I may be a bit biased because I’m most happy in front of a Bridgeport with a face full of metal chips, but hardware startups should take one giant cue from web startups: prototype more sooner (homage to Brad Feld and David Cohen).

I’ve heard countless war stories from (now successful) hardware entrepreneurs of rampant feature creep, horribly difficult-to-assemble products, millions of venture dollars sunken into scrapped tooling, and most often: products that simply fail to move off store shelves. We spend so much time worrying. Worrying about that draft angle on the mold tool. Worrying about that microcontroller that keeps overheating. Worrying about impressing buyers at retail stores. Worrying about costs. Oh the costs! But far too often hardware startups fail to worry about the most important thing: am I building a product that people will love.

Most investors/engineers/buyers/CMs/designers/bloggers will say “they just know” great product. I’ve even been known to mutter it on occasion. But the truth is, we’re just making educated guesses. I can play with a product for a few hours and decide if I want to keep playing with it or even buy one. “Just knowing” great product is an illusion. It’s the engineering equivalent of 20/20 hindsight. What makes a truly revolutionary product is foresight. It’s seeing a dozen prototype models and being able to pick out the perfect one that will be irresistible to millions. After all, it’s only the products that quietly weave themselves into every day life that are truly great.

A later-stage foam model of Dyson’s Airblade hand drier. Dyson is known for exquisite designs born from copious prototyping.

Building these types of products is hard. Really, really hard. Some say it takes luck. Some say it takes vision. Some say it takes fantastic dedication or a great team. It takes all of these things. But more importantly, hardware entrepreneurs need to master setting aside their perfectionism. I suggest teams build weekly prototypes. They don’t have to be full revisions. A foam model of a tiny feature by itself is more than enough, but you should find yourself in a shop at least once a week.

These prototypes become invaluable to the product development process. Give prototype models out to users or even random people around town. “Does this handle feel right to you?” Carry prototypes around in your pocket. Give them to “fringe customers” that aren’t in the sweet spot of your market’s bell-curve. You’ll be amazed at how effective this kind of thinking can be towards quelling some of that worry. Doing it more and more often will only hone your critical thinking about how features and products will be viewed by customers. Prototyping more will save you countless dollars and years of development time. Even though spending a day per week in the shop will seemingly slow the development process down, your first hit product will come much sooner.

Note: many may recognize my pedaling of Eric Reis’ Lean Startup methodology. Readers should be aware that the ‘lean’ actually comes from the manufacturing world and our beloved Taiichi Ohno. So to all you skeptics out there: hardware can be lean too!