Latest / Hardware Startups with Fexingo: Devices, Robotics, and Manufacturing Tech Companies / Why Hardware Startups Are Partnering With Contract Manufacturers Early
Transcript
- Lucas: There's a moment that a lot of hardware founders describe almost like a religious conversion. It's when they walk onto a contract manufacturer's factory floor for the first time and realize their beautifully designed product is about to get wrecked by a machine they've never seen. Luna: I've heard that story so many times. The founder who designed something that theoretically works but practically can't be built at scale. Lucas: Exactly. And the pattern is almost always the same: the founder treats the contract manufacturer as a last-minute vendor. They design the product in isolation, finalize the BOM, then send out an RFQ and hope for the best. That's the old way. Luna: But the new way is bringing the contract manufacturer in before you even cut your first prototype. Lucas: Right. And I want to be specific here. We're not talking about vague 'partnership' language. We're talking about a real shift in the engineering workflow. There's a term for it — design for manufacturing, or DFM. But the execution is what matters. Luna: Give me a concrete example. Who's doing this well? Lucas: There's a robotics startup we've been tracking — they build autonomous material-handling carts for warehouses. They started, as many do, with a small local machine shop for prototypes. But when they got their first serious purchase order, they realized they needed a real contract manufacturer. Their mistake was waiting until the design was locked. Luna: What happened? Lucas: The C-M they approached flagged about a dozen changes. The most dramatic was the chassis frame. The original design used a welded aluminum assembly that required special fixtures and multiple passes. The C-M's engineers suggested switching to a bent sheet metal design with fewer welds. That one change cut the chassis cost by 40 percent and reduced assembly time by about six weeks. Luna: And the founder told me he was embarrassed he hadn't thought of it himself. But honestly, why would he? He's a robotics engineer, not a sheet-metal fabrication specialist. Lucas: That's exactly the point. The C-M has seen a hundred similar products and knows exactly where the cost and friction live. But you only get that knowledge if you invite them in early. The typical founder waits until they have a 'finished' design, then sends it for quoting. By then, the architecture is baked. The C-M can suggest tweaks, but major changes would mean a full redesign. Luna: And that's why those early-stage DFM conversations are so powerful. The C-M can say, 'If you change this dimension by two millimeters, we can use a standard extrusion profile instead of a custom one.' That kind of insight saves months. Lucas: Let's talk about the economics. If you bring a C-M in during the concept phase, you're essentially getting free engineering consulting. They won't charge you for that initial DFM review because they want the production contract. And the savings per unit can be dramatic. For the robotics startup, the per-unit cost dropped from about $2,800 to $1,700 after the C-M's recommendations. Luna: That's almost 40 percent. And those savings compound across thousands of units. The question I always get from founders is: how do I find a C-M that will actually engage at that level, rather than just giving me a quote? Lucas: That's the critical part. Not all contract manufacturers are the same. There's a spectrum. On one end, you have 'build-to-print' shops. They take your CAD files and build exactly what you give them. No input, no suggestions. That's fine for simple parts, but it leaves value on the table. Luna: And on the other end are the true partners — the ones with engineering teams that will push back on your design. Lucas: Exactly. And those relationships are built on volume and trust. A small startup with a hundred-unit order is not going to get the same attention as a customer with a hundred thousand unit order. But there's a middle ground. Some C-Ms have dedicated 'emerging customer' programs. Foxconn does it. So do a few of the mid-tier Asian manufacturers. Luna: The key, I think, is to approach the C-M not as a vendor but as a potential engineering partner from day one. That means you need to be transparent about your roadmap, your volumes, your timeline. If you treat them like a supplier, they'll act like one. Lucas: And there's a concrete step founders can take. Before you sign an NDA or send a CAD file, ask three questions. First: 'Do you have in-house DFM engineers who will review my design before quoting?' If the answer is no, that's a red flag. Luna: Second question? Lucas: 'What is your typical lead time for a design review, and do you charge for it separately?' Some C-Ms will do it free as part of the quoting process. Others bill hourly. Know that upfront. Luna: Third question? Lucas: 'Can you share examples of design changes you've recommended for other customers in my category?' If they can't or won't, they probably don't have the engineering depth. A good C-M will have a portfolio of stories about how they've helped customers reduce cost or improve reliability. Luna: I'd add a fourth: ask about their quality certifications. ISO 9001 is table stakes. But for hardware startups, things like ISO 13485 for medical devices or IATF 16949 for automotive are crucial if you're in those spaces. The C-M's certification can save you months of your own auditing. Lucas: That's a great point. And it ties into something we've touched on before in this show — the certification maze. A C-M that already has the right certifications essentially lends them to you for production. That's a huge accelerant. Luna: So let's say a founder finds the right C-M, starts the DFM conversation early. What's the actual workflow