Technical note
I Stopped Buying Inverters on GW Numbers — Including SMA Solar's 2023 Figures
2026-09-28 · Adewale Okoye
Shipment volume is the least useful number in most solar distributor buying guides — and it's the one everybody leads with.
I get why. "X gigawatts shipped" sounds like proof of something. It's easy to put on a slide. It's the first thing a sales rep tells you, and the last thing anyone actually verifies.
I don't buy on it anymore. Not because scale doesn't matter — it does, and I'll get to that — but because in five years of managing component purchasing for a distribution and installation business, I've never once had a problem that a bigger GW number would have prevented. Every problem I've actually had was a documentation problem, a spec-interpretation problem, or an aftermarket problem.
So here's my position: treat reported shipment volume as a screening filter, not a buying criterion. If two suppliers both clear your filter, the number tells you nothing about which one ships your order correctly. I manage roughly $6M a year across 11 suppliers, and I report into both operations and finance — which means I'm the one explaining delays, not the one quoting headlines.
The SMA Solar 2023 numbers, and why I don't anchor on them
This comes up constantly, because SMA Solar is one of the names people benchmark against. SMA's 2023 inverter sales volume gets quoted in the low-20-GW range — 20.5 GW is the figure I've seen most often for inverter output sold, with group revenue somewhere in the region of €1.9 billion. I want to be straight with you: I'm working from memory and from secondhand summaries here, not from a line-by-line read of the SMA Group annual report. If you're making a six-figure commitment, pull the annual report and read the segment notes yourself. Don't take a number from a blog post, including this one.
But the part that actually matters applies whether you're looking at SMA or anyone else:
The headline figure isn't standardized across vendors. Depending on who's reporting, "shipments" can mean inverter output sold, units shipped, AC nameplate, DC-coupled capacity, group level or segment level. A 20 GW year for one company and a 20 GW year for another might be measuring two different things. If I remember correctly, there was a stretch where half the datasheets I compared mixed AC and DC ratings without labeling them clearly. Same trap, bigger scale.
And 2023 data is a lagging indicator. It tells you what demand looked like two years ago. It doesn't tell you whether your 400-unit order gets allocated in a tight quarter, and it definitely doesn't tell you whether the firmware on the unit you receive is still being maintained in 2031.
What a GW number does tell you is that a company can run a supply chain at volume. That's a real signal. It's just not the signal you need when you're choosing between two quotes.
Microinverter OEM programs: the label isn't the product
From the outside, a microinverter OEM arrangement looks like a branding exercise — same hardware, different logo, better margin. The reality is that you're buying into somebody's certification, firmware, and warranty structure, and those three things are what you'll live with for the next decade.
When I evaluate a microinverter OEM or a microinverter distributor relationship, the first thing I ask for isn't a price list. It's the certificate chain. Concretely:
- Safety certification under IEC 62109-1 and -2, or UL 1741 / UL 1741 SB for North American deployments
- Grid-code compliance for the actual market — IEEE 1547-2018 in the US, EN 50549-1 across much of Europe, VDE-AR-N 4105 or 4110 for Germany, plus whatever the local network operator genuinely accepts
- Who holds the declaration of conformity — the OEM, the brand, or the distributor? And if it's the distributor, what happens to that document if you change distributors?
- Module-level certifications if you're bundling — IEC 61215 and IEC 61730 for the panels themselves
Certificates expire, and standards get revised. We had a shipment held because the version of a grid-code certificate on file had been superseded and nobody on the supplier side flagged it. Nine days of delay on 400 units. No one's GW number prevented that.
The firmware question is the one that keeps me up. If you buy under an OEM label, ask who pushes firmware updates, how long they've committed to doing it, and what happens if the OEM relationship ends. Get it in writing. Looking back at our first OEM contract, I should have asked that before we signed rather than eight months in, when we needed a fix for a grid-code update. At the time the pricing was good and the lead time was short, and I let those two things carry the decision. That was a mistake.
Documentation throughput is the actual competitive advantage
Here's the claim I'd defend hardest: the operational difference between a good supplier and an average one isn't quality or price — it's how fast their documentation moves.
The most frustrating part of this job is that the same issues recur even when you've communicated clearly. You'd think a written spec sheet plus a signed PO would eliminate ambiguity, but interpretation varies wildly between sales, engineering, and logistics at the same company.
Where that shows up in real money: every time I email a rep to chase a datasheet, a certificate, or a shipping document, that's a day. Sometimes two. Multiply it across 60–80 orders a year and you're talking weeks of internal time — mine, plus whoever in accounting is holding payment until the paperwork lands.
What changed things for us wasn't finding better suppliers. It was moving verification earlier and making it digital. We built a checklist that runs before the PO goes out — certificates pulled from the supplier's portal rather than an email attachment, with expiration dates logged. Once that existed, the "surprise, the certificate lapsed" category of delay basically disappeared.
That's the efficiency argument in this industry, and it isn't about automation for its own sake. A supplier who publishes current certificates, spec sheets, and firmware release notes in a portal you can pull from removes a manual step from your process. A supplier who emails you a PDF when you ask is doing the same thing five years later, just slower. Over a year, that gap compounds.
I'm not saying email-based suppliers are bad. Some of our most reliable partners still work that way, and for high-touch custom orders a human who knows your account is worth more than any portal. But for standardized, repeat orders — which is most of what a distributor buys — the difference is measurable.
What I actually screen on now
If you're building a solar panel distributor buying guide for your own team, this is roughly the sequence I'd use:
- Market fit first. Does the product carry a current certificate for the grid code and safety standard in the market you're selling into? Not "similar to" — current.
- Documentation access. Can you pull datasheets, certificates, and firmware notes yourself, or does everything route through a rep?
- Aftermarket structure. Who handles RMA, who owns spare parts, and what's the warranty transfer path if you sell on to a project owner?
- Spec alignment. Confirm AC vs DC ratings, temperature derating, and monitoring compatibility against what you're actually installing.
- Scale and financials — last, and mostly as a filter. Pull the annual report and look at the segment your product sits in, not the group headline.
Notice where the GW number lands on that list. It's fifth, and it's a filter, not a score.
"But scale means bankability"
Fair pushback, and I'll concede most of it. If you're supplying a utility-scale project, or anything that touches project finance, lender requirements are real and tier-1 scale is often a hard gate. A supplier that can't demonstrate volume may not clear it at all. In that context the shipment number isn't decoration — it's a contractual checkbox.
What I'd push back on is treating that checkbox as the whole decision. Bankability gets you into the room. It doesn't tell you whether your specific order arrives with the right firmware, the right certificate revision, and a serial number registered in the warranty system. Those are the things that generate the phone calls.
Worth saying too: a supplier can be large and still be slow on documentation. Scale and process efficiency aren't the same thing. I've worked with small distributors whose document workflow was cleaner than companies ten times their size, and I've worked with the reverse.
Where I land
Use the SMA Solar 2023 shipment figures, and everyone else's, the way you'd use a credit score. It tells you whether to keep reading. It doesn't tell you whether to sign.
What tells you whether to sign is whether the certificate chain, the spec alignment, and the aftermarket path hold up under your own process — and whether the supplier's side of that process moves fast enough to keep up with yours. That's the difference I've actually felt over five years, and no gigawatt figure has ever shown up in a delay report.