The Tuesday Call
It was 4 p.m. on a Tuesday in November 2024.
I was at my desk at Signalcrest Sourcing—we handle sourcing and supply for turn signal parts, mostly for B2B distributors and fleet repair shops—when the phone rang.
On the other end was a purchasing manager at a mid-size auto parts distributor in Ohio. They had just taken on a vehicle safety inspection contract for a local fleet. The previous Friday, their first shipment had arrived: 400 aftermarket turn signal switches, ordered from a different supplier.
The problem: the lever action felt wrong. They grabbed three samples and bench-tested them. Two didn't conduct.
They called their original supplier. The response: "It's universal fit."
Turns out that "universal" was literal—a one-size lever head marketed as compatible with multiple models. On the actual vehicles, it wouldn't seat. The stalk shape didn't match. The connector was pinned the wrong way.
The fleet inspector stood nearby, took one look, and rejected the batch.
Next inspection: Monday morning, 9 a.m.
They needed 400 working turn signal switches in four days.
(Quick side note: FTC advertising rules do require that "universal fit" claims for aftermarket parts be substantiated with actual evidence. I'm not suggesting you audit every compatibility claim with a microscope—but if a supplier's answer to "is this supposed to be a drop-in?" is just "it's universal, don't worry," that's a red flag worth logging.)
The Problem with 400 Units
In the auto parts world, turn signal switches are one of those categories with dozens of SKUs and low per-unit pricing. Most factories don't want to turn on a line for 400 pieces.
I made calls. One factory quoted a 2,000-unit minimum. Another said they could do 400, but lead time was 18 days.
It took five calls to find a channel partner we'd worked with before—a factory whose standard MOQ for turn signal switches is 500 units, but we had history.
I called them. "400 pieces. Can you make an exception?"
There was a pause on the line. "400 is low... but you're an old customer."
They agreed to pull capacity from another line and slot us in. 400 from the 500-unit MOQ, with the remaining 100 going into their own stock.
So far, so manageable. We placed the order Wednesday, the factory confirmed production Thursday, and a batch shipped out Friday at 3 p.m.
The Friday Surprise
We waited for delivery. Friday at 4 p.m., the warehouse messaged me: "Shipment arrived. But the connector on some of these doesn't match."
My first reaction: What?
A colleague sent photos—of the 400 switches, 65 had a different harness connector spec. The factory had pulled those 65 from another batch and apparently didn't cross-check the drawing.
I won't pretend I kept my blood pressure flat.
(Honestly, that one's on me. I assumed "established partner" was enough. I didn't ask for an off-line photo on Friday morning. I figured a small order like this wouldn't hit a snag. It did. Note to self: even with repeat partners, first-off inspection photos on Friday, no exceptions. And connector specs get their own page, not page three of the attachment.)
Electrically, those 65 pieces would work. But the housing color and clip retention were different. For a fleet inspection with centralized parts review, uniformity was non-negotiable. An inspector comparing against the original would flag it immediately.
The problem was now simple: 65 wrong, 335 usable. Split the shipment or don't?
Weighing the Risk
I laid out the options:
- Send the whole batch back and redo 400. Best case, Monday morning production, then standard freight—Tuesday arrival. That misses the inspection entirely.
- Ship 335 now, send 65 by Monday morning. Would the client accept a split delivery?
I called the client's purchasing manager directly.
He was quiet for a few seconds.
"Fine. But the 65 have to be there by 10 a.m. Monday. Our inspection starts at 1 p.m."
I did the math—
65 pieces on a weekend overtime run. Ship Sunday night. Air freight into Ohio, wheels down by 8 a.m. Monday. Courier from the airport to the client's dock, delivery between 9 and 10.
Freight cost: roughly $600. Order value: under $2,800.
Margin on this whole order, after freight? Flat.
"We'll do it."
That's not a rational commercial decision. Spending $600 in freight against a $2,800 order is a loss on paper. But the client's situation was real—they had a fleet inspection contract. Missing that week didn't just kill one order. It could cost them an entire quarter of supply work.
(In the moment, my test is simple: if I'm the buyer on the other end of this line, what do I need the person on this side to do? That's usually the answer.)
The Weekend
I didn't sleep well on Saturday.
Sunday night, the replacement 65 went out on a plane.
Monday morning, 8:42 a.m., the courier handed the box to the client's receiving desk.
By 11 a.m., the client sent photos—the full batch of turn signal switches installed, fleet vehicles queued up for inspection.
Result? All passed.
That afternoon I texted: "How'd it go?"
The purchasing manager replied: "Thanks for not dropping the ball on us."
What I Kept Thinking About
That client became a repeat customer. The next quarter, they placed a single order for 3,000 switches.
But what actually made me want to write this down wasn't the outcome. It was the first sentence of that phone call—"We called a few places. Soon as they heard 400, they said they couldn't do it."
I get why factories set MOQs. Changeovers, tooling, scheduling—fixed costs sit there on the floor. 400 units and 4,000 units have completely different per-unit economics. I don't hold it against a supplier who sets a minimum. That's their business model, not their character.
The issue is what a buyer hears when they're told "we don't do small orders." It sounds like you're not worth it. And that feeling is real.
Three years of handling emergency orders—over 200 of them at this point—have made me more convinced of one thing: a small order isn't "not worth doing." It's an order you haven't seen the full shape of yet. A customer starts with 200 units, comes back for 2,000. That happens more than I initially expected (though part of that is just the client growing, not all supplier magic). But if the trial order is handled badly, none of what comes after happens.
And another thing: a small batch is a test signal. Ship 400, the client installs them, drives them, sends feedback—you learn more about how your part performs in a real bay than you'd get from a month of lab sheets.
If You're Stuck in the Same Spot
A few practical things, not theory:
First: put the connectors on their own page. That 65-piece mismatch was a drawing interpretation gap. Now, for every part number, we check harness, connector, and pinout on a single standalone sheet. Not buried in attachment three.
Second: split shipping is more realistic than asking for everything at once. 335 now plus 65 by Monday works better than demanding all 400 land together. The main line gets moving—the client's core need—while the tail catches up.
Third: compatibility isn't just a part number match. Aftermarket turn signal switches often cross models and year ranges. With a universal turn signal switch, the wiring logic matters more than the outer shape. LED turn signal bulbs are the same story—same socket, but the flash rate can be off depending on the flasher relay the vehicle runs. That's a common issue, not always a defect. (Rear turn signal bulb upgrades to LED are where this shows up most—people mistake frequency change for a bad bulb.)
Bottom Line
I've asked myself since: was a 400-switch order worth that level of effort?
On pure margin, no.
But the value of a small order was never in the order itself. It's in the signal—whether you treat a customer's problem as real at the moment it's smallest. Do it, and they remember. Don't, and they remember that too.
400 switches isn't a lot. But caught in a 72-hour window, it's a very real problem.
And if the cost of solving a real problem is one order that breaks even, I'll take that trade.