I'm a Procurement Manager. Here's What SMA's 19.5 GW in 2023 Taught Me About Solar Battery Costs

In March 2024, I sat down with a spreadsheet and a sense of dread. I manage procurement for a 40-person solar installer in Colorado, and we'd just decided to add battery storage as a standard offering. My boss gave me the deadline: "Figure out what this costs, and make a pricing model by the end of the month."

I've been tracking every equipment dollar for the last seven years—about $2.1 million in total purchases across our projects. But storage was new. And if there's one thing I've learned in procurement, it's that new products are where the budget goes to die.

Why SMA's 2023 Shipments Mattered

The first thing I did was look at what the big manufacturers were shipping. If you're an installer, you need to make sure your hardware ecosystem isn't going to be abandoned in two years. SMA is a stock brand for us, so I pulled their 2023 figures.

According to SMA's 2023 Annual Report, they shipped about 19.5 GW of inverters last year—up significantly from previous years (I think the 2022 number was around 12.7 GW). Source: SMA Investor Relations, ir.sma.de. That kind of volume tells you two things: string inverters aren't going away, and the market is moving toward hybrid and storage-ready systems. If SMA is shifting its sales mix that way, we need to be there too.

So I started digging into battery sizes and prices. And I made a mistake almost immediately.

The First Mistake: Thinking "kWh" Was the Whole Story

My initial approach to battery selection was stupidly simple: a customer wants 10 kWh of backup, I quote a 10 kWh battery, done. I even created a comparison table with the nominal capacities and the listed prices from a few battery manufacturers. Easy, right?

It took me two days to realize how wrong that was. The first red flag: "10 kWh" on the spec sheet is rarely 10 kWh of usable energy. Some batteries have a 90% depth of discharge, some have 100%, but the inverter efficiency and the battery management system can eat into that. More importantly, a 10 kWh battery with a 3 kW continuous output can't run a 5 kW load. So the "size" you're buying is actually a combination of capacity (kWh) and power (kW).

I remember thinking: the entire industry is using the same words but meaning different things. I said "10 kWh battery" to a sales rep, and they heard "10 kWh of usable capacity at 100% DOD." My customer heard "10 kWh of backup, so I can run my AC." We weren't even close.

The Hidden Cost That Almost Bit Me

Then I nearly signed off on a battery that was 18% cheaper on paper. The manufacturer's spec showed an attractive price per kWh. It wasn't an SMA-branded battery, but they said they "supported SMA inverters."

Spoiler: they did, but not natively. To connect their battery to the SMA's Sunny Boy Storage, I needed an extra gateway module that cost $500, plus a communication adapter, plus a new firmware version from the battery vendor that required a (paid) commissioning call. The cheap battery ended up costing more than the SMA High Voltage battery once I added all the pieces.

That's the classic overconfidence failure. I'd been burned by this exact thing in 2022 with a different vendor, and I have a note in my procurement system that says, "always verify compatibility before doing price comparisons." I skipped it because I was rushing. And the result was a $500 mistake that I caught before we ordered—but only after two emails, one hold-on-the-phone, and a technician's time. That "free support" call after the sale was not free.

This is where the total cost of ownership mindset saved me. The listed price for the battery was, say, $1,200 less than the SMA alternative. But the extra gateway, the extra programming, the additional 45 minutes of labor, and the risk of a warranty headache if the integration wasn't perfect—that completely eliminated the gap. I built a cost calculator after that burned me. The line item for "integration friction" is now one of the first things I model.

When the Drawing Didn't Tell the Whole Story

Around the same time, a customer emailed me a drawing of their roof. They had a nice array of 15 panels, and they'd circled a spot next to their electric panel. "We want a battery that covers the whole house," they wrote.

The drawing didn't say anything about loads. It didn't show whether they had electric heat, an EV charger, or a well pump. If I'd used that drawing to size a battery, I would have installed a system that either ran out of power in an hour or was two times too big.

We had a conversation where they said "whole house" and I heard "typical whole house, say 30 kWh per day." They actually meant "the essentials I can't live without for a two-day outage." That's a communication failure that could have cost them an extra $8,000. Now, here's what I've learned: in a solar system drawing, the energy flow matters more than the panel layout. Battery sizing isn't about roof area. It's about load profile—what runs, when, and for how long.

The result is that every proposal we do now starts with a load calculation. We have a template that asks about outages, appliances, and whether they have an electric car. We even include a line in the contract that says "design assumes the load list provided by the customer." That one line has saved us from a ton of scope creep.

What Battery Sizes and Prices Actually Look Like

So, to answer the question everyone asks: how much does a solar battery cost? In 2024, from the quotes I've generated for actual projects, an installed battery system lands in the $10,000–$20,000 range before incentives. The hardware itself—the battery module and the inverter—is maybe $4,000–$8,000, but you have to add permitting, electrical labor, the current transformer, a transfer switch, and the inevitable miscellaneous stuff. In our pricing, I include a 10% contingency buffer, because if there's one thing that never changes, it's that some part will take longer than expected.

For context, here's a rough breakdown of what we use as standard tiers:

• A 9.6 kWh usable LFP battery with 5 kW continuous output—packaged with an SMA inverter—typically quotes around $12,000 installed.
• A 16 kWh system with 8 kW output—like two battery modules in series—comes in around $16,000–$18,000.
• A 23 kWh system with 10 kW output—a serious whole-home backup stack—can hit $22,000 or more.

But the price per kWh is not the only number. A 10 kWh battery that only gives you 9 kWh before the inverter shuts down is a different deal from one that gives you 9.5 usable. And if the battery's round-trip efficiency is 90% instead of 97%, you're losing money on every cycle. I think the net number is more useful: the cost per usable kWh over the warranty's expected cycles. That's the real measure, not the salesman's "per kWh" figure.

One More Caution: Green Claims

I can't talk about batteries without mentioning the "eco" side of the pitch. A lot of manufacturers market batteries as "fully recyclable" or "100% sustainable." Be careful. Our own vendor's marketing materials used the word "recyclable," but when I checked with our local recycler, they didn't have a program for that chemistry. Per FTC Green Guides (ftc.gov), a product can only be called "recyclable" if recycling facilities are available to at least 60% of consumers in the area where the claim is made. If we'd repeated that claim to a customer, we'd have been misleading them. It's the kind of liability that keeps a cost controller up at night.

The Takeaway

The industry is changing faster than a lot of us are comfortable with. What seemed like a specialty product in 2020—a battery that could back up your essential loads—is now a mainstream line item. The fundamentals haven't changed: you have to know your customer's loads, match the AC and DC sides, and account for the total cost of ownership, not just the sticker price. But the execution has transformed. Prices per kWh have dropped year over year, and the list of "compatible" products is starting to stabilize around a few big ecosystems. SMA's shipment numbers and their tight integration with batteries (both theirs and third-party) are one sign of that.

If you're on the fence about adding storage to your offerings, I get it. The upfront research is a pain. The compatibility curves are real. But the days of telling a customer "we don't do batteries" are ending. The bottom line: a cheap battery that doesn't integrate is not a deal. A battery that's the wrong size is not a deal. A solar drawing that doesn't include the load profile is just a picture. Do the math—including the hidden costs—and you'll be in a position to give your customers what they actually need, at a price that keeps you in business.

Take it from someone who almost learned the hard way: in solar storage, the last thing you want is a shock from the system.