In Q1 2024, I sat down with our engineering lead to review BOM costs for a new industrial data logger. We're a 180-person company that builds environmental monitoring equipment. Our annual interconnect spend runs around $180,000. Not huge, but I track every invoice.
This wasn't a consumer electronics project—no transparent smartphone displays, no YouTube reviews of the best multimeter for electronics. This was industrial equipment that had to work in a factory for a decade. The product needed a high-speed board-to-board connector. Not a barrel jack for power or a TRRS jack for audio—those are commodity parts with predictable pricing. This was a 0.8mm pitch, 30-position vertical connector for differential pairs at 28 Gbps.
We'd been using a generic design from a distributor I won't name. Worked fine for three years. But lead times had crept from two weeks to five, and the rep stopped answering emails.
Time to look at alternatives. Engineering suggested Samtec's QSH series.
I pulled quotes from eight vendors over three weeks. Eight. My spreadsheet had columns for unit price, minimum order quantity, and lead time. That was it at first.
The numbers told a clear story:
Simple math: over our annual volume of roughly 2,000 units, Samtec would cost an extra $3,100 per year. That's 1.7% of our interconnect budget. My spreadsheet highlighted Vendor A in green.
My gut said something else. And I couldn't explain why.
Maybe it was the Samtec rep who answered my email in 20 minutes. Maybe it was the datasheet that actually listed insertion loss at 28 GHz without me having to ask. Maybe it was the sample that arrived in three days, not "three to five business days meaning next week."
But gut feel isn't a procurement strategy. I needed data.
Here's where it got complicated.
While waiting for final quotes, I had our engineering team send detailed specs to all eight vendors. One key requirement: "0.8mm pitch, 30 positions, vertical mating." Basic stuff.
Vendor A came back: "Standard pitch, 30 pin, right angle. Confirmed."
Wait. Right angle?
I went back through our email. I'd written "vertical mating orientation." Their engineer read "standard" and assumed we meant their standard configuration—right angle. Nobody was lying. Nobody was sloppy. We were using the same words but meaning different things. Discovered this when the sample arrived and nothing fit our existing footprint.
That's when I added a "sample verification" column to the spreadsheet. And a "spec acknowledgment signed by engineering" requirement. And—most importantly—a column for "engineering support responsiveness."
Samtec sent a $0 sample with a handwritten note from their FAE offering to walk through our stack-up. The others sent PDFs.
The most frustrating part? This wasn't the first time. Two other vendors sent samples last year with spec mismatches. Same problem: we said one thing, they heard another, nobody caught it until the part was in hand.
I went back to my spreadsheet and rebuilt it. Not just unit price. Everything.
What I found:
Over three years, the unit cost difference between Vendor A and Samtec was about $9,300. If I remember correctly, the exact figure was $9,240, though I might be misremembering. But hidden costs—setup fees, expedited shipping when 5-week lead times became 6, engineering hours resolving spec mismatches—added up to roughly $6,800.
Net difference: $2,500. Over three years. On a $540,000 total spend.
That's 0.46%.
And that's before factoring in my own time.
I went with Samtec QSH.
Then I immediately second-guessed myself. What if the higher unit price became a problem in a budget review? What if the board rejected the PO because "you overpaid"? What if I'd just fallen for good salesmanship?
The two weeks between placing the order and receiving the first production run were stressful. I checked the tracking number daily. I reread the datasheet three times. I may have refreshed their shipment portal more than once.
Delivery arrived on day nine. Correct spec. Zero defects in the first 500 units.
That's when I relaxed. Mostly.
First: my TCO spreadsheet was a joke before this project. One tab. Now it has four. The difference isn't the tool—it's the discipline of asking "what else could go wrong?" for every line item.
Second: the communication failure with Vendor A wasn't their fault or mine. It was a system problem. We had no shared vocabulary for specs. Samtec sent their own spec sheet with our requirements pre-filled and asked us to confirm. That's a better system.
Third, and this is the part that still bugs me: I almost made the decision on unit price alone. If that sample hadn't arrived with the wrong orientation, I might have signed with Vendor A and spent the next year dealing with setup fees and spec revisions.
I still kick myself for not building that verification column earlier. If I had, I'd have caught the mismatch before wasting three weeks.
Luck isn't a procurement strategy either. But sometimes it's what saves you from a bad one.
Here's the honest part: Samtec QSH is not the right connector for every application.
If you're building a low-speed, cost-sensitive product—a consumer gadget with a two-year lifecycle—the unit price premium doesn't make sense. Vendor B at $4.90 or Vendor A at $4.20 will do the job. Save your budget for the components that matter.
If your production volume is above 50,000 units annually, the unit price gap becomes dominant. At that scale, engineering support savings get diluted. You're better off negotiating hard on volume pricing with a domestic distributor.
And if your team already has a deep relationship with a connector vendor who answers technical questions within hours, don't switch just because a spreadsheet says you should. That relationship has value that doesn't show up in any TCO model.
But if you're like us—mid-volume, engineering-constrained, tired of spec mismatches eating your timeline—Samtec QSH was worth the premium. At least, that's been my experience with production runs under 5,000 units a year. For our situation.
That's the thing about "best." It's not an absolute. It's relative to your constraints, your team, and your tolerance for the kind of hidden costs that never show up in the quote.
My spreadsheet still highlights the cheapest option in green. I just don't trust it anymore.