(FORTUNE Magazine) – Look up, look down Keep your ear to the ground Keep your ear to the ground --Heather Nova

As a venture capitalist and a writer, I am often asked to comment on the state of the venture capital market and on the outlook for startups in particular. The questioning has increased as Internet stocks have pulled back, and it is safe to say that Silicon Valley is abuzz with prognostications. While I am generally bullish about the amazing opportunities for innovation and about the overall outlook for Internet startups, there is one issue that causes me to pause--something I will call "supply-driven capitalism."

One way to gain perspective on the state of the venture capital and startup markets is to contrast the demand for startups and the supply of startups. Startup demand is driven by needs of consumers or business. Demand increases when their needs are left unmet by current corporations, or when technology or innovation allows those needs to be serviced better, faster, or cheaper. The Internet is an extremely disruptive technology that creates many such opportunities in virtually every market. As such, I would characterize the demand for startups as quite high.

The supply of startups is driven by a combination of entrepreneurial interest and capital availability. It would be hard to characterize entrepreneurial interest as anything but extremely high, with many students from top-tier MBA programs forgoing $150,000-a-year careers in investment banking or consulting to enjoy the thankless, lower-salary career of a startup founder. (Of course, there is that great potential upside and the reward of creating something new.) Venture financing has followed these MBAs. In fact, venture capitalists invested a record $22.7 billion in 1,557 U.S. companies in the first quarter, according to the National Venture Capital Association.

With startup demand and startup supply both at an all-time high, it is difficult to determine whether there might be an imbalance in the current market. Entrepreneurs might be collecting more money from venture capitalists, but does that show any actual demand for startups? How much innovation is really needed, and do we have enough companies providing it?

To gauge this supply-and-demand curve, check out the origin of today's typical startup. My take is that too many of today's startups are "supply-driven." The ideas for these startups are born purely in the mind of the entrepreneur, not in the need to solve a problem in the market. While this take is almost impossible to measure definitively, my guess is that it indicates an imbalance between startup demand and supply.

To shed more light on this situation, let's look back several years to a time when there were fewer startups, and those that were around had less exposure. Frequently the idea to start a new company would come from an entrepreneur's seeing an obvious need for a product or service--often while at his current job. Take, for instance, a systems integrator who develops an application for a customer and finds that two or three other companies have an interest in the product. Eventually the systems integrator realizes that there is more value in the application business and shifts direction. This is clearly a "demand-driven" startup.

Many successful companies over the years have started with this demand-side bias. Cisco was started when several people at Stanford University noticed the need to integrate heterogeneous computer networks. They had demand for the product right there at Stanford. Soon others wanted the product, and before you knew it, a company was born. Likewise, Michael Dell began offering direct sales and support of branded PC products in Austin, Texas. Buying PCs was difficult, and setting them up was even tougher. So people appreciated Dell's hands-on relationship. And when Dell couldn't get his hands on branded PCs, he built his own. The customer valued the relationship so much that he moved his business to Dell's brand.

There are many more examples like these, and all have a similar theme. The market began to emerge along with the supply of a product. There are numerous stories of venture capitalists attending sales calls and receiving a bird's-eye view of the intensity and breadth of the demand for the new product. In other words, there was some certainty of demand before the formation of the startup.

Today an increasing number of entrepreneurs start companies with little more than a gut feeling that someone will eventually need or want their product or service. This form of "ivory tower" capitalism is quite dangerous. The startup is left with the enormous burden of establishing a market rather than participating in one. As they say on the beach, it's easier to surf a wave than to create one.

A second risk to the supply-driven startup is the relatively low barriers to entry. Let's face it: If you can think up a concept for a company in your head, there are probably five other people who can dream up the same concept. And as the supply of capital for startups increases, you can bet that you will be one of many, not one of few. Take the oft-cited example of the e-tail market for pet supplies or the many vertical exchanges in the B2B arena. Last I counted, there were something like six venture-backed seafood exchanges. I doubt that the canneries are crying out for that much innovation just yet.

I suspect that Platonic idea creation is much more intellectually appealing than combing the earth for problems to solve. But bear in mind that even a sexy Internet company like eBay (in which Benchmark's an investor) was born of demand rather than supply--Pierre Omidyar's girlfriend wanted a place to trade Pez dispensers online. The company became a powerhouse after the market voted its support for the idea. Entrepreneurs and venture capitalists alike would be well served by returning to the boring, but surely more successful, world of demand-driven capitalism.

J. WILLIAM GURLEY is a partner with Benchmark Capital, a venture capital firm. Except as noted, neither he nor Benchmark has a financial interest in the companies mentioned. To receive an expanded version of Above the Crowd, visit www.news.com, or to subscribe to the e-mail distribution list, please send an e-mail to subscribe-above_the_crowd@atc.revnet.com. If you have feedback, please send it to atc@benchmark.com.