Background
The client is a European AI hardware startup that had developed an innovative smart camera product powered by edge AI. They had a working prototype, had raised a seed round, and were ready to move to mass production. They had been talking to several factories in Shenzhen and Dongguan but had not yet selected one.
Their situation was typical of early-stage hardware companies: they had a great product design but limited manufacturing experience in China. They were worried about IP protection, quality control, tooling ownership, and the risk of getting locked into a single supplier. They wanted legal support that was practical and manufacturing-savvy — not just contract templates from a lawyer who had never set foot in a factory.
Key Challenges
IP protection before sharing designs. The client needed to share detailed design files with potential factories for quoting and DFM analysis, but they were concerned about IP leakage. They wanted to know how to protect themselves before they started disclosing their designs.
Tooling ownership and risk. They had heard horror stories about companies losing access to their tooling when factory relationships went bad. They wanted to make sure their tooling investment was protected — both legally and practically.
Factory selection risk. They were evaluating three factories with very different quotes and capabilities. The cheapest quote was 30% lower than the most expensive one, but they were not sure if the cheaper factory could deliver the quality they needed. They needed a framework for evaluating factories beyond just the price.
Quality control and acceptance standards. They knew they needed quality control but were not sure what kind, how much, or how to structure the acceptance criteria in their contracts. They did not want to overspend on inspections but also did not want to receive a shipment of defective products.
Supply chain contract structure. They needed a manufacturing agreement that covered all the important issues — pricing, payment terms, quality standards, delivery schedules, IP ownership, confidentiality, liability, and termination — but they did not know what was standard in Chinese manufacturing and what was negotiable.
Geason’s Approach
We structured the engagement around the NPI (New Product Introduction) timeline, providing legal support at each stage rather than handing over a static contract.
Phase 1: Pre-selection IP protection. Before the client shared any detailed designs with factories, we put in place a multi-layered IP protection system. We registered their trademarks in China (classes 9, 11, and 35), filed software copyright registrations for their embedded software, and advised on patent strategy (filing two design patents for the product’s distinctive visual features, plus one utility model for a structural innovation). We also drafted a strong mutual NDA specifically designed for manufacturing discussions — covering not just the designs themselves but also manufacturing processes, cost structures, and business information.
Phase 2: Factory evaluation and selection. We provided the client with a structured factory evaluation framework covering quality systems, production capability, financial stability, compliance track record, and IP protection practices. We accompanied them on factory visits to all three candidates, taking detailed notes and asking the questions that factory auditors typically ask. After the visits, we produced a comparative analysis with our recommendation — not just which factory was cheapest, but which offered the best overall value considering quality, capability, and risk.
Phase 3: Manufacturing agreement negotiation. We drafted a comprehensive manufacturing agreement tailored to the client’s product and situation. Key provisions included: clear IP ownership (client owns all tooling, designs, and improvements), tooling maintenance and storage responsibilities, detailed quality standards and acceptance procedures, delivery schedules with penalties for delays, confidentiality obligations that survive termination, and clear terms for ending the relationship (including the right to retrieve tooling and transfer to another factory). We negotiated the agreement with the factory’s legal representative, resolving several important points that the factory had initially pushed back on.
Phase 4: Tooling and NPI support. During the tooling and NPI phase, we provided ongoing support: reviewing the tooling acceptance procedure, advising on quality control protocols (pre-shipment inspections plus in-process checks at key stages), and helping structure the relationship with a third-party quality inspection company. We also reviewed all change orders and price adjustments to make sure they were documented properly.
Phase 5: Mass production preparation. Before mass production started, we conducted a compliance review of the entire manufacturing arrangement to make sure everything was in order. We verified that all tooling ownership documentation was complete, all quality standards were documented and agreed, all payment milestones were clear, and the client had a plan for what to do if quality or delivery issues arose.
The Non-Obvious Judgment
The most important non-obvious point in this case was about tooling ownership as an operational issue, not just a legal one.
Many companies think that if their contract says “the customer owns the tooling,” they are protected. But in practice, ownership is about more than the contract language. It is about: Can you actually get your tooling if the relationship ends? Can you move it to another factory? Do you have the technical documentation? Is the mold properly maintained?
