When Google first announced its Glass prototype, complete with a snazzy video, it generated a lot of excitement. Through augmented reality projected onto the lenses, users could seamlessly navigate an urban landscape, send and receive messages and take photos and videos. It was a completely new way to interact with technology.

Yet criticism quickly erupted. Many were horrified that hordes of wandering techno-hipsters could be surreptitiously recording us. Others had safety concerns about everything from people being distracted while driving to the devices being hacked. Soon there was a brewing revolt against "Google Glassholes."

Now, however, Wired reports that Google Glass is taking on a new life and gaining traction as an industrial tool. From factory floors to operating rooms, augmented reality is proving effective at improving productivity, safety and documenting procedures. There's a lesson in all of this. When you introduce a revolutionary product into the market, target the few, not the many.

Your Idea Will Always Be Wrong

Think for a second about planning a surprise night out for your spouse or significant other. When they arrive home, you inform them that you have a surprise in store and whisk them away for an evening that you have meticulously planned in detail, from the appetizers to the perfect place for after dinner drinks.

There's a chance you might get lucky and predict exactly what you partner is in the mood for on that particular day, but more than likely, you will be sorely disappointed. The evening may even devolve into one of those recrimination filled episodes in which you both begin to wonder if you even know each other at all.

This happens because humans are complex creatures and our preferences often defy prediction or rational explanation. As we have seen with Google Glass, this is especially true when large numbers of people are involved, because we not only have to predict what users want, but a variety of other stakeholders as well.

The truth is that what can seem like a brilliant idea in a conference room or even when tested among a limited number of customers can fail miserably in the real world. That's why your first assumption always needs to be that your idea is wrong. Maybe it's off by a little and can be tweaked later on, or maybe it's doomed from the start, but you can be sure it's wrong in some way.

The Right Idea At The Wrong Time Is Still Wrong

By now, most people know the story of how Xerox PARC invented the technology that led to Apple's Macintosh. It's one of those cautionary tales about how big companies lack vision and wily young entrepreneurs are able to dream big and change the world. Yet that is a gross misapprehension of what actually happened.

The truth is that Xerox had the bigger idea. It set out to build the "office of the future" and transform how people worked. The product it built, the Star, was not a standalone computer, but an integrated system that included multiple, networked computers connected by an Ethernet, a laser printer and software for email and other applications. It was years ahead of its time.

Steve Jobs, on the other hand, had a much smaller idea. He built a much simpler, easier to use computer that was cheap enough for consumers to buy for their homes. You couldn't actually do much with it, except play video games, write documents and run simple spreadsheets, but it was fun and everybody who bought one felt like they were on the ground floor of something big.

Go to any office today and the computers you'll find look a lot more like the Xerox Star than Apple's Macintosh, but still it was Apple that succeeded and Xerox that failed. As innovation expert Tim Kastelle has put it, the right idea at the wrong time is still wrong.

Identifying A Visionary Customer

The problem, as Steve Blank explains in The Four Steps to the Epiphany, is that most new products target large addressable markets. That may work with an incremental innovation, but for something truly new, you first need to identify "visionary customers" or people who want or need a product so badly that they don't care if it's not quite perfect.

Consider the case of self-driving cars. It's an exciting vision that may very well change the world one day. It will fundamentally change what it costs to distribute products, help free up urban congestion and, most importantly, will save countless lives. The possibilities are profoundly exciting.

Yet a more apt market for self-driving technology is not on city streets or even rural highways, but in a remote Australian iron mine, where driving a truck is hazardous work and the supply of labor is tight. The need for the technology is so great at these mines the economics are a real no-brainer and, with little traffic in the area, safety concerns are minimal.

The same is clearly true for Google Glass. It might seem cool to have the functions of a smartphone projected right before your eyes, but it's far from essential. Having goggles that can walk you through a complicated procedure at work, on the other hand, is tremendously useful and produces measurable economic value.

Getting From Few To Many

While focusing on a just a few, visionary customers is not the ultimate goal, it will get your product into the hands of people who will pay you and help you refine it by exposing flaws that you're not likely to be aware of at first. They will also be a great source of insight about which features you should develop next.

That's exactly what's happening now with Google Glass. As the technology gets battle tested in real world work environments, the technology itself is being improved. The company is also building out an ecosystem of partners who are designing industry specific applications and building traction in the marketplace.

As it stands now, what was once a flop looks like it may end up being a significant commercial success. Of course, Google is an amazingly profitable company and can easily afford the costs associated with the failed Glass launch. Most of us don't have that luxury. So if you're dreaming up a new product, build for the few, not the many.