I manage $180,000 in component spending annually for a 45-person manufacturing operation. I've signed purchase orders for ceramic capacitors, tantalum capacitors, relays, and connectors across 6 years and roughly 200 vendor quotes. This is the checklist I run before every one of them.
Six checkpoints. None of them take more than 20 minutes. All of them exist because I skipped one once and paid for it.
A broker listing "Kemet Electronics" on their website means absolutely nothing. Kemet has a formal authorized distributor network, and any seller outside it is gray market or counterfeit. In Q2 2024, a broker quoted us 22% below our regular channel price on $4,200 of capacitors. I asked for a certificate of authorization. They went silent. Then they came back with a "factory direct" story and a 15-day lead time. We passed.
The check: pull the distributor from Kemet's official partner list, then verify the lot number against the OEM's traceability portal. Ten minutes of work. Cheap insurance against a recall.
Kemet Electronics sits under Yageo Corporation (acquisition closed in 2020). Does that matter for your order? Yes — it affects supply terms, lead times, and sometimes minimum order quantities. A distributor quoting from a 2019 price sheet is quoting from a different company.
I keep a one-page ownership map for every brand on our BOM. When Yageo consolidated a product line, we caught a SKU re-pricing two months before our contract renewal. That single check saved us about $1,900 on the next order.
I almost chose a vendor quoting $0.22 per unit over one quoting $0.24. Almost. Then I ran the numbers.
The $0.22 vendor charged $180 setup, $75 per change order, and a 2% surcharge for card payment. The $0.24 vendor included setup and accepted standard ACH. By order 3, the "cheaper" quote was 17% more expensive. Same product. Same brand. Completely different total.
Look, price sheets lie by omission. Your TCO model needs: unit price, setup fees, change-order fees, freight terms, payment surcharges, MOQ penalties, and the cost of handling defective parts. (Mental note: revisit this whenever a vendor updates their fee schedule — they love adding line items.)
A ceramic capacitor is a ceramic capacitor, right? No. This is the mistake I made early on.
Even for something as specific as a platinum blood pressure monitor, the BOM calls for low-ESR MLCCs with tight voltage derating and a verified failure mode. A general-purpose MLCC from the same product family passes the datasheet check but fails field reliability. I've seen it happen. We used a generic 10μF/16V part once — same tolerance as the spec'd component. By month 3 of production, we had early-life failures. ESR drifted with temperature and ate the ripple margin. The datasheet didn't lie. The application changed.
Before approving a substitute, check the actual operating conditions: temperature range, ripple current, and — for medical or safety-critical applications — the consequence of failure. Efficiency doesn't help you if the product fails in the field.
This is the same calculus IT teams use when choosing between HPE vs Broadcom infrastructure. Both make solid products. But once you build around one ecosystem, switching costs get ugly. Components work the same way.
If one vendor holds more than 60–70% of your component spend, you have a concentration risk. Run the numbers on qualifying a second source: what does it cost, and how long does it take? If the answer is under 5% of annual spend and under 90 days, build the second source now.
Don't trust your memory. Set triggers that force a TCO re-evaluation: (a) 5 orders placed with the same vendor, (b) any corporate change at the OEM (merger, acquisition, capacity shift), or (c) a defect rate above 0.5% at QBR.
We started doing this after a 2023 audit revealed we'd been paying a 6% "expedited handling" fee on standard-lead-time orders for 11 months. Nobody caught it because nobody was looking. A single calendar reminder would have saved us $450.
It took me 3 years and roughly 150 orders to understand that "cheap" and "expensive" don't mean anything without a TCO model. The first time I caught a 17% hidden-cost gap, I was annoyed. The second time, I built this checklist so I wouldn't have to remember.
A few things to watch for:
That's it. Six checks. Maybe 40 minutes total. The alternative is finding out at audit time, and audit time is always more expensive.