I’m reading a book called “Founders at Work”, by Jessica Livingston, one of the founders of YCombinator. I bet she started it as a research project to figure out how best to incubate companies. It’s a series of interviews with founders of information technology companies, mostly in software, mostly in Silicon Valley. But very few of them, if any, came out of “incubators.” In fact, most of them started their companies while working at full-time jobs elsewhere, or in the few months after leaving their full-time jobs and before getting funded.
I’ve been reminded of a couple of things by reading this book, which might better be called “adventures in software engineering.”
First of all, there’s no such thing as a one-product company. Many of the recent buzzy names in the industry, familiar ones like Bloglines, Flickr, and Blogger, were ideas that were developed into products, acquired users, and got sold. They weren’t even businesses, much less companies. They got sold to bigger companies that actually WERE companies, and needed new product lines. Some of them sold the company before they sold even a single product.
Second, it’s mind boggling how much software today is free or remarkably inexpensive. Quickbooks, FlickrPro, TypePad Pro – these are the more expensive forms of free or bundled software. Most Web 2.0 tools, and many older tools are free. Very few companies actually make a business of selling software. Microsoft is one, Adobe another. Intuit would be a third. The others exist on advertising, or on licensing deals, or on other less obvious business models. So software’s a difficult product to build into a company, from scratch.
Third, most of the companies in this book were incubated by the surrounding environment, not in a physical space. Because they started in places with good lawyers, good mentors, good PR people, and good programmers, they were able to operate on the cheap for a long time, receiving sustenance from the community. Incubators are necessary only for companies that are doing research.
These companies were not doing research. They had already found both the problem and the solution. And they were not building fabs, like Intel. They were not doing research, like Genentech. They were not searching for copper, coal or gold.
The Silicon Valley software startup environment over the past thirty years is unique in history. Because it’s no longer a manufacturing economy, it doesn’t need big infrastructure, or even big teams. “Companies” of two or three people are sold for millions.
Think of the differences between a software company and, say, a mining company or an electric utility. Any company in the manufacturing sector, the natural resources sector, or the agricultural sector cannot be started in a garage by people in their early 20’s with ideas. A hospital chain cannot be started by two guys in a garage with an idea.
And yet the model for starting tech companies has become the accepted model for business startups. Two guys in a garage has become the paradigm for success.
We are facing an interesting future. In most Western nations, an aging population demands health care. In most developing nations, a young population demands technology and resources. Layer on top of these demands the pressure to save the planet from destruction by both sides of this equation.
We might be past the era of the simple software startup. The next generation of venture capital may be given to companies with far greater infrastructure demands and more basic problems to solve: clean air, clean water, affordable housing. Ironically, although the problems are more basic, the tools to solve them will be far more sophisticated. Creating software will increasingly look like a piece of cake.
Will Silicon Valley be able to retain its relevance in the environment I’m predicting? Not without getting on airplanes and going to where the deals are happening. And not without radically re-thinking the concept of two guys in a garage.