Latest / The Buyer's Side with Fexingo: Procurement, Sourcing, and B2B Purchasing Decisions / How One Buyer Used Digital Twin Modeling to Cut Prototype Spend
Transcript
- Lucas: You know that moment in a product development cycle when the buyer has to sign off on a prototype order — and it's basically a blank check because the supplier knows you're committed? Luna: I've heard that story a dozen times. 'We need ten thousand units for testing, and we need them in three weeks.' And the price comes back at double what the production unit should cost. Lucas: Exactly. That's the prototype premium. And it's not just the price — it's the sunk cost. Once you've paid for the mold or the tooling, you're locked into that supplier, that design, that timeline. But there's a way to break that cycle, and it involves building the product before it exists physically. Luna: You're talking about digital twins. Lucas: Digital twins. And not the kind you see in flashy Industry 4.0 presentations. I'm talking about a very practical application: using a high-fidelity simulation to run prototype iterations in software before you ever cut metal or shoot plastic. The buyer I want to tell you about is a senior procurement manager at a mid-sized industrial equipment company — let's call it Midwest Pumps, not the real name, but a real 2024 case. Luna: What were they buying? Lucas: A new pump assembly for an agricultural irrigation system. The pump had to handle abrasive sediment, operate at variable speeds, and fit into an existing housing. The engineering team had a design, but they knew the tolerances on the impeller and the volute would need multiple iterations. Traditionally, they'd go to a supplier — say, a precision injection molder — and order a few hundred prototype parts. Each iteration would cost around $40,000 in tooling modifications and material, and take four to six weeks. They were budgeting for at least four rounds. Luna: So $160,000 and four to six months just to get a working physical prototype. And that's before you even test it in the field. Lucas: Right. And the buyer, her name is Sarah, she had a different idea. She'd been reading about how automotive and aerospace companies use digital twins to validate designs, and she wondered if she could do something similar with her supply base. But she didn't have a million-dollar simulation lab. What she had was a CAD model and a copy of a cloud-based simulation tool that her company already licensed but barely used. Luna: So she built the digital twin herself? Lucas: She didn't build it alone. She partnered with the engineering lead and the supplier's application engineer. Together, they created a digital twin of the pump assembly that included the material properties — the exact polymer blend the supplier would use, the shrinkage rates, the thermal expansion. They ran the simulation for flow dynamics, stress points, and wear patterns. And here's the key: she invited three potential suppliers to contribute to the model. Not to bid on a spec, but to co-develop the virtual prototype. Luna: That's a different dynamic. Usually, you keep suppliers at arm's length during design because you don't want to tip your hand on volumes or pricing. Lucas: Exactly. But Sarah's logic was, 'If I can get them to help me validate the design virtually, I'm not sharing proprietary pricing — I'm sharing a model that makes their job easier when it's time to manufacture.' And it worked. The three suppliers each ran their own simulations and came back with suggestions: one recommended a different gate location on the mold to reduce warpage, another suggested a slight change in the polymer blend to improve flow, a third identified a potential cavitation issue at high RPM that the original design missed. Luna: So she got three rounds of engineering feedback for free, essentially. Lucas: Better than free — she got it in two weeks, not six. And when the digital twin was finalized, she had a virtual prototype that had already been through 12 iterations. The physical prototype they ordered on the first try worked. No revisions. The tooling was right, the tolerances were right, the part fit the housing on the first assembly. Luna: So what was the actual cost saving? You mentioned 35 percent on prototype spend. Lucas: Sarah tracked it. The original budget for prototype tooling and parts was $400,000. She spent $260,000. The saving came from three things: no rework tooling charges, fewer physical iterations, and faster time to market — which the CFO valued at roughly $50,000 in avoided delay costs. So total saving was about $190,000 direct, plus the $50,000 indirect. And the product launched three months ahead of the original schedule. Luna: Three months early in agricultural equipment — that's a whole selling season. Lucas: Exactly. The product hit the market in time for spring planting. The revenue impact was substantial. But what I find most interesting is the negotiation leverage it created. When Sarah went to award the production