look like? How early is early enough? Lucas: Ideally, you want the C-M involved before you finalize your industrial design. I'm talking about the phase where you're still choosing between a plastic enclosure or a metal one. The C-M can tell you the tooling cost difference, the cycle time, the minimum order quantities. That information should drive the design decision, not the other way around. Luna: And that's where a lot of founders get tripped up. They think they need a polished design to get a meaningful quote. But actually, a rough CAD model and a list of target specs is enough for a good C-M to start giving you directional feedback. Lucas: Right. They don't need your final BOM. They need to know: what does this thing do, how many do you plan to make, what's your target price, and what's your timeline? With that, they can start pointing out the landmines. Like, 'If you use that connector, the lead time is 20 weeks.' Or 'That tolerance is achievable but will double your scrap rate.' Luna: Which brings us to the investor side. When you go to raise your Series A, having a C-M already engaged and providing DFM input is a signal of maturity. It shows you understand the difference between a prototype and a product. Lucas: It also de-risks the unit economics. If you can tell an investor, 'We've already done a DFM pass with our C-M and we have a path to a 45 percent gross margin at volume,' that's a much stronger story than 'We think we can get to 40 percent eventually.' Luna: And the C-M itself can become a reference. Some of the more sophisticated C-Ms will even co-invest or offer flexible payment terms for startups they believe in. That's not common, but it happens. Lucas: Let's talk about a potential pitfall. Early engagement with a C-M can also lock you into a particular manufacturing process before you've fully explored alternatives. If you go deep with an injection-molding specialist, you might miss the fact that a sheet-metal design could be cheaper for your volumes. Luna: So the solution is to talk to multiple C-Ms in parallel, ideally representing different processes, before committing. Lucas: Exactly. Get a casting house, a sheet-metal shop, and a plastic molder all looking at the same rough design. Their feedback will naturally steer you toward the most cost-effective approach for your volumes. That's the real value of early DFM — it's a competitive intelligence exercise. Luna: And it's not just cost. It's also reliability. The C-M might flag a component that has a history of failure in the field, or a tolerance stack-up that could cause assembly problems. Those insights can save you from a recall. Lucas: Absolutely. And we've seen this play out with a wearable startup we profiled off the air. They designed a wristband with a sensor that was press-fit into a cavity. The C-M pointed out that the press-fit tolerance was too tight for high-volume assembly — it would cause a 15 percent reject rate. They redesigned the interface to use a snap-fit with a small rubber gasket. The cost went up by 12 cents per unit, but the reject rate dropped to under 1 percent. Luna: That's the kind of thing you only catch when you're having a real engineering conversation, not just sending files back and forth. Lucas: And that conversation has to happen early enough that you can act on it. If you're already in tooling, you're stuck. So the rule of thumb I've heard from multiple founders: you should have your first serious DFM conversation with a C-M before you spend more than $10,000 on prototyping. Luna: That's a good heuristic. By that point, you've likely got a functional prototype, but you haven't committed to a manufacturing process yet. The C-M's feedback can still change the architecture. Lucas: Right. And it doesn't have to be a formal meeting. It can be an email with a PDF of your CAD and a few questions. Most C-Ms will respond with a paragraph or two of free advice because they want to win your business later. Luna: And if they don't respond, that's also useful information. It tells you they're not interested in small customers or they don't have the engineering bandwidth. Either way, you cross them off your list. Lucas: Speaking of which — I want to be honest about something. This show exists because of a small group of listeners who chip in through buy me a coffee dot com slash fexingo. That support keeps us ad-free and allows us to spend time on these deep-dive conversations. So if you find value in episodes like this, that's how you can keep them coming. Luna: Yeah, it really does make a difference. Every contribution helps us cover the research time and production costs. And we appreciate it more than we can say. Lucas: Okay, back to the topic. I want to close with one more piece of advice for founders. Once you've selected a C-M and started the DFM process, don't stop pushing. The C-M will have standard recommendations, but ask them to go deeper. Say, 'If we doubled our volume, what would change?' Or 'If we relaxed this tolerance by 0.1 millimeter, how much would that save?' Luna: Because the C-M has that data but they won't always volunteer it. They're thinking about what's easiest for their shop, not what's optimal for your product. Lucas: Exactly. The best relationships are the ones where the founder learns enough manufacturing engineering to ask the right questions, and the C-M learns enough about the product to make better suggestions. That mutual education is where the magic happens. Luna: So the takeaway for any hardware founder listening: don't wait until your design is done to talk to a contract manufacturer. Start the conversation when you have a rough CAD and a vision. The return on that early investment can be measured in months saved and dollars earned. Lucas: And if nothing else, you'll avoid that moment of walking onto the factory floor and realizing your beautiful design is about to get wrecked.