Engineering isn't manufacturing
Every new prototype felt like progress. Looking back, they realized they weren't reducing the uncertainty that mattered next.
The situation
I spoke recently with the cofounder of a company building headphones incorporating a visual experience that moves with the music. The founder shared this story because they thought other hardware founders might recognize parts of the journey.
Like many first time hardware founders, they didn't begin with a background in consumer electronics or manufacturing. They had an idea they were excited about and wanted to see if they could bring it to life.
From the beginning, the path seemed fairly straightforward. Build a working prototype, continue improving it until it felt like a product, then take it to a factory to figure out manufacturing.
The plan
The founders spent the next couple of years trying to execute that plan.
They hired contractors, experimented with different approaches, and slowly worked through the technical challenges of integrating the novel visual feature into a pair of headphones. Some contractors made little progress.
Others disappeared altogether. One finally produced a prototype that worked, even if it was held together with glue and rough 3D prints.
For the founders it represented a huge milestone. They finally had proof that the concept was possible.
With a working prototype in hand, they thought they were entering the final stretch. The plan was to keep refining it until it looked and behaved like a production product, then hand it off to a factory.
Where uncertainty remained
They continued prototype iterations. Over time, the prototypes became noticeably better. The industrial design improved, the technical challenges became more manageable, and videos of the product began spreading online, attracting millions of views and a growing audience eager to buy the headphones.
Looking back, the founder realized something - although the prototype kept improving, they weren't getting closer to manufacturing. They still didn't understand tooling, assembly, supplier capabilities, be required to manufacture the product consistently.
Every engineering milestone made it feel like production was getting closer. Looking back, the founder realized they were answering the wrong questions.
What needed to be proven first
Everything shifted once they began working directly with a headphone manufacturer in Asia. The conversations stopped revolving around building a better prototype and started revolving around building something that could be manufactured.
Decisions that had previously been made in isolation were now shaped by the people who would eventually produce the product.
The founder realized the factory had information they needed much earlier, and involving them sooner would have changed many of the engineering decisions they spent months making.
The pattern
One of the interesting things about hardware is that every stage creates a different set of questions. Early prototypes answer whether something is technically possible. Manufacturing partners answer a different question entirely: whether that same idea can be built repeatedly, economically, and at scale.
Progress on one set of questions doesn't automatically reduce uncertainty in the next. It's possible to spend months improving a prototype while learning very little about production.
If you're in a similar spot
If you're investing heavily in another prototype iteration, it can be useful to ask whether you're trying to build a better prototype, or whether it's time to bring manufacturing partners into the conversation.