CoStars on the horizon.

Today, it's everywhere. All the major real estate firms in the country have one. But when Andrew Florance started CoStar Group's database in the late 1980's, he had a hard time convincing brokers that they needed computers, let alone databases.

"We found ourselves trying to sell this new product and having difficulty because no one had hard drives," he laughs. "So we first had to sell the brokerage firms on the idea of personal computers."

So how did he come up with the concept of property information software?

"The thing that drove me was that I came out of commercial real estate, my brother was a broker, my mother and father were in real estate and I knew that people wasted a lot of time trying to collect information," Florance explains. "Brokers were spending 20% of their time collecting basic data. Our thought in starting CoStar was that it would be a tremendous time saver for the industry, as well as enabling brokers to make more intelligent decisions."

According to Florance, who started the company in his senior year at Princeton University, keeping the business going for the first few years was an uphill battle. He was persuasive enough to convince a Washington, D.C. lawyer and his friend to invest about $1 million in CoStar, but as he and his staff went ahead with the project, they realized they had underestimated the market.

"At that time, an established brokerage firm in a major city would track 2,000 or 3,000 buildings and consider itself very knowledgeable. We discovered there were usually three to four times as many properties in the market. The size of the job was getting bigger and bigger as we went," he notes. "In the course of our work, we had photographed 1.7 million buildings. When you think about that, it's sort of stunning how many properties are out there."

In addition to the sheer volume of work it had to do, CoStar was facing competition from much better funded sources. At the time, major market players, such as Dow Jones and Goldman Sachs, were trying to sell their own databases. Today, all of CoStar's competitors from the 1980's have abandoned the enterprise, while the CoStar Group is expanding on the international arena.

"Back in 1994, we were just trying to make the concept work in Washington, D.C.," Florance says. "Then, we opened in New York City and the project just exploded. We are currently in 50 U.S. cities and we are going into research operations in 20 new ones. We are the number one player in the U.K. and we will have a lot of growth in the U.S., the U.K. and Canada in the next few years."

In addition, the CoStar Group has signed contracts with companies like CB Richard Ellis, Cushman & Wakefield, Trammell Crow and Jones Lang LaSalle, all of which have satellites on the continent. So though Florance is concentrating on the English-speaking world right now, he thinks that the day of worldwide operations is not far behind.

"We will have a foreign language version of our system done by the fourth quarter of this year," he promises. "We want to finish virtually every important city in the U.S. because our customers want to turn to CoStar for all their information, but in the next 10 years, we will probably be in 20 additional overseas markets. Our top clients are all international and we hope to grow international with them over time."

In general, Florance compares the CoStar database to the Zagat Restaurant Survey--everyone uses it and everyone wants to be in it.

"It's in the owner's interest that his data be in our system," he explains. "If I own a major property and my goal is to keep it leased up, CoStar is the most powerful tool there is and it's free. If you put your building into our system you are exposing it to anyone who might want to lease. If you don't list it, thousands of brokers won't even know that you exist."

It's true that sometimes the company gets calls from disgruntled landlords who are upset about their buildings getting a lower grade, but Florance argues it's part of the business.

"Some of the information we list will not be favorable, but it's more important that the property be visible," he says. "Probably every single day someone emails me and is upset that we classified his building as class C, but it's part of the nature of what we do. If Zagat rates you as 4 instead of 5, you'd still be foolish not to be on Zagat."

Of course, Florance admits that CoStar's databases might have changed the way things are done in the real estate business, but he thinks the changes are mostly for the better.

"You look at things like the down cycle we've had in 1993, when the supply and demand balance got really out of whack, and I don't think that would happen again because all the banks and all the brokers have enough information about what's really happening in the market," he offers. "So we can avoid that severe swing that causes so much damage. The flip side of that is we've redefined what makes a firm compatible or non-compatible in the industry. Some brokerage firm somewhere might be losing 20% of its deals. But I don't think it's a bad thing. That's just the way business evolves."