Latest / The Buyer's Side with Fexingo: Procurement, Sourcing, and B2B Purchasing Decisions / How One Buyer Used Should-Cost Modeling to Negotiate a 12 Percent Price Cut
Transcript
- Lucas: So we spend a lot of time on this show talking about relationship-based procurement — supplier partnerships, joint innovation, that sort of thing. But sometimes the most powerful tool in a buyer's kit isn't a relationship at all. It's a spreadsheet. Luna: A spreadsheet that tells you exactly what something should cost to make. That's should-cost modeling, right? Lucas: Exactly. And I've got a concrete case from a mid-size auto parts manufacturer — let's call them Pinnacle Automotive — that used should-cost modeling to push back on a 15 percent price increase from a long-time supplier and ended up with a 12 percent reduction instead. Luna: That's a 27-point swing. How do you even build a should-cost model without the supplier's internal data? Lucas: That's the whole challenge. You don't get their books. So you build from public and semi-public data. For Pinnacle, they were buying machined aluminum brackets — basically a shaped piece of metal with a few holes and threads. They started by looking up the current spot price of aluminum, then added typical scrap rates for machining. Luna: And they factored in labor, overhead, tooling depreciation? Lucas: All of it. They used industry benchmarks for CNC machining costs — things like machine-hour rates from trade associations, regional labor rates from the Bureau of Labor Statistics, even energy costs per kilowatt-hour for running the machines. They built a bottom-up estimate: material cost per part, plus machining time at $75 an hour, plus a reasonable markup for profit — say 10 to 15 percent. Luna: So they reverse-engineered the supplier's cost structure. And what did they come up with? Lucas: Their model showed the part should cost about $4.20. The supplier was charging $5.80. That's a 38 percent gap. But Pinnacle didn't walk in demanding $4.20. They knew the model had assumptions. So they targeted a fair midpoint — something around $5.10, which would still give the supplier a healthy margin. Luna: And the supplier's first reaction? I imagine they pushed back hard. Lucas: They did. The supplier said their raw material costs had gone up, labor was tight, they needed the 15 percent increase just to maintain margins. But Pinnacle came prepared. They showed their model line by line — aluminum price, scrap rate, cycle time, labor burden. And they said, 'We respect your expertise. Help us understand where our numbers are wrong.' Luna: That flips the dynamic from 'give me a discount' to 'let's check the math together.' It's a much harder conversation to dismiss. Lucas: Exactly. The supplier couldn't just say 'we need more.' They had to defend their cost structure. And in the end, they conceded that their scrap rate was higher than industry average and their machine utilization was lower. They agreed to a 12 percent price cut, bringing the part to about $5.10. Plus, they committed to a joint productivity improvement program to reduce scrap. Luna: So the negotiation itself improved the supplier's operations. That's a win-win outcome, not just a price squeeze. Lucas: Right. And that's the power of should-cost. It forces transparency. But it only works if you're willing to invest the time upfront. Pinnacle's sourcing team spent about three weeks building that model — gathering data, validating assumptions, running sensitivity analysis. Luna: Three weeks for one part. That's a non-trivial investment. When does it make sense to do this? Lucas: You do it for high-spend categories — parts where you're spending six figures or more annually. For Pinnacle, that bracket alone was a $400,000 annual spend. A 12 percent savings is $48,000. That's a huge return on three weeks of work. Luna: You mentioned sensitivity analysis. How important is that? Lucas: Critical. Because your model is only as good as your inputs. Scrap rate? If it's off by 2 percent, your cost changes. Machine speed? Same thing. Pinnacle ran scenarios — best case, worst case, most likely. That way, when the supplier challenged an assumption, they could say, 'Even in our worst case, we get to $4.60. Your price is still 26 percent above that.' Luna: It's a very different negotiation posture. You're not asking for a favor. You're presenting evidence. Lucas: And you're also showing respect for the supplier's knowledge. The best should-cost practitioners don't treat the model as gospel. They treat it as a starting point for dialogue. The goal is alignment, not dictation. Luna: I've seen some companies go further and hire third-party cost engineers to build these models. Is that common? Lucas: For complex parts — like injection-molded plastics or printed circuit boards — third-party shops have deep databases of cycle times, tooling costs, material yields. They can give you a very