We addressed this in three ways. First, we included detailed provisions in the contract about tooling storage, maintenance, record-keeping, and return. Second, we set up a separate tooling ownership certificate — a standalone document that both parties sign, clearly identifying each mold, its owner, and its location. Third, we made sure the client received full tooling documentation (2D drawings, 3D files, material specifications, maintenance records) when the tooling was accepted — not just at the end of the relationship.
The result was that the client did not just “own” the tooling on paper — they had practical control over it. If they ever needed to move production, they could do it quickly and with minimal disruption.
A second important judgment was the quality acceptance structure. The client initially wanted a simple “we inspect and either accept or reject” approach. But we convinced them to set up a more nuanced system with multiple acceptance stages: first article inspection (FAI), production validation testing (PVT), and ongoing statistical process control (SPC). This structure caught problems earlier, reduced the number of full rejections, and gave both sides clear criteria for what counted as acceptable quality.
Key Outcomes
- IP protected before sharing designs: trademarks registered in 3 classes, 2 design patents filed, 3 software copyrights registered, NDA in place with all candidate factories
- Factory selected: chose the mid-priced option (not the cheapest) based on overall capability and quality track record — a decision that proved correct when the cheapest factory later had quality issues with another client
- Manufacturing agreement: comprehensive contract with strong IP protection, clear tooling ownership provisions, and structured quality acceptance
- Tooling properly documented: 6 molds documented with ownership certificates, full technical files delivered to client upon acceptance
- NPI completed on schedule: EVT, DVT, and PVT phases completed on time, with quality issues identified and resolved in each phase
- First mass production shipment: 5,000 units shipped with 0.3% defect rate — well below the industry average of 1-3% for similar products
- Dual-sourcing plan: by the end of the engagement, we had helped the client qualify a second source for one critical component, reducing their supply chain risk
Since the initial engagement, the client has scaled production to 50,000 units per year and has started developing their second product. They continue to work with us on manufacturing legal support, IP enforcement (they discovered a copycat product on a Chinese e-commerce platform and we helped them take it down), and China entity setup as they prepare to establish a local presence.
FAQ
Q: How can we protect our IP before sharing designs with factories?
A: Several layers are important. First, register your core IP in China before you start sharing — trademarks, copyrights, and any patents you can file. Second, use a strong NDA specifically designed for manufacturing discussions — not a generic one-size-fits-all NDA. Third, share information on a need-to-know basis: you do not need to send every design file to every factory you are talking to. Start with high-level information for quoting, and only share detailed designs after you have narrowed down to your top candidate. Fourth, consider using watermarking or other tracking measures on sensitive design files.
Q: Who should own the tooling — us or the factory?
A: We almost always recommend that the customer owns the tooling, not the factory. If the factory owns it, you are locked in — you cannot move production to another factory without starting over with new tooling. If you own it, you have flexibility. But ownership is not just a contract clause. You need proper documentation, a clear maintenance agreement, and a plan for what happens if the relationship ends. The most important thing is practical control, not just legal ownership.
Q: How much does manufacturing legal support typically cost?
A: For a typical hardware product going from prototype through first mass production, legal support usually ranges from RMB 50,000 to 200,000 depending on complexity — the number of custom parts, the number of suppliers, and whether you need help with factory selection, IP strategy, or dispute resolution. For early-stage startups, we often structure this as a fixed-fee package covering the full NPI phase so you know the total cost upfront.
Q: Is it worth paying more for a better factory?
A: Almost always yes. The cheapest factory quote rarely results in the lowest total cost. Quality problems, delays, rework, and management time spent troubleshooting usually cost far more than the initial price difference. We have seen many cases where a company saved 15-20% on the per-unit price but ended up spending 2-3x that amount fixing quality problems and dealing with delays. The goal should be the lowest total cost of quality, not the lowest per-unit price.
Q: What should we do if we discover our product is being copied in China?
A: Act quickly but strategically. First, gather evidence — purchase samples, take screenshots, document everything. Second, evaluate your options: administrative complaint (through the IP bureau or market supervision), civil litigation, platform takedown (if it is sold on e-commerce platforms), or a cease-and-desist letter. The right approach depends on what the infringer is doing, how big they are, and what you want to achieve. Sometimes a quick platform takedown is enough; sometimes you need to take more aggressive action. The important thing is to have a plan before you act — because once you confront the infringer, evidence can disappear.