I manage procurement for a mid-sized B2B facility with about 60 employees. My job comes down to one thing: making sure we spend money where it matters and don't waste it where it doesn't. That's why, when operations asked for a UPS for the new server rack six years ago, I did what anyone in my position would do—I looked for the cheapest option that wasn't obviously a fire hazard.
What I didn't know is that the people selling power protection are counting on that instinct.
Search "eaton-ups" and you'll get a page of listings that all look interchangeable. Same VA rating, similar wattage, similar price. It's easy to convince yourself they're the same product with different logos on the front.
The problem is that you're comparing the wrong number. VA rating tells you how much load the UPS can handle. It doesn't tell you how long the battery lasts, how clean the output power is, or what happens when the unit has been running for three years. Those are the numbers that actually matter, and they're the ones no comparison shopping site will show you. Some distributors won't even publish them.
Let me bust the assumption that caused more bad purchases than anything else: people think a higher VA rating means longer runtime. Actually, runtime is determined by the wattage of what you plug in and the battery capacity. Two 1000VA units from different manufacturers can have completely different runtime curves. The VA number is a ceiling, not a timeline.
I've seen the reverse, too. The assumption is that a 1000VA UPS is way more than enough for a small rack. The reality is that the load adds up faster than you think—servers, network switches, firewalls, modems, even the controller for the door access system. When I audited our 2023 spending, I found one recurring thread: nearly every "unexpected shutdown" traced back to a load calculation that left off one device. I once watched a $12,000 server shut down because we forgot the PoE switch was on the same circuit as the rack. The UPS was sized perfectly for the servers. It wasn't sized for the thing that connects everything else.
When I say "cheap," I do not mean the price tag. I mean the total cost nobody tracks until it hits the ledger.
In 2023, I compared costs across eight vendors. Vendor A quoted $1,850 for an Eaton 9PX. Vendor B quoted $1,490 for a comparable unit from a brand I'll leave unnamed. I almost went with B. The spreadsheet said we'd save $360 per unit, and we needed four. That's $1,440. The numbers pointed one way; my gut said something felt off—their emails kept getting slower, and two references never called me back.
I signed the PO anyway.
Even after choosing Vendor B, I kept second-guessing. What if their batteries were actually fine? What if I was being paranoid about slow email responses? I didn't relax until the units arrived—and they did arrive, on time, with full documentation. For the first two months, everything looked great.
Then the beeping started.
Within 11 months, two of the four units had dead batteries. The warranty process took four days just to get a response, and replacement batteries cost $220 each plus shipping. The third unit had a firmware bug that triggered random false alarms. Their "fix" was to wait for an update that took another two months. The fourth unit was fine. I'll give them that.
To be fair, their pricing was competitive for what they offered. But what they offered wasn't the same thing. We replaced all four with Eaton units in Q2 2024, and I built a cost calculator the same week.
Here's the math no sales rep would show you:
$1,440 saved upfront − $1,760 in replacement batteries and shipping − $540 in electrician overtime for an emergency swap = −$860 net loss, plus three weeks of elevated downtime risk.
After tracking 40+ power protection orders over six years in our procurement system, I found that roughly 30% of our budget overruns came from emergency battery replacements and zero-notice installations. We now have a policy: any UPS over 1500VA gets a battery refresh schedule and a spare set of batteries in storage. That single policy cut our emergency spend by half.
While we were fixing that mess, I learned about something else most UPS buyers don't consider: the transfer switch. If you have a generator—and a facility our size should—the UPS has to be wired behind the switch correctly. Otherwise, when the utility power drops, the generator starts, the transfer switch flips, and the UPS sees dirty power and decides to stay on battery. 25 minutes later, the battery drains, and everything dies anyway.
"How to wire a generator transfer switch" is a search that sounds like a weekend project until you realize it involves code compliance, grounding, and interlock kits. We paid a licensed electrician $1,800 to install a Reliance transfer switch kit on our panel. It was worth every penny, but it was also a cost nobody included when we set the original UPS budget. Add it to your total cost of ownership spreadsheet, or you'll be unpleasantly surprised halfway through the project when the electrician hands you a change order.
Here's a clue about how confused the market is: one of the searches that led someone to our blog was "hunter douglas battery charger." Hunter Douglas makes window blinds. They do not make UPS systems. The person was probably looking for a spare battery charger for a motorized blind, and Google showed them power protection products instead.
The mix-up points at a bigger problem: people treat any device with a battery as though it's part of the same category. It isn't. A UPS has to detect a power failure in milliseconds and deliver clean power from its batteries. A battery charger's only job is to top up a battery until it's needed. If you conflate the two, you end up with a rack full of equipment that has power—until the wall outlet sneezes, and then nothing.
This confusion is costly. If you're searching for a battery charger, you're probably not planning to protect a critical load. And if you're searching for a UPS, you probably haven't thought about the transfer switch, the generator, or what happens when the batteries age out in year three.
I'm not saying every cheap UPS is garbage. I'm saying the purchase decision should start with your risk, not your price threshold.
If one hour of downtime costs your business less than $500, buy a decent consumer UPS and call it a day. If it costs more—if you're running a data center, a medical practice, a manufacturing line, or any kind of 24/7 operation—then you should be looking at Eaton Corporation's UPS line, calculating runtime at your real load, and planning for the transfer switch before the generator folks show up. The investment in an Eaton 1000VA UPS or a larger modular unit is insurance you pay once for years of not thinking about it.
We run Eaton UPS units now, from the 1000VA models in our network closet to the 9PX units in the server room. I recommend them not because they're perfect (I don't think they are), but because their support actually answers the phone, their batteries last, and their power management software doesn't lie about what's happening to your equipment.
Per FTC advertising guidelines, claims like "protected" or "battery backup" have to be truthful and substantiated. The same standard should apply to your internal procurement decision. Verify the specs, measure your real load, and calculate the whole chain—generator, transfer switch, UPS, load. The sticker price is the start of the conversation, not the end of it.
This approach worked for us, but we're a facility with stable grid power and predictable staff hours. If you're in an area with frequent outages or you run production shifts that never stop, the calculus is different—you may need dual UPS units or a larger generator. I can only speak to my context, and in my context, the cheapest option was the most expensive one I ever approved.