It was 7:40 on a Tuesday morning in February 2023, and I was staring at two pages of numbers. I'd handled close to $180,000 in cumulative LED component spending across six years — every invoice logged in our procurement system — and I thought I'd gotten past being fooled by unit price. That morning I almost wasn't.
Two quotes sat in front of me. Identical specs: 100–277VAC input, 700mA constant current output, power factor ≥ 0.9, for 1,800 units of the 4-inch downlight we were building for a retail chain retrofit.
The first came from an Illinois-based led driver distributor — basically a middleman. Unit price: $8.80.
The second came from a direct manufacturer. Unit price: $9.90.
At 1,800 units, that's a $1,980 gap. Not life-changing money, but not nothing either. I manage the LED components budget for our company — roughly $125,000 a quarter, closer to $180,000 accumulated across recent years — and shaving a little off every order adds up.
I went with the $8.80 quote.
Six weeks later, that order had cost us $19,768. The $9.90 quote, all-inclusive, would have been $17,820.
I'll walk through what happened.
What we were actually solving for
We're a small OEM lighting assembler, about 85 people, based in Indiana. LED drivers — the constant current drivers that sit inside a finished fixture — make up roughly 40% of my purchasing budget. Because making a fixture is basically housing, heatsink, driver. The housing and heatsink are simple. The driver is where problems live.
Late 2022, we won a retail chain retrofit project. 34 stores, roughly 50 fixtures per store. 1,800 custom 4-inch downlights total.
Not the biggest order we've taken, but the client was clear on two things from the first call: an 11-week delivery window, and "no surprises."
I should have taken that last word more seriously.
Three suppliers, three different stories
I'd collected quotes from three sources.
Our incumbent supplier, a large Chinese manufacturer, was at $10.40. Reliable, consistent lead times, UL/cUL documentation on file, but their pricing had never been aggressive and their MOQ tiers didn't leave much negotiation room at our volume.
The Illinois distributor came in at $8.80. They claimed direct sourcing, two-week delivery, and they sent a datasheet that looked reasonable — voltage range matched, ±5% current accuracy, five-year warranty.
And there was Moso, a direct LED driver manufacturer, at $9.90. I'll be honest: I almost didn't look closely at the Moso driver quote. My instinct at that moment was: if a distributor can do $8.80, why would a factory quote $9.90?
I get why I thought that. I just wish I'd paused for thirty seconds before deciding.
When the pretty price started cracking
First batch landed in two weeks. We tested 20 units out of the pallet. Three had output current drift beyond 1% — not alarming, but worse than our usual. I emailed the distributor. They responded quickly: "Within ±5% tolerance, meets spec."
They weren't wrong. I just hadn't specified tighter. That one's on me — but it's a data point.
Second batch, around 800 units, is where it broke. We needed UL/cUL paperwork per batch for the client's safety consultant. The distributor forwarded the original file — an outdated two-year-old report with somebody else's factory name on it. The consultant rejected the shipment.
It took us close to four weeks — or rather, closer to five once you counted the revision cycle — to get a new file. The distributor charged $450 per document. Then there was a $380 "expedite fee" to re-issue a replacement batch.
By April, one box turned up with physical damage — around 60 units, cracked housings. Could have been transit, could have been inbound. We couldn't prove it either way. The distributor said it wasn't their problem. We ate 60 × $8.80 = $528.
Around that point, I opened a spreadsheet and started tracking every dollar that had actually left our account on that P.O.
What the order really cost
Let me lay it out.
- Base price: $8.80 × 1,800 = $15,840
- Document re-issue fees: $830
- One inspection rework internally: roughly $1,420 (staff time plus a delayed sub-assembly)
- Damaged units: $528
- My own hours on this: call it $1,150 of internal load
Total: about $19,768, or $10.98 per unit.
So the "saving" of $1,980 cost us $3,928 in additional burden. Net loss on the decision: roughly $1,948, plus a strained client relationship that took another two months to smooth over.
The Moso quote — $9.90 — was $17,820 all-in. No document charges, no surprise expedites, no rework cycle.
12.5% higher on unit price. 12.6% lower on actual landed cost.
What I got wrong
To be fair, I don't think anyone was being deliberately deceptive. There are good led driver distributors out there. My mistake was comparing part of the cost against part of the cost.
I wasn't buying LED drivers. I was buying:
- The landed cost per working unit
- Somebody who picks up the phone when a batch is off
- Compliance documentation that will hold up to a client's safety consultant
- Delivery inside an 11-week window, not "probably"
- My own time I don't have to spend chasing invoices, files, and email threads
Strip it down and the $9.90 Moso led driver and the $8.80 distributor quote were not the same purchase at all.
The TCO framework I use now
The conventional wisdom is to always get three quotes. My experience with 200+ orders over six years suggests that relationship consistency and total cost tracking matter more than marginal unit savings — especially for anything with a compliance or deadline component.
So I changed our quote template. It's not fancy. Every supplier now gets scored across six cost lines:
- Base unit price × quantity
- All add-on fees — tooling, documentation, packaging changes
- Freight and handling
- Estimated rework or scrap rate (based on historical supplier data)
- Compliance documentation cost, if the client requires it
- Internal burden — hours we expect to spend, at an internal hourly rate
Then divide the total by usable units. That's the number that matters.
The first quarter after I built this thing, I discovered one of our existing suppliers was actually 8% cheaper per landed unit than the one we'd been treating as our lowest-cost source. We just hadn't seen it, because we'd always compared base price.
One piece of conventional advice I no longer trust
"Get three quotes" is fine for commodity buying. It falls apart for anything with certification, deadlines, or a client attached.
The advice assumes quotes are comparable. They usually aren't. And if you're sourcing LED power supply wholesale — whether for OEM production or resale — treating an all-inclusive supplier quote and a stripped-down one as comparable is how you end up writing your own led power supply wholesale cost guide the hard way, invoice by invoice.
I wish I'd understood that before I found out on a spreadsheet.
What I'd say to the next person in my chair
If there's one thing worth carrying out of this whole mess, it's a shift in the question you ask first.
Before "who's cheaper," ask "cheaper than what."
Two quotes that look adjacent on a spreadsheet are frequently not the same purchase at all. One of them includes the phone call when something goes wrong. The other one doesn't.
In LED drivers specifically, I've come to think that the cheapest quote is rarely the cheapest order. It just defers the cost — into another invoice, a rework cycle, or a delivery schedule that quietly slips while a client waits.
You end up paying for the saving twice.
That's the part nobody puts on the quote.
