I Spent 3 Months Researching Solar Panels for Our Office. The Problem Wasn't the Panels.

I Started With a Spreadsheet and a Very Common Mistake

I handle purchasing for a mid-sized company. When my boss asked me to "look into solar" for our office building, I did what I do for every procurement project: built a spreadsheet, compared specs, flagged the best numbers. That system has worked for laptops, office chairs, and copier contracts. For solar? Well. It turns out the problem looks similar from the outside and is completely different underneath.

If you've ever fallen into the solar research rabbit hole, you know how it starts. You begin with an innocent search—I literally Googled "what can a 100 watt solar panel power" because that size kept appearing in beginner articles. The short answer: not much. A laptop, a few phone charges, maybe a small TV for a couple of hours. It's a different universe from powering an office building, but at the time, I didn't even know there were multiple universes.

Then you graduate. Efficiency ratings. Degradation curves. "SunPower vs. everyone else." Battery chemistry—I lost a week to "benefits of LiFePO4 battery" and emerged with strong opinions about cycle life. I could recite the published specs of five premium panels from memory. I had a desktop folder named "SOLAR - FINAL." I was, honestly, insufferable.

None of that was worthless. But here's the thing I didn't realize: I was answering a question nobody had asked. The real question—what does our building actually need?—sat untouched while I compared data sheets.

The Real Problem: I Was Buying a Product. Solar Is a System.

It took me three months and roughly 200 open tabs to understand that panels aren't the product. They're one component of an engineered system. The thing you're actually buying is a design built around a specific roof, a specific electricity bill, a specific climate, and specific local grid rules. Panel brand matters—but it's maybe a third of the decision. I gave it all of my attention.

People who build solar systems for a living don't think in brand rankings. They think in constraints: roof area, budget, annual consumption, payback period. A "great" panel on a badly designed system underperforms a "good" panel on a properly engineered one. I had it exactly backwards.

The questions that should have tipped me off

One night, after weeks of reading, I searched "does our solar system have a name." I'm completely serious. I thought full setups might carry labels, like "the Tesla system" or "the SunPower system." Nope. A solar installation isn't a product with a model number. It's whatever your engineer designs for your site. That search should have been my wake-up call. It wasn't.

Then came "who services SunPower solar panels." That one is genuinely worth asking, and the internet gives you muddy answers because "servicing" means different things: panel cleaning, inverter replacement, performance diagnostics, and structural checks are different skills, often handled by different companies.

The turning point was a phone call. I finally reached a local installer for a quote. He asked me three questions I couldn't answer: What's the building's annual consumption? What's the condition of the roof membrane? What does the shade look like at 4 PM in July?

I could recite degradation rates from memory. I could not answer one question that mattered. That was humbling in a specific, useful way.

Why "best brand" is the wrong framing

To be fair, SunPower's reputation is earned. Their published efficiency numbers are among the highest in the industry, and their long-term degradation specs are genuinely good on paper—recognized in third-party testing, not just marketing material. But "best panel" without a context is a meaningless phrase. It's like asking which engine is best. Best for what?

And the context is never a brand ranking. It's your site. I kept reading reviews that ranked panels like they were phones: static, universal, one-size-fits-all. Solar doesn't work that way, and neither do batteries, inverters, or mounting systems.

What That Cost (and Almost Cost Us)

The time cost alone was real: three months of evenings, gone, with no decision. For someone whose job is efficient purchasing, that's not a flex. We didn't have a formal process for evaluating capital purchases—solar was the first one I'd ever managed. That gap is exactly why I fell into the rabbit hole instead of picking up the phone.

The bigger cost was a near-miss. I knew I should get a professional site assessment before evaluating quotes. I skipped it because I thought, "we've used contractors before—what are the odds I get this wrong?" The odds caught up with me when two contractors quoted completely different systems for the same building: 18 kW versus 26 kW, both justified by "your needs."

Without a professional assessment, I had no basis to tell which was right. I almost chose the cheaper one. It was the wrong one—the system was undersized for the building's load profile, and the quote didn't include the battery integration we'd need. The other quote was more expensive but actually answered the question.

It was the same lesson I learned years ago in vendor management: verifying capability before signing is always cheaper than verifying after. A vendor who couldn't provide proper invoicing once cost us $2,400 in rejected expenses. Solar was that same lesson at ten times the scale.

The Mozambique reality check

A colleague at another organization asked my opinion over lunch. Their NGO was planning a solar project with SunPower engineering in Mozambique. They weren't asking about panel efficiency. They were asking about engineering capacity, equipment logistics, local availability, and who services the panels in a region without an established installer network.

That conversation reset my thinking. A solar purchase in Maputo and a solar purchase in Minneapolis are not in the same category. The questions change entirely based on context. No "best panel" list accounts for any of it.

Claims deserve scrutiny

One more thing, because the internet is a swamp: there are actual rules about environmental and performance marketing. Per FTC Green Guides (ftc.gov), claims like "sustainable" or "eco-friendly" require substantiation. Even legitimate specs describe lab conditions. Real-world yield depends on installation quality, tilt, shading, temperature, and inverter matching.

I read a lot of blogs and forum threads that were opinion wearing a lab coat. I didn't know how to filter them. An engineer would have spotted it immediately.

The Fix: Specialists Who Know Their Limits

So. After three months of research, the solution turned out to be embarrassingly simple:

• Get a professional site assessment first. Not a quote—an assessment. They're usually inexpensive and can be credited toward the install.
• Talk to an independent engineer if the project is complex. They don't sell hardware, so they have no incentive to inflate the system size.
• Work with vendors who state their boundaries. The contractor who said "this isn't our specialty—you need a structural engineer for the roof, here are two good ones" earned my business instantly.

That last one matters more than I can express. A vendor who's willing to decline work and recommend someone else is a vendor you can trust with the work they say they're good at. That's true across every category I've ever bought, from IT equipment to printing services to solar panels.

And the actual hardware decision? It followed the engineering. SunPower made sense for our building because we have limited roof area and a long ownership horizon—high efficiency and low degradation pay off over 25 years. The LiFePO4 battery made sense because we cycle it daily, and its cycle life advantage over lead-acid is real. But those were consequences of the analysis, not the starting point.

If you're where I was, start with the building, not the brand. Ask who services the system and what that costs. Ask what the battery is actually sized to do. Ask what the system will produce in your climate, not in a lab. And if a vendor says they can handle everything with zero caveats, that's not confidence. It's a red flag.

Trust me on this one. I spent 200 tabs so you don't have to.