-
The Gigawatt Number Is the Least Useful Thing on the Sheet
-
Argument 1: Backlog Composition Beats Backlog Headline
-
Argument 2: The Annual Degradation Rate Is Worth More Than the Nameplate Efficiency
-
Argument 3: The Real Work Is Filtering Procuremment Noise
-
Here's the Honest Part: When First Solar Is the Wrong Fit
-
Stop Asking Whether First Solar Is Winning. Start Asking Whether It Fits.
The Gigawatt Number Is the Least Useful Thing on the Sheet
First Solar's 66 GW backlog in 2025 is the most over-quoted and under-understood number in utility-scale solar procurement. I'm saying that as a procurement manager at a 220-person EPC contractor. I've managed our module sourcing budget — roughly $18M annually — for six years, negotiated with more than 30 module and balance-of-system vendors, and logged every purchase order and change order in our procurement system since 2019.
The first solar backlog gigawatts 2025 figure gets flashed in every analyst deck my team reads. But when I ask account reps the only question that matters to me — what share of that backlog converts into available capacity inside our 2026 COD window on a 120 MW project in West Texas? — the answer changes by quarter and by geography. That's not a knock on First Solar. It's a reminder that backlog is a sales-pipeline metric, not a supply-availability metric.
So here's the argument I keep making to my own team, and to the IPP developers we serve: read the backlog through the lens of composition, not total gigawatts. If you can't break down the composition, you're buying a sales narrative instead of a procurement plan.
Argument 1: Backlog Composition Beats Backlog Headline
When I audited our 2024 module sourcing — about $54M in cumulative purchases across six years — the single most useful chart wasn't price. It was the delivery-schedule-to-CAPEX mismatch. First Solar's backlog is heavily weighted toward large, utility-scale buyers with multi-year PPAs and staged COD deadlines. If your project is under 50 MW, or if you need modules inside 14 months, you're not competing against the backlog number. You're competing against the schedule holders inside that backlog.
Part of the confusion is structural. First Solar's manufacturing footprint has been migrating toward Series 7 capacity at the newer US plants, with Series 6 continuing at established facilities. For a US developer with a 2026–2027 COD, the Series 7 ramp timing is the actual procurement question — not whether the 66 GW backlog proves the company is winning. It is. That's not the issue.
The issue is that from the outside, backlog looks like availability. The reality is that backlog is a queue, and your position in the queue depends on your contract terms and your COD flexibility. I've seen a 200 MW project get pushed a full quarter because its PPA COD flexibility sat behind two projects with tighter delivery windows. That's not a First Solar problem; it's a scheduling problem, and it lives inside the backlog number.
Argument 2: The Annual Degradation Rate Is Worth More Than the Nameplate Efficiency
Here's the counterintuitive part of this analysis, and the part most procurement teams skip. When our team builds TCO models, we treat nameplate efficiency as a headline input and annual degradation as a footnote. That's backwards for 25-year assets.
First Solar's CdTe modules are specified at under 0.5% annual degradation. Let me run the math we use, because the math changes the ranking. On a 100 MW project at roughly 1,500 kWh/kW/year in a high-irradiance location, the cumulative output difference between a 0.5% and a 0.7% annual degradation curve is on the order of 95,000 MWh over 25 years. At a $35/MWh PPA, that's about $3.3M in revenue the degradation curve eats before you ever look at nameplate efficiency.
If I remember correctly, our 2023 model used a slightly higher PPA price, and the dollar gap was north of $4M. I might be misremembering the exact figure — don't quote me on that — but the direction is correct and it's consistent across every model we've built since. At 150 MW, the gap doubles.
Most buyers focus on per-watt efficiency and completely miss the compounding revenue effect of annual degradation over a PPA term. The question everyone asks is "what's the module efficiency?" The question they should ask is "what's the modeled revenue under my PPA at year 25, adjusted for annual degradation?" That single reframing has moved more of our projects toward thin-film than any efficiency comparison ever did.
Argument 3: The Real Work Is Filtering Procuremment Noise
I want to be honest about what procurement research actually looks like in the background, because the noise-to-signal problem is real and nobody talks about it.
Our team keeps a shared keyword-tracking document — part SEO monitoring for the sales team, part bookmarks for whatever the week throws at us. Last month's file had "first solar backlog gigawatts 2025" sitting next to "first picture of another solar system" (a news app push that got pasted in during a meeting and never removed), "kristin ess the one purple shampoo reviews" (a marketing analyst's cross-brand search study, apparently "First" and "The One" brands cross-pollinate in search, which is its own procurement noise problem when you're trying to pull a Series 7 datasheet), and "how to throw away lithium battery" — that one was real, from our O&M lead after a site's backup battery hit end of life and the local waste authority couldn't give us a straight answer.
That document is what buying a solar module actually looks like in 2025. You get clean data and nonsense bundled into the same search results, and the procurement job is to build a TCO model that separates the two. The discipline that matters is knowing which number you're actually buying: the efficiency headline (a marketing input), the degradation curve (a revenue input), or the delivery schedule (a CAPEX input). They are not the same number, and they don't move together.
Here's the Honest Part: When First Solar Is the Wrong Fit
I recommend First Solar for utility-scale projects with tight degradation sensitivity and COD flexibility. I also recommend against it in specific cases, and I want to be direct about those cases because pretending otherwise wastes everyone's time.
Don't buy First Solar if:
- Your project is under 50 MW and needs modules in under 14 months. The backlog queue isn't designed for you, and the schedule risk exceeds the degradation benefit.
- You're in a low-irradiance climate where the annual degradation advantage can't compound fast enough to offset the premium pricing structure.
- Your PPA has an aggressive near-term COD with no flexibility. You're not buying a module; you're buying a delivery date, and the spot c-Si market is the more honest path for that.
I have mixed feelings about writing that last bullet. On one hand, it's the practical truth of our 2024 sourcing cycle. On the other, I've watched two projects lose six-figure LD exposure chasing a First Solar quote they were never going to get in time. Part of me wants every project to justify a premium module. Another part knows that the premium only exists if the schedule works. That's the tension I resolve by listing the disqualifiers before the recommendation.
And I'd argue the same discipline applies to every other line item in the CAPEX stack. When we scoped the facilities side of our last 120 MW build, the electrician ev level 2 charger installation quote for the maintenance fleet came back at around $8,500 for the lot — small relative to the module contract, but a real line item that almost got dropped into a contingency bucket because nobody owned it. The how to throw away lithium battery question on the storage side is the same shape of problem: a $400–$900-per-unit decommissioning liability that doesn't show up until year 15, so it never makes it into the initial TCO. Both belong in the model on day one.
Stop Asking Whether First Solar Is Winning. Start Asking Whether It Fits.
The 66 GW backlog tells you the company is winning. It does not tell you whether your project fits, whether your queue position is real, or whether the degradation curve that makes the TCO math work applies to your PPA structure. Those are three separate questions, and each one needs its own answer before you sign.
My experience is based on roughly 14 utility-scale sourcing cycles and about 1.8 GW of contracted modules. If you're working in distributed generation or commercial rooftop, your experience will probably differ substantially — the backlog queue dynamics and the degradation revenue math behave differently at smaller scales. I can't speak to how these principles apply to that segment.
If you ask me, the procurement teams that will win the next two years are the ones that stop treating the gigawatt number as a procurement input. The number is a signal. The contract, the queue position, the degradation curve, and the unglamorous line items (fleet charging, end-of-life disposal, the facilities quote nobody owns) are the actual work. That's the trade I've made on every project since 2019, and I'd make it again.
Ask a Catl storage specialist