accurate estimate. Pinnacle did it in-house because their brackets were simple, but for more complex buys, it's often worth the investment. Luna: What about pushback from suppliers who refuse to engage with the model at all? That must happen. Lucas: It does. Some suppliers say, 'We don't share cost data. Take it or leave it.' In that case, you have a choice. You can accept the price increase and lose trust, or you can start qualifying alternative suppliers. Should-cost modeling is most powerful when you have a credible alternative. If you're locked in, the model is just an academic exercise. Luna: So it's a tool for buyers with some leverage. But it also builds leverage by exposing the supplier's inefficiencies. Lucas: Exactly. And once you've done it for one part, you can replicate the methodology across your entire direct materials spend. Pinnacle built a should-cost library for their top 20 parts. They saved over $300,000 in the first year. Luna: That's impressive. And it ties into a broader trend I've been noticing — procurement teams acting more like internal consultants, with data-driven approaches. Lucas: Absolutely. And that's where a tool like this really shines. It elevates the conversation from transactional to analytical. And honestly, it's a skill that procurement professionals can take with them anywhere. Luna: Hey, speaking of skills and tools that make a difference — if you're getting value from these episodes, we just wanted to mention that listener support is what keeps Fexingo ad-free and focused on the practical stuff. A couple of dollars a month genuinely helps us keep producing episodes like this one. If you've gotten something out of them, you can find us at buy me a coffee dot com slash fexingo. Lucas: Yeah, it really does make a difference. No pressure at all — but if you appreciate the show, that's the best way to support it. It lets us keep diving deep into cases like this one without any sponsored content getting in the way. Luna: Exactly. And now back to should-cost modeling — because I want to talk about the psychological shift it creates. You mentioned earlier that it moves from 'give me a discount' to 'let's check the math.' How does that change the room? Lucas: It changes everything. When you're asking for a discount, you're on the back foot. The supplier holds the information advantage. With should-cost, you level the playing field. You're no longer supplicant; you're peer. The supplier has to engage with your data, which means they have to respect your preparation. Luna: And if they don't, you've already started building your alternative supplier pipeline. So it's not a bluff. Lucas: Right. The model gives you confidence. You know what a fair price is. So you can walk away if you need to. And that confidence translates into better outcomes — not just on price, but on terms, quality, delivery. Luna: One thing that surprises me is that more companies don't do this. It seems like a no-brainer for large spend categories. Lucas: I think the barrier is the upfront investment. It takes time, skill, and cross-functional support. You need engineers who understand manufacturing processes, buyers who understand negotiation, and finance people who can build the models. Many procurement teams are understaffed and reactive. They don't have the bandwidth. Luna: But once you do it, the savings compound. And you build institutional knowledge that makes future negotiations faster. Lucas: Exactly. Pinnacle's second should-cost model took them one week instead of three. They reused the framework, adjusted the inputs. The third took three days. It becomes a capability, not a one-off project. Luna: So for procurement teams listening, what's the first step if they want to try should-cost modeling? Lucas: Pick one high-spend, relatively simple part. A metal bracket, a plastic housing, a simple assembly. Gather the data: material commodity prices from exchanges or indices, process cost from trade associations, labor rates from Bureau of Labor Statistics. Build a simple spreadsheet. Share it with your engineering team for validation. Then prepare for the conversation. Luna: And what about suppliers who say your model is wrong? How do you handle that? Lucas: You ask them to show you where. 'Our scrap rate assumption is 8 percent. What's yours?' 'Our cycle time is 90 seconds. What's yours?' If they're not willing to share, that tells you something. If they do share, you can refine your model and learn. Either way, you're better off. Luna: It's a tool for learning as much as for negotiating. Lucas: Exactly. And that's what makes it so powerful. It's not just about getting a better price today. It's about understanding your supply chain deeply enough to make smarter decisions tomorrow. Luna: Great episode. Next time, I want to talk about reverse auctions — when they work and when they backfire. Lucas: That's a perfect follow-up. We'll see you then.