconnection relays. The other bidder, the one that looked about 12% more expensive, included all of that in the package. Once I added replacement logistics, downtime, and the engineering hours we'd have spent stitching together third-party warranties, the "cheap" option was the expensive one.
That experience changed how I compare bids. My cost model now includes:
- Base hardware price — the $/kWh everyone quotes, which is usually the DC block only.
- Balance-of-system costs — inverters, transformers, switchgear, controls, containers.
- Installation and commissioning — including months of utility interconnection delays if your IEEE 1547 paperwork isn't in order.
- Degradation and throughput guarantees — what the manufacturer commits to in writing, and what happens if they miss it. A 0.2% annual degradation difference doesn't sound dramatic, but over a 10-year PPA it changes project economics.
- O&M over asset life — thermal management power, filter cleaning, firmware updates, remote monitoring fees, spare parts.
- Decommissioning or repowering — nobody prices this in, but it's a real line item at end of life.
Take a rough example: two offers for a 5 MW / 20 MWh system. One integrator quotes $190/kWh and guarantees 80% capacity after 10 years. Another quotes $215/kWh and guarantees 86% retained capacity with a stronger throughput warranty. On the initial quote, the first one wins by around $500,000. But a 6% capacity difference at year 10—combined with round-trip efficiency and O&M terms—changes the revenue envelope over a decade by more than that upfront gap. I'm not saying the pricier bid is always the better buy. I'm saying you can't know which one is better until you run the TCO.
I'm not saying you need to model all six lines before your first call with a supplier. I am saying that if you only ask "what's your $/kWh?" you're handing the negotiation to whoever can hide the most cost in the fine print.
Duration Decides Which Technology Is Actually Cheaper
The second thing I had wrong was treating "energy storage" as one product category. A 4-hour lithium-ion system and a 100-hour iron-air system aren't two flavors of the same thing. They're different tools designed for different jobs.
The conventional wisdom says lithium-ion is the cheapest storage technology, period. In my experience, that's only true within certain durations. Lazard's levelized cost of storage reports have shown for years that the LCOS ranking shifts as discharge duration moves from 4 hours to 8, 12, or 100 hours. For short-duration peaking and same-day arbitrage, lithium is hard to beat. But when the job is multi-day firming—say, three days with no wind and heavy cloud cover—the cost picture changes.
Let me be clear: this is not a knock on lithium. We still spec lithium BESS systems for plenty of applications; most projects I'm involved with include a lithium component. The problem is using it for the wrong duty cycle. If you buy a lithium system sized for 100 hours of storage, you're paying for capacity you might use a dozen times per year. The TCO per MWh delivered is terrible, not because the chemistry is bad, but because the duty cycle doesn't match the asset.
That's where long-duration storage changes the conversation. Form Energy's battery energy storage platform is probably the best-known example right now—iron-air chemistry, built for discharge durations up to 100 hours. The company has been in the news for scaling up its Weirton, West Virginia factory and moving its first utility-scale projects into commercial operation. The reason Form Energy news keeps showing up in power-sector trade publications isn't that iron-air beats lithium on the initial $/kWh. It's that the total cost of serving a multi-day profile is competitive, and in some scenarios, cheaper.
One honest caveat: this was accurate as of the market data I'd seen through early 2026. Prices keep moving—lithium costs have been sliding for two years—so verify current numbers before you lock in a budget. Don't hold me to the specific $/MWh figures; the ones I saw in Q1 will probably be outdated within a couple of quarters.
Supplier Structure Changes the Cost Picture
Here's the thing that caught me off guard the most: the same cell chemistry from two different manufacturers can have very different total costs, because the system design and the commercial package aren't the same.
Most buyers focus on the obvious stuff—rated power, capacity, price—and completely miss what the manufacturer is willing to stand behind. The question everyone asks is "are these components UL certified?" The better question is "what exactly is certified, and what happens when a local code official asks to see it?" UL 9540A thermal runaway propagation testing and UL 9540 system listing matter, because fire marshals and authorities having jurisdiction (AHJs) are asking for them more and more under NFPA 855.
NFPA 855 is a good example. It imposes siting limits, spacing requirements, and fire protection measures based on technology and system configuration. A manufacturer that has already worked through those details—and has the documentation to prove it—can save you months of engineering time. A manufacturer that treats compliance as an afterthought? You'll be the one explaining the system to the fire marshal.
This is also where energy storage system wholesale and OEM/private label decisions come in. If you're buying from an integrator or rebranding a manufacturer's hardware under your own name, the TCO question doesn't disappear—it gets harder. You now own the compliance story, the spare-parts chain, and the service network. That can work well; some of the best deployments I've seen came through OEM partnerships. But it only works if the underlying BESS manufacturer has real system-level testing and a warranty they've actually honored before. Otherwise, the risk transfers straight to your balance sheet.
From the outside, a battery module looks like a commodity. The reality is that thermal management, controls, monitoring, container layout, and commissioning support differ wildly between manufacturers. Those differences don't show up on the initial quote. They show up on your O&M budget, three years in.
"I Can't Sell TCO to My Budget Committee." Yes, You Can.
I've heard the objections. "Our procurement policy requires the lowest responsive bid." "My CFO doesn't want to hear about degradation curves." I get it—I've sat in those meetings.
Here's the thing: a TCO model is exactly how you win that meeting. A side-by-side LCOS comparison turns an argument about upfront price into a data-driven conversation about net cost per MWh delivered. I've done it myself. The last time we shortlisted suppliers, the bidder with the lowest unit price lost once we ran the degradation, warranty, and O&M numbers. We placed the order with a manufacturer that looked more expensive on paper and saved—don't hold me to the exact figure—roughly $180,000 over the project's life.
And honestly, "cheapest upfront" isn't the low-risk option people assume. The expensive failure mode is buying a system that underperforms, fails an interconnection review, or loses its support chain when a supplier rebrands its product line. That's what blows up project budgets.
Another objection I understand is "this is too much analysis for what I'm buying." Fair enough—if it's a one-off pilot with a small budget, skip the deep TCO model. But a multi-MWh BESS is a seven-figure asset. A week of spreadsheet work is a rounding error compared with the cost of picking the wrong manufacturer.
So no, I'm not going to soften the conclusion: if you're evaluating BESS manufacturers without a TCO model, you are leaving real money on the table.
The Bottom Line: Make Suppliers Prove Their TCO
Energy storage procurement has gotten more professional since around 2023, and that's a good thing. The buyers getting the best results are asking suppliers for LCOS schedules, degradation guarantees, and a warranty coverage map—not just a unit price.
That's the level of detail you should expect from any serious energy storage system specification guide. If a manufacturer won't put their performance assumptions in writing—or keeps steering the conversation back to $/kWh—that tells you something about how their product will behave over the next decade.
Personally, I'm done negotiating on unit price alone. I load the TCO spreadsheet, compare suppliers on delivered MWh cost, and the ones willing to back their numbers get the order. If you ask me, that's not a neutral position. It's the one that protects the budget.