Latest / The Buyer's Side with Fexingo: Procurement, Sourcing, and B2B Purchasing Decisions / How One Buyer Used Multi-Sourcing to Secure Chip Supply
Transcript
- Lucas: Back in early 2025, a European automotive OEM — one of the top ten globally — made a quiet but significant shift in how they bought microcontrollers. For years, they had relied on a single fab, a single supplier, for a critical class of 32-bit MCUs used in engine control units. One fab. One source. And when that fab had a power outage in late 2024, it shut their assembly line for eleven days. Luna: Eleven days of downtime. At a modern auto plant, that's easily tens of millions in lost revenue, not to mention the reputational hit with dealers. Lucas: Exactly. So the procurement team — led by a director named Sarah Kline — got approval to restructure the sourcing model. Instead of single-source, they would multi-source that MCU across three different fabs, in three different countries. Not just three suppliers, but three distinct manufacturing sites, each with independent power grids, water supplies, and logistics chains. Luna: Right. The classic resilience play. But the instinct of most buyers I talk to is that multi-sourcing costs more. You lose economies of scale, you split volumes, you increase qualification and testing overhead. Did Sarah's team quantify that trade-off? Lucas: They did, and that's the real story here. The team built a probabilistic cost model that treated disruption not as a binary event but as a frequency and severity distribution. They used historical blackout data, geopolitical risk scores from a third-party provider, and even weather patterns for each fab location. The model told them that under single-source, the expected annual disruption cost was about 4.2 percent of the total cost of the MCU category. Luna: So they could pay up to 4.2 percent more and still break even on a risk-adjusted basis. Lucas: Right. But here's the twist — they actually accepted a unit-cost premium of 7 percent. That's above the break-even threshold. Why? Two reasons. First, the model's confidence interval was wide; the 4.2 percent was a median estimate, but the tail risk was much worse. A single catastrophic event — like a fire or a prolonged factory shutdown — could cost 15 to 20 percent of category spend in a single year. They were buying insurance against the tail. Luna: And the second reason? Lucas: Strategic flexibility. By having qualification already done at two backup fabs, they could rapidly shift volume if the primary supplier had allocation issues. In the chip market of 2024 and 2025, allocation was the bigger headache than outright shutdowns. Suppliers were rationing capacity, and single-source buyers often got shorted because they lacked leverage. Multi-sourcing gave Sarah's team the ability to say, 'If you can't give us our full allocation, we'll move 20 percent to your competitor.' That alone made the 7 percent premium worth it. Luna: And those kinds of conversations, honestly, are what keep this show going — real procurement decisions that move the needle. If these episodes have helped you think differently about your own sourcing challenges, it's because of listeners like you who support the show. We keep this podcast ad-free, and listener contributions are what make that possible. If you find value in the Buyer's Side, consider buying us a coffee — buy me a coffee dot com slash fexingo. No pressure, just a small way to keep the conversation going. Now, back to Sarah's multi-sourcing model. Lucas: So the implementation wasn't straightforward. They had to re-qualify the MCU at two new fabs — that's a six-month process involving electrical testing, thermal cycling, and certification by regulators. That cost roughly one point two million dollars in engineering time and testing fees. Sarah's team had to decide: absorb that as a one-time cost or amortize it into the unit price? Luna: Which did they choose? Lucas: They amortized. They negotiated a three-year agreement where the supplier covered 40 percent of the qualification cost in exchange for a guaranteed volume floor. That way, the OEM didn't have to write a big check upfront, and the supplier got a predictable revenue stream. It's a good example of risk-sharing in supplier relationships. Luna: So the premium wasn't just a straight markup. Part of it was the cost of redundancy, part was the cost of qualification, and part was the price of optionality. Lucas: Precisely. And that's the key conceptual shift. Traditional procurement thinking treats multi-sourcing as a pure cost increase. But Sarah's team reframed it as buying an option — the option to redirect volume, the option to negotiate from strength, the option to survive a black-swan event. When you think of it as an option, you can apply option-pricing logic to figure out what premium is worth paying. Luna: That's a really powerful framing. I've seen companies pay much more than 7 percent for financial options on commodities. Why not treat operational flexibility the same way? Lucas: Exactly. And the result? In the twelve months after implementation, the primary fab had a two-week force majeure event due to a typhoon. The OEM shifted 30 percent of volume to a backup fab within a week. No assembly line stoppage. The 7 percent premium paid for itself in that single event — they avoided a disruption that would have cost them roughly 8 percent of category spend. Luna: So the ROI was immediate. But I'm curious about the downsides. Multi-sourcing can dilute buyer leverage. If you're splitting volumes across three suppliers, none of them sees you as their top customer. You might lose priority during allocation crunches. Lucas: That's a real risk. Sarah's team mitigated it by designating one supplier as the 'lead' with 50 percent volume, and the other two at 25 percent each. The lead supplier still had the largest share, so they had an incentive to prioritize the OEM. And because the OEM had proven they would actually shift volume — the typhoon event demonstrated that — the lead supplier knew they couldn't take the relationship for granted. Luna: So the multi-sourcing structure itself creates a credible threat that strengthens your hand with the primary supplier. It's not just about redundancy; it's about bargaining power. Lucas: Exactly. And that's a lesson that applies beyond automotive chips. Any buyer in a constrained market — medical devices, aerospace, industrial electronics — can use this playbook. The key is to model the disruption probability honestly, set a premium threshold based on tail risk, and structure the split to maintain leverage with your key supplier. Luna: One more thing — how did they handle inventory? If you have three sources, do you hold safety stock for each? Lucas: Good question. They actually reduced total inventory by about 15 percent. Because the supply chain was more reliable, they could lower safety stock levels. Before, they held eight weeks of buffer to cover single-source risks. After, they held five weeks — two weeks for each source, with some overlap. The inventory savings partially offset the unit-cost premium. Luna: So the net cost impact was smaller than the headline 7 percent. Maybe closer to 3 or 4 percent after factoring in inventory savings and reduced disruption risk. Lucas: Exactly right. Sarah's team calculated the net cost increase at 3.8 percent. And they considered that a small price for the resilience and negotiating leverage they gained. In their internal presentation to the CFO, they called it 'the cheapest insurance we've ever bought.' Luna: That's a great line. And I think it captures the mindset shift that procurement leaders need to make in this era of frequent supply shocks. Efficiency is still important, but resilience has become a competitive advantage.