We Lost 63 Hours to a Blackout—Here's How I Chose the Right Solar Backup System

The power went out at 2:47 PM on a Tuesday in October 2024. I know the exact time because the UPS alarms in our server room started beeping, one after another, down the hallway. I watched the whole office block go dark outside my window and thought: this is going to hurt.

It hurt worse than I expected. The first blackout lasted about six hours. The second one, two days later, went 63 hours. Our 210-person office lost three full working days—unanswered customer emails, stalled internal systems, and a payroll run we had to finish at an off-site location. When the lights finally came back on, my VP pulled me aside and said, "We can't have this happen again. Figure out what we need."

I'm the office administrator here. When I took over purchasing in 2020, my approach was simple: find the lowest quote that meets the spec, present it to finance, move on. That worked fine for office supplies and cleaning services. It did not work for solar. I handle about $450,000 in annual purchasing across 17 vendors, and I report to both operations and finance. Solar was outside my normal lane. But when your VP asks you to figure something out, you figure it out.

The Search for a Solution

I started researching in November 2024. My first instinct was to buy a solar generator with 30 amp rv outlet. It sounded perfect on paper—portable, rechargeable, with enough power to plug in standard equipment. I nearly ordered one before a colleague pointed out that a 30-amp RV outlet is designed for an RV, not a two-story office. That kind of unit will run a fridge and a few lights for a few hours. It will not keep servers, network switches, and a phone system alive through a two-day outage.

The category I actually needed was home backup power battery storage, scaled up for commercial use. There are a ton of options out there, and my early research pointed me in the wrong direction because I didn't know what to compare.

Here's my initial misjudgment: I assumed the most important number was price per watt. It felt familiar, like comparing paper vendors by the ream. I collected quotes from three solar installers. One came in considerably lower than the other two. I nearly signed it—I actually put the contract in front of my finance manager. Then I asked the installer about the panel degradation rate. He sent me a spec sheet with a number he couldn't explain, and when I pressed him, he said, "Most of my customers just want the most powerful panel for the least money."

That answer should have been a red flag. It turned out to be.

The Number That Matters

Here's something vendors won't tell you: the efficiency rating printed on a solar panel spec sheet is only half the story. The other half is the degradation rate—how quickly the panel's output fades over time. Most buyers focus on the initial watt rating and completely miss this. National Renewable Energy Laboratory (NREL) research puts typical module degradation between 0.5% and 0.8% per year. After 25 years, a 400 W panel at the high end of that range could be producing closer to 300 W. That's a 25% loss of energy generation, compounded annually.

That's when I looked at the SunPower Maxeon 6 solar panel. Its published degradation rate is around 0.25% per year, which means it retains about 92% of its original capacity after 25 years. For a commercial array, the difference between 0.25% and 0.7% annual degradation is tens of thousands of dollars in lost generation over the life of the system.

I checked Maxeon's published specs against FTC advertising guidance (ftc.gov), which requires performance and environmental claims to be substantiated with evidence. The numbers were right there: over 24% efficiency, the 0.25% annual degradation figure, and a 25-year product warranty. I could verify what I was being told. That mattered to me because I've been burned before by a vendor who promised things they couldn't back up—and finance rejected the invoice because it didn't match the agreed terms.

Meanwhile, the cheap installer couldn't commit to an installation date. Every time I asked when they could start, I heard "as soon as possible." I meant: give me a week. They seemed to mean: whenever we get through our queue. When I finally pinned them down, they said they were booking past March 2025.

March would've been fine. April would've been too late—that's when spring storm season hits our region. I sat down with the numbers and realized something: if we went through just one more multi-day outage while waiting for a cheaper system, the lost revenue would outweigh any price difference. The premium I'd pay for a reliable installer wasn't for speed. It was for certainty. And certainty has a price.

When the Guesswork Ended

The European part of this story came when our London office heard what we were doing. The manager there emailed me a simple question: "what size solar battery do I need uk?" I told her what I'd learned the hard way: battery sizing isn't about square footage. It's about loads and hours. You list every device, add up the average watts, multiply by the backup hours you need, then account for the battery's depth of discharge—for lithium-ion, you don't want to run it below about 20% (the deeper you drain it, the faster it wears out).

For our headquarters, that worked out to a 60 kWh storage bank paired with SunPower photovoltaic panels on the south-facing roof. For London, it came to about 30 kWh, enough to keep two small offices and their network equipment running. I also pointed her to the MCS (Microgeneration Certification Scheme) standards so she could verify that any UK-based installer followed the same rigor we were applying here.

We paid a noticeable premium for the SunPower system. The cheap installer's quote was lower, no question. But the cheap quote didn't include a guaranteed install window, didn't have a degradation curve I could trust, and didn't have a warranty structure that gave me confidence the company would still exist 20 years from now. Looking at total cost of ownership—the price of the panels, the storage, the installation, the degraded output over time, and the risk of one more blackout—the decision was obvious.

The Storm That Proved It

Installation happened in early February 2025. The panels went on the south-facing roof, the battery bank went into a repurposed storage room, and the system passed inspection without issues. Then, on March 27, 2025, we got hit with another major storm. The power went out across the region for 18 hours.

Our office stayed lit. Servers stayed up. Phones rang normally. The only sign that anything was wrong was the silence from every other building on the block. I stood in the breakroom watching the battery status screen—around 40% remaining, with the panels starting to recharge it as dawn broke. That was the moment the whole thing paid for itself.

The system has continued performing since then. As of September 2025, the summer generation numbers have beaten the installer's projections, and finance no longer asks skeptical questions about the premium we paid. But the real lessons from this project go beyond solar panels:

• Price per watt is the wrong number to compare. Find the degradation rate and the warranty terms. A difference of 0.4% per year can mean tens of thousands of dollars over a 25-year system life.
• Size battery storage by calculating loads, not by guessing. The UK team's "what size solar battery do I need" question had a single correct answer—and it came from a load calculation, not intuition.
• When a deadline matters, uncertainty costs more than a premium. A vendor who cannot commit to a date isn't cheaper; they're just deferred. And deferred costs in an emergency are always the more expensive ones.

I still buy office supplies the way I always have—lowest quote, best delivery terms. But for anything where timing and reliability matter, I've changed my approach. Sometimes the most expensive thing you can purchase is a promise with an asterisk. This time, I paid for the promise that didn't need one.