contract, she had three suppliers who had all invested engineering time in the digital twin. They knew the design intimately. They were essentially locked into competing for the business, but they also had lower switching costs because the design was already validated. So she got a 12 percent lower unit price than the original target cost. Luna: How did she sell this to the CFO? I mean, digital twin software costs money, and the engineering time isn't free. Lucas: She framed it as a capital efficiency play. She said, 'We're going to spend $20,000 on software licenses and maybe 40 hours of engineering time. The alternative is $400,000 in prototype spend and a six-month delay if anything goes wrong. The ROI on the digital twin is ten to one in the worst case.' The CFO bought it because she had numbers from a pilot project in a different division that had tried a similar approach on a smaller part. Luna: So this wasn't her first rodeo. She had proof of concept. Lucas: She did. She started with a simple bracket — low risk, low complexity. That pilot saved 20 percent on prototype costs and took two weeks off the timeline. Once she had that data, she could argue for scaling it to the pump assembly. That's a classic procurement move: start small, prove the model, then go big. Luna: I've heard procurement pros say digital twins are only for big companies with dedicated simulation teams. But Sarah was at a mid-sized firm — maybe 500 employees, $200 million in revenue. She didn't have a dedicated team. Lucas: That's the myth she busted. The cloud-based simulation tools are getting cheaper and easier to use. She used a tool that cost $500 per month per user. The learning curve was about a week for the engineer. The key wasn't the technology — it was her willingness to bring suppliers into the model early. Most buyers are afraid that early collaboration means losing control of the design. But Sarah found that control comes from the data, not from secrecy. Luna: What about intellectual property concerns? If you share your CAD model with three suppliers, what's to stop them from using that design for a competitor? Lucas: She handled that with a simple joint development agreement. Each supplier signed a non-disclosure that specifically covered the digital twin data. And she only shared the model after stripping out any proprietary performance specs that weren't necessary for the simulation. The suppliers only saw the geometry and material properties they needed to run their simulations. No secret sauce on the control algorithms or the system integration. Luna: So the digital twin became a shared artifact that aligned everyone around the same goal, rather than a weapon for one side. Lucas: That's a perfect way to put it. And it changed the conversation from 'What's your price per part?' to 'How do we make this design manufacturable at the lowest total cost?' The suppliers started suggesting cost-saving ideas because they wanted to win the production business. One supplier suggested a change that eliminated a secondary machining operation — that alone saved $3 per unit. Luna: And Sarah got credit for that cost reduction, even though it was the supplier's idea. Lucas: Right. Because she created the environment where the supplier felt invested in the design. That's the trust-based sourcing we've talked about in previous episodes, but here it's enabled by a specific tool. Luna: You know, one of the things I appreciate about this show is that we don't run ads. It's just us digging into real procurement stories. If these conversations have helped you think differently about your own supplier relationships, there's a simple way to support that choice. It's buy me a coffee dot com slash fexingo. No pressure, no perks, just knowing that people who find value here help keep it ad-free. Lucas: Yeah. And listeners like Sarah's story — they're exactly why we do this. So back to the digital twin: what's the one thing a buyer should take away from this episode if they want to try it? Luna: Start with a simple part. Not the high-risk, high-complexity assembly. Pick a bracket, a housing, a cover — something with known specifications and a clear cost baseline. Run the digital twin with one supplier you trust. Measure the results. Then scale. Lucas: And don't underestimate the power of a joint development agreement. It's a legal document, but it's also a relationship tool. When you and the supplier sign it, you're both saying, 'We're in this together.' That shifts the dynamic from adversarial to collaborative. Luna: Sarah's case also shows that procurement can be a driver of innovation, not just a cost center. She didn't wait for engineering to come to her with a finished design. She went to them and said, 'Let's try something new.' Lucas: Exactly. And that's the kind of proactive sourcing that gets you a seat at the strategy table. Next episode, we'll look at how one buyer used a supplier's factory floor data to renegotiate a long-term contract. Thanks for listening.