This is the final article in a multi-part series about sourcing components for a 2000 unit build. These posts are meant for new hardware teams going through the process for the first time.
- Part 1: Sourcing COTS parts
- Part 2: Sourcing custom parts
- Part 3: Final Assembly
- Case Study: How Kinderlab Robotics Built the First 1000 KIBOs
KIBO: A robot toy that teaches technology to 4-7 year-olds
This article tells the story of Kinderlab Robotics, the company behind the KIBO – a robot toy for children aged 4-7 years old. Children build their own mobile robots with KIBO, decorate them, and program them to do fun things with sound, lights and motion.
The product has many materials – laser cut wood, injection molded plastic parts, printed circuit board assemblies (PCBA), cable harnesses, fasteners and so forth. It is a perfect case study for light assembly.
Kinderlab was co-founded by Mitch Rosenberg – a veteran entrepreneur and rock star marketer, and Marina Umaschi Bers – an award-winning Tufts professor whose research involves the design and study of innovative learning technologies to promote positive youth development. By June 2014, they had taken the KIBO through a few engineering iterations, and were ready to go into production. Perfect timing, since they ran a successful Kickstarter campaign that exceeded their fundraising goal and got them quite a lot of pre-orders. Lastly, KIBO is an eminently giftable product, and they had to gear up for a big holiday season.
Building the Supply Base
As Kinderlab embarked on the production journey, it faced two challenges: Sourcing COTS and custom parts, and finding a way to do the final assembly for their product. Together with their product design partner, Smithwise (formerly known as Boston Device Development), Kinderlab set out to build their supply base. They quickly found solutions for the PCBA and cable harnesses. They also found a domestic solution for the die cut wood that creates the signature look and feel of the product.
However, they faced a challenge on the plastic parts. They looked into short-run molding solutions like Protomold. Based on their volumes, the part cost was too high. So they ran a competitive Request For Quote (RFQ) process for those parts. They originally thought that with moderate quantities (less than 50,000 a year), and the need for flexibility on minimum order quantities (MOQ), they would be better off working with suppliers in the US. Much to their surprise, the clear winner was a Chinese injection molding company with an office in Florida. The MOQ was only 1000 pieces. Smithwise was able to work with the supplier to come up with a strategy for using family molds to minimize the tooling cost. So this was a good partner to work with.
Now it was time to find a way to do the final assembly. They had four choices:
- Build it themselves (two co-founders… four hands…)
- Build it with the help of interns and friends and family
- Build a full-on production facility
- Outsource the final assembly and packout to a contract manufacturer.
The last two were not good options. Since they had sourced all their parts themselves, and it was a relatively simple product to build, the contract manufacturer would not make enough money on labor, testing and packout to make it a good deal.
Building a full-on professional production facility also made no sense, since they wanted to keep costs down and stay flexible and agile. Four hands and 1000 units would have meant that nothing else got done for months. So the only smart thing to do was to build the units with interns and friends and family – which was what they did.
This is Mitch, the CEO, with a stack of fully assembled KIBO kits ready to ship. He was very kind to give me a full factory tour.
Components are shipped in bulk and are sorted into bins on shelves, ready to be assembled into sub-systems.
The assembly procedures call for sub-systems to be assembled in batches. Here is an assembly station where sub-systems are built.
Completed sub-systems are binned and shelved, ready for final packout.
There are three KIBO product configurations. This picture shows the high-end configuration, with all of the optional sensor, light and sound modules.
Once the plastic totes are fully packed, there is one more step: Pack the totes into shipping boxes, then pack the boxes onto pallets for outbound shipping. (Yes, this facility is also the logistics center.)
There you have it. A complete manufacturing facility for what most manufacturers will consider a low volume production run. When you walk through this space, it is clear that they have solved for optimal flow – there is a place for how material flows from one place to the next, and it is logical and efficient. They navigated through the pitfalls by putting leveraging procedures and processes which are being carefully followed, even though they were technically a DIY production facility without “professional manufacturing staff”.
Will this facility scale up rapidly if they experience exponential growth – to 100,000 units a year? Probably not. But they are at a stage where they value flexibility and agility above scalability. And it makes total sense.
The one cardinal sin in the Lean Startup philosophy is to scale up before product/market fit. In their case, they made the smartest choice regarding production management. They retain control and are able to be super responsive to customer feedback. Kudos to Mitch and Marina for setting up a great system to fulfill their customers’ demand!
About the Author
Elaine Chen is a startup veteran and product strategy and innovation consultant who has brought numerous hardware and software products to market. As Founder and Managing Director of ConceptSpring, she works with executives and leaders of innovative teams to help them set up and run new product innovation initiatives with the speed and agility of a startup. She is also a Senior Lecturer at the MIT Sloan School of Management and the Martin Trust Center for MIT Entrepreneurship. Follow her at @chenelaine.