Start a conversation

Shenzhen

From working prototype to first production run

The prototype proves the idea. The series proves the drawings, the tooling, the supply chain and the tolerance for variation — four different problems.

France → China

A prototype is a single object made by someone who understands it. A production run is a thousand objects made by people who will never meet you, from documents that have to be unambiguous without you in the room. The gap between the two is where most hardware projects lose their schedule and their margin.

The four steps that sit in the gap

  1. Design for manufacture. The factory reviews the design and tells you which features are expensive, fragile or impossible at volume. This review is cheap and it is the single highest-return hour of the project.
  2. Engineering samples from real processes. Not 3D-printed look-alikes: parts made with the actual moulds, the actual materials, the actual assembly sequence.
  3. A golden sample, signed by both sides. One physical reference that settles every later argument about what « correct » means.
  4. A pilot run. A few hundred units on the real line, to find the defects that only appear when people repeat a gesture five hundred times.

What the drawings have to carry

  • Tolerances on the dimensions that matter, and explicitly none on the ones that do not — over-specifying raises the price of everything.
  • Named materials and finishes, with an accepted equivalent where you can live with one. Otherwise a substitution will be made without you.
  • The acceptance criteria: what is a defect, what is cosmetic, and what the acceptable rate of each is.
  • Packaging and labelling, which are part of the product and are almost always specified last and badly.

The tooling decision

Tooling is the point of no return: once a steel mould exists, changing the design costs real money and real weeks. The instinct to order tooling early, to « gain time », is the most expensive instinct in hardware. Tool when the engineering samples have survived a round of real use, not before.

A useful rule: nothing goes to tooling that has not been held, used and broken by someone who is not the designer.

Questions we get on this.

How long does it take to go from prototype to first production run?

It depends on how much tooling the product needs and how complete the documentation is. The variable that moves the schedule most is not the factory's speed but the number of design changes made after engineering samples — each one restarts part of the sequence.

Should I pay for tooling myself or let the factory own it?

Paying for tooling and stating in writing that you own it keeps the option of moving production later. Tooling paid by the factory is usually cheaper up front and effectively locks you in. Whichever you choose, the ownership and the physical location of the moulds should be written down before the first payment.

Keep reading on this.

ShenzhenTooling: the cost people discover too lateUnit price gets all the attention in a first quote. The mould is what actually locks you to a supplier, and it is rarely discussed until it is paid for.China sourcingSamples: how many rounds, and what the last one is forThe sample loop is where the project's time is really spent. Its purpose is not approval — it is producing a physical reference both sides are bound by.China sourcingWriting a specification a Chinese factory can actually quote againstThe factory is rarely the problem. The specification given to the factory usually is — and it is the cheapest thing in the whole project to fix.ShenzhenHong Kong or Shenzhen: where the goods should leave fromTwo ports an hour apart, with different paperwork, different costs and different reasons to choose them. The decision is rarely about distance.

If this raised a question about your own project, tell us where you are — the first conversation usually clarifies more than a proposal.

Talk about your product →Read the full guide