Visit Your Local PBS Station PBS Home PBS Home Programs A-Z TV Schedules Watch Video Donate Shop PBS Search PBS
I, Cringely - The Survival of the Nerdiest with Robert X. Cringely
Search I,Cringely:

The Pulpit
The Pulpit

<< [ Broadly Speaking ]   |  Going to the Dogs  |   [ Sharper Than Ever ] >>

Weekly Column

Going to the Dogs: Why Venture Capitalists Have Such a Hard Time Making Money on the Internet and How to Change That

Status: [CLOSED]
By Robert X. Cringely

One thing that is clear, almost a decade into the era of the commercial Internet, is how hard it is to make money online. ISPs, with the exception of AOL and MAYBE Earthlink, just scrape by. EBay makes money, as do some online brokers and travel sites, but most of the other profits go to equipment and software providers, not to web sites or data services. We have been through half a dozen business models in search of one that really works. Portals, search engines, B2B (business-to-business), B2C (business-to-consumer), P2P (peer-to-peer), industry-specific exchanges, and banner ads have all pretty much come and gone. Even pornography is suffering from too much competition according to my friend Franchesca Scalpi, who runs Flynt Digital. Yet every month there are more people on-line. Internet use is now up to 170 million in the U.S. and 400 million worldwide. Surely with numbers like those there ought to be many ways to make money. The problem is two-fold. First, like any new information industry, it takes time to find what you are good at. Alexander Bell intended his telephone to be used to bring music to remote audiences. Television was supposed to be a videophone. Second, the people who have been leading this process of discovery, the venture capitalists, generally don't know what the heck they are doing.

VCs don't like to see themselves as boobs, but many are. Some have become very successful investing with the "spray and pray" technique of funding a lot of companies and hoping for a few big winners. Some VCs are more selective, thinking they are adding value by doing so, and perhaps they are, but generally not in the meager profit fields of the Internet. The Internet is something totally new for VCs used to the high profit margins of traditional high tech. The Internet is more like the grocery business, with painfully thin margins and high capital requirements. You can't spray and pray as an Internet investor, yet that's generally what has been taking place.

Part of the problem can be traced to something that every VC seems to know. It is a mantra they repeat verbatim and, by doing so, reveal their weakness. In describing the various elements of risk faced by a venture (technical risk, financial risk, competitive risk, etc.), when they get to market risk, the VCs will always say the same line: "Will the dogs eat the dog food?" This line, which suggests to me by its universality that there is a secret VC training camp somewhere in the mountains west of Sand Hill Road, is intended to mean that however good a job you do at starting a company and building a product, there is no way to really know if people will buy the thing until it is on the shelf and ready for sale. But what it also says is that VCs can't relate to customers.

I don't want to over-labor this point, but it bears further explanation. At the risk of sounding like some televangelist who gets a 15 minute sermon and two calls for contributions from a single line of scripture, this line says that customers are different from VCs (they are dogs), and the VCs can't know in advance what customers will do. And why should they? Normal people aren't millionaires, yet many VCs are. Normal people aren't used to building and crushing companies and occasionally forcing them to mate, yet that's what VCs do all the time. The only area where the typical VC could qualify as a dog is for products aimed specifically at VCs as customers. But VCs as a class are too small in number to make a reasonable customer base for anyone except business jet builders.

Want to know why Webvan failed or, better still, how it even came into being? It is simply because venture capitalists aren't like you and me.

None of this mattered much in the days when venture money was plentiful, and spraying and praying seemed a perfectly reasonable technique, but those days have passed. Now the VC community has, for the most part, lost confidence in its ability to make obscene profits with very little effort. Reality has finally hit home, and the VC industry is struggling for an alternative to spray and pray. Fortunately, I stand ready to provide just such an alternative.

I think there is just now emerging a new archetype for Internet success, one that we can only hope the venture capital community comes to understand and embrace. Maybe they'll even come to think they invented it.

