When a devastating earthquake rumbled through northwestern Turkey a few years ago, entire city blocks were left in ruins. Yet some buildings remained standing amid the wreckage, lonely testaments to the value of sound construction.

But not always. That would be too simple.

PETER FENTON: The first-mover strategy has been overused and misunderstood. The irony of the dot-com land grab of 1998-1999 is that when you look back at the most successful companies of the last decade—such as data-storage company EMC Corp. (www.emc.com), personal-computer maker Dell Computer Corp. (www.dell.com), and maker of customer relationship management software Siebel Systems (www.siebel.com)—none of them was the first to market.

At our venture-capital firm, Accel, we see thousands of companies each year that claim first-mover advantage. However, first movers rarely achieve advantage by simply being first to market. Instead, many succeed by being the first to achieve economies of scale or scope, to redefine a customer experience for the better, or to take advantage of structural resources in a unique way.

It’s important to consider the underlying structure of the market. If you’re in a highly technical category—semiconductors, for example—being the first to market actually does matter. The market window can be very small. There may just be three to six months for a new design to take hold. The early products get built into larger networks and become central components of other companies’ offerings, which can be hard to change down the line. If you’re not there with the right product at the right time, you don’t have a business.

On the other end of the spectrum are what I consider to be operational businesses. Dell is an example. In this area, being first can actually be a disadvantage. There is a lot of value in learning from the inefficiencies of others in an established market and coming up with a structurally better business model than the early movers have. This approach may sometimes take years to execute but is often a better strategy.

GEOFFREY MOORE: The three first-mover winners that come to mind from the last few years are online retailer Amazon.com Inc. (www.amazon.com), auctioneer eBay Inc. (www.ebay.com), and Web portal Yahoo Inc. (www.yahoo.com). Each one genuinely got an advantage from moving first. They built a base of customers who didn’t readily switch.

Because of these high-profile successes, the idea of being a first mover became a shibboleth that, like all interesting ideas, became badly overused. I think you have to put restrictions on the idea of first-mover advantage. It’s much more limited than was originally thought.

For one, being the first mover doesn’t mean being first under the sun. Some people have said that the advantage really goes to the first company to achieve scale, because of the operational efficiencies that can come with size. The phrase I would use is "first into the tornado." The "tornado," which is the term I use for the time when a market goes into hyper-growth, is when a company really benefits from being in the lead. For example, PeopleSoft Inc. (www.peoplesoft.com) was the first mover in enterprise resource planning software, but SAP AG (www.sap.com) was first into the tornado. Computer makers Apple Computer Inc. (www.apple.com) and Hewlett-Packard Co. (www.hp.com) were first movers in the personal digital assistant category with the Newton and 95LX, but Palm Inc. (www.palm.com) was first into the tornado with the Palm Pilot.

In addition, first-mover advantage doesn’t mean anything if you can’t create barriers to the next person coming in or if you can’t discourage customers from switching from you. Many of the Internet pure plays didn’t have barriers to entry or exit.

Finally, there can be a second-mover advantage. If there are mistakes to be made and the market can’t be locked up, then being second or third, and watching others fumble, is a huge advantage.

FENTON: The best example of second-mover advantage isn’t from the Silicon Valley start-up world. It’s Wal-Mart Stores Inc. (www.walmart.com), a very late entrant in the retail market.

Because he wasn’t first, founder Sam Walton could go study his competitors: what they were doing right, what they were doing wrong. When Wal-Mart operated out of its single store in Arkansas, Walton would go hang out at the competitor’s store to try to figure out its strengths and then incorporate them into his own strategy.

However, through ruthless execution, Wal-Mart achieved a number of "firsts" that drove real long-term advantage. One was geography. Because Wal-Mart studied how the competitive landscape had played out, it spotted promising locations where competitors hadn’t yet established themselves. It moved there first, typically building big retail sites in rural areas that could only support one store. Competitors couldn’t dislodge it.

It was better to come later and learn.

Our theory with Walmart.com [a joint venture between Wal-Mart and Accel] is that you can take a lot of the same principles Walton used and learn from what other online businesses have done.

MOORE: I am wondering if it is mostly important to be first in the area that differentiates you overall.

FENTON: Right. Dell wasn’t first in the personal-computer market, but it was the first to change the customer experience of buying computer systems and to make the supply chain more efficient. If you’re able to do something like that, then achieve scale and defend against competitors’ efforts to overtake you, you gain a big advantage.

MOORE: Some companies, even if they move early, try to stay hidden for a prolