07 Jul China ODM factory designs: 7 pitfalls to avoid
If your company needs a new electronic product made then the fastest and lowest cost way to get this done would seem to find an ODM factory in China already making very similar products. They have the leanest design and cost structure, the experience, and the large volume which allows for the lowest cost. On top of that the fee they charge to customize the product to your specifications is often only symbolic.
It all sounds tempting, but having developed products with factories in China and Taiwan for 20 years I’d like to share a few pointers:
1. Be aware you have no ownership
The changes you are requesting will be modifications to the design which is the Intellectual Property (IP) of the factory. You may be working with the factory for a year, and pay them quite a bit of money in the process, but if in the end you decide to part ways you will not be able to take the design elsewhere, so you will need to start from scratch. One more reason to pick your partner very carefully.
2. Make sure you have a firm spec
It’s best to have your specifications completely clear and frozen before the start of the project, because that is what you will get a competitive quote on. If the spec evolves during the process it is much harder to control the additional costs charged for additional features.
3. Your ideas may leak
Remember that this factory is making similar items for several other brands, perhaps even for your biggest competitor. The large Taiwan ODM’s tend to be pretty serious about their Chinese walls between accounts, but with the Chinese ones this is not a given, and so every improvement you introduce has a good chance to find its way into other products coming out of the same factory, and if your product idea is strong enough it likely will become a standard item to be shown at the next trade show.
4. Bring big orders, or expect serious delays
Most factories have very limited R&D resources, and their mission is crystal clear: keep the manufacturing lines humming. Every client requiring R&D support goes on to the little white board with the top 5 clients ranked by likelihood of bringing in large orders soon. A factory may accept your project when their business is slow, but as soon as a bigger fish comes along you will be put on hold. The best you can do is to keep the factory engaged by focusing on the large sales your new product will generate.
5. Don’t try to be Steve Jobs
If your team keeps coming with improvements on the user interface, or if you’re picky on getting all the aesthetics just right, be aware that most factories have a very short attention span. If your project requires too many rounds of optimization, no matter if they are caused by your changes or their mistakes, they may go for lower hanging fruits first. They may stall, charge you ever higher fees or flat out refuse to do any further modifications.
6. Accept that Firmware stays a Black Box
Plastics and PCB layouts can be copied fairly quickly, but recreating firmware may take a lot of time, and a factory who has spent years perfecting it will not let it go, which means you will never get to see it. Unfortunately firmware is often their weakest spot, and so the issue most likely to create long delays, and meanwhile your clever engineers and consultants are biting their nails because they are not allowed to help the engineers who keep on banging their head against the same wall for ages.
7. Don’t base a B2B product on a consumer platform.
More likely than not, the platform the China factory uses is based on some local IC. These low cost IC’s are generally tightly tailored to support one standard reference design only, and when you try to customize you tend to run into a lot of limitations, not to mention bugs. Furthermore lifecycles in the consumer industry tend to be short, ASIC’s can go End Of Life (EOL) without much warning, and all your customization effort will have to be redone.
Despite these pitfalls China still is the place to go for manufacturing, see “Why can’t the US build consumer electronic products?”. If all you want to change is the logo and perhaps the color of the plastic then you’re likely fine with an ODM as long as you maintain a strong local presence and do good QC. But if you want to make substantial changes, and are not yet 100% sure what the final spec should be, then I really do not recommend picking a factory as your development partner, their business model is just not set up for it, and it’s a marriage that has a high likelihood to end in tears, especially if you don’t bring the purchasing power of Apple. In fact, if your company is considering to raise funding it is even more important to own your IP.
As for myself, in the beginning I worked quite a bit with both Chinese and Taiwanese ODM’s, and in the end some of them gave me good results, but it always generated a lot of frustration, and importantly took a lot longer than planned: in electronics market windows tend to close fast. So I have completely given up on this approach and at www.titoma.com we do all engineering in-house in Taiwan, giving us & our clients complete control.
Do you have positive experiences, or more tips to prevent trouble? I’d love to hear from you!
Keesjan Engelen is founder of www.titoma.com and has been developing electronic products in Taiwan for 20 years. He also wrote: Kickstarter Prototype to Production: $100K is not enough