The first company I found that follows this new archetype is, a real estate Internet startup based in suburban Minneapolis and just now emerging from stealth mode. Real estate is an industry that ought to be perfect for automation. Millions of people buy and sell homes each year. Searching for homes online is a no-brainer, as is shopping for services like mortgages and title companies. And yet recent history shows us it is very hard to make money with a real estate Internet startup. Microsoft couldn't make money with its HomeAdviser. is in bankruptcy. The only company that is even close to successful is, thanks to a successful IPO and lots of money in the bank. But isn't profitable, either. For all the profit potential in real estate on the Web, I couldn't find a single pure Internet play that makes money.

Given this grim reality, why then would Roald Marth and Elizabeth Chesen choose to start It all comes down to dogs and dog food.

The typical venture-funded Internet startup begins with a target market ("Millions of people are looking for new homes!") and then builds an organization to serve that market. Jeff Bezos, founder of, was an investment banker, not a retailer of any sort, when he made a list of items that could be sold over the World Wide Web and decided to start with books. Bezos knew nothing about books or retailing, which is to say he knew nothing about dogs or dog food. Amazon, like most of the real estate sites, relied on size to take it to eventual profitability. But will be profitable by the middle of next year, only about a year after its founding. This is because is a dog food company founded by dogs to serve dogs.

In the U.S., homes are sold by agents who work as independent contractors for brokers. Nearly all real estate Internet startups are built on the idea that their revenue will come from home buyers. This makes perfect sense if, as a startup founder, you don't really know the real estate industry. But it turns out to be very hard to get enough revenue this way to be profitable. gets its revenue from agents, not home buyers. For $99 per month, gives an agent a personal web page to list their homes for sale, an e-mail address, a wireless e-mail device, a portal to link their listings to the world, even a digital camera to take pictures that are uploaded to the personal page. Simple workflow tools manage the agent's page and handle a continual stream of e-mail, keeping everyone in touch with the home buying transaction as it goes forward. Home for sale listings are instantaneous, which is a big advantage over Multiple Listing Service sites that are often weeks behind. It's all good technology packaged in a way to be truly useful to an enterprising agent.

This sounds very nice, but the key to profitability for lies within its capital structure. is owned by brokers. These brokers are in a position to require the agents who work for them to purchase services as a condition of their employment contract, just like a mechanic has to own his tools. needs to have 10,000 agents as customers in order to be profitable. Since just one broker who is on the board of employs more than 6,000 agents, the road to profitability is short. can build a good business just from its own investors, which is something Homestore could never do.

The change of paradigm here is that is not only owned by dogs, it is owned by alpha dogs, and that alone assures its success. With thousands of brokers employing millions of agents, has a scalable business opportunity that can quickly dwarf previous Internet real estate efforts. I believe this company, in which I own no stock and with whom I have no business relationship, will be a big success.

But real estate isn't the only area where this approach can work. A Utah company called TermSeek is about to do much the same thing in the language translation industry. There are approximately three million professional translators in the world who work as independent contractors for about 18,000 translation agencies. Most translators use computers and the dominant translation application is TRADOS. Microsoft is an investor in TRADOS, which is based in Northern Virginia. In this example TRADOS is analogous to Homestore.

TermSeek is just about to roll out their application, which is cleverly called TermSeek. They believe it to be faster and easier to use than TRADOS. But the real advantages of TermSeek over TRADOS lie in the business model. For TRADOS, those 18,000 translation agencies effectively act as first line technical support, helping solve problems for contract translators. They provide this support because they have no real choice and aren't paid for it. TermSeek, on the other hand, distributes its software over the Internet, but through links from the translation agencies, which receive a sales commission. TermSeek effectively does pay them for support through these sales commissions. Which application will be more popular with translation agencies? And TermSeek is leased, not sold, so translators pay by the month and their up-front costs are low. Which application will be more popular with translators? Again, I own no stock in TermSeek and have no business relationship with the company.

Just as's broker-owners can push their product, TermSeek's translation agencies will push the translation software. Both are closed systems that can be profitable entirely on their own. Both are companies founded by dogs for dogs.

Comments from the Tribe

Status: [CLOSED] read all comments (0)