Latest / Hardware Startups with Fexingo: Devices, Robotics, and Manufacturing Tech Companies / How Hardware Startups Use Value Engineering to Cut Costs
Transcript
- Lucas: If you're building a hardware startup right now, you're probably watching your bill of materials climb. Component lead times have eased since the pandemic peak, but prices on many specialty ICs are still elevated. So how do you bring costs down without redesigning everything from scratch? Luna: That's the question every hardware founder asks when they hit volume production. Is there a systematic way to do it? Lucas: There is, and it's called value engineering. It's not just swapping out a cheaper resistor. It's a structured process where you analyze every line item in your BOM and ask: does this component deliver more performance than the product actually needs? Can we get the same functionality from a standard part instead of a custom one? Luna: So it's about matching capability to requirement. Not just cutting cost, but making sure you're not paying for headroom you'll never use. Lucas: Exactly. Let me give you a concrete example. I was talking to the team at a smart home hub startup. Their first prototype had a BOM cost of around $87 per unit. That's for a device that retails for $129. After they sell through distribution, that's a razor-thin margin. They needed to get the BOM under $65 to make the business work. Lucas: So they did a value engineering pass. First they ranked every component by cost. The top three items were a custom Wi-Fi module, a specialized audio codec, and a power management IC they'd designed in-house. Those three accounted for 42 percent of the total BOM. Luna: The custom parts are usually the biggest cost drivers. What did they do about them? Lucas: They took a hard look at the Wi-Fi module. The custom one had a Bluetooth LE co-processor with extra GPIOs they never used. They found a commercial module from a major supplier that had the same footprint, slightly less flash storage, but all the necessary certification. Cost: from $12.50 to $6.80 per unit. That's almost fifty percent savings. Luna: And the testing on that? Did they have to redo any certification? Lucas: They did need to re-run FCC and CE pre-compliance because the module's RF characteristics were different. But the module vendor provided pre-certified reference designs, so it took just two weeks instead of the typical six. The audio codec was easier. They realized the hub's speaker was only used for basic chimes, not high-fidelity audio. So they swapped the dedicated codec for a PWM output from the main MCU with a simple filter circuit. That saved $2.10 per unit. Lucas: The interesting one was the power management IC. They'd designed a custom PMIC because their first engineer thought it gave better battery life. But the actual battery was only used as a backup during power outages. A standard off-the-shelf PMIC with a simple charging circuit performed within spec, and the cost dropped from $8.40 to $3.20. Luna: Those three changes alone brought the BOM down significantly. But I imagine they had to re-validate the whole system to make sure nothing broke. Lucas: Right, and that's the critical part of value engineering. You can't just swap components and ship. They put the redesigned boards through a full battery of functional tests: 1,000 hour burn-in, temperature cycling from -20 to 60 Celsius, and drop tests. The new design actually passed with fewer failures than the original because the commercial parts were more mature. Luna: It's counterintuitive. The cheaper solution turned out to be more reliable because it used parts that had been mass-manufactured for years. Lucas: Exactly. Their total BOM ended up at $62.30. That's a 29 percent reduction — $24.70 per unit. At their initial volume of 15,000 units, that's over $370,000 in direct savings per batch. And they didn't sacrifice any features that mattered to customers. Lucas: If this kind of practical insight is useful for what you're building or running, the way we keep this show ad-free is listener support. You can buy me a coffee at buy me a coffee dot com slash fexingo. Luna: Yeah, it's a small gesture that lets us keep digging into these stories without sponsors. And we genuinely love hearing what you're working on. Lucas: So back to the value engineering process. After the initial cost reduction, they looked at the remaining line items. They found they were using a more expensive usb c connector than needed — it was rated for 100W power delivery, but the hub only drew 5W. They switched to a 15W rated connector, saving another $0.18 per unit. It adds up. Luna: Those kinds of over-specifications are really common when startups rush to prototype. They pick a component that's 'good enough' from their bench stock, then never revisit it. Lucas: That's exactly why value engineering works best as a formal milestone, not a one-off cost-cutting scramble. You do it once after you've stabilized the design and before you lock in the tooling. The startup I mentioned did it right after they hit 500 pre-production units. At that point, they had enough data to know what the product actually needed. Luna: When do you think it's not appropriate to do value engineering? I can imagine a case where you're trying to hit a very aggressive price point and you might over-optimize and lose quality. Lucas: Absolutely. The trap is focusing solely on unit cost without considering total cost of ownership. If you put in a cheaper component that fails twice as often, your warranty costs explode. I've seen startups swap a $0.50 regulator for a $0.20 one, then deal with a 5% field failure rate. That wipes out all the savings. Lucas: You need to run reliability testing on the proposed substitutes. The smart hub team had a rule: any component change had to survive at least 1,000 hours of accelerated life test at 85°C before approval. That kept them honest. Luna: And that's a good discipline for any hardware startup. Because the real cost is not the component price, it's the cost of a recall or bad reviews. Lucas: Right. So value engineering is powerful when done right. A few takeaways for anyone listening: first, rank your BOM by cost early — Pareto analysis usually shows 20% of components drive 80% of cost. Second, question every custom part. Can a standard part do the job? Third, involve your contract manufacturer in the review — they often know cheaper equivalents that still meet specs. And fourth, always test the revised design as thoroughly as the original. Luna: And one more thing: don't wait until you're in high-volume production. Do a value engineering pass during the prototype phase or after the first pilot run. The later you leave it, the more expensive changes become because of tooling and certification re-runs. Lucas: Great point. The startup we talked about locked in their final BOM at around 500 units. That gave them confidence to order the injection molds and commit to a $200,000 tooling investment. They knew the cost per unit was viable. Luna: That must have been a huge relief for the founders. They could go to investors with a clear unit economics story. Lucas: Exactly. And that's the hidden benefit of value engineering: it forces you to understand your product deeply. You can't just outsource it to the manufacturing team. The founders were in the lab testing each swap, reviewing the data. It made them better hardware engineers in the process. Luna: I think that's a good note to end on. Have you seen any other hardware startups recently that used a similar approach with different products? Lucas: Yes, I've seen it in robotics, medical devices, even consumer electronics. The principles are the same. The next time you're looking at your BOM, ask yourself: is every dollar earning its keep? That's value engineering in a nutshell. Luna: Thanks for joining us on Hardware Startups with Fexingo. We'll be back with another episode soon. Lucas: If you want to support the show, remember you can buy me a coffee at buy me a coffee dot com slash fexingo. And as always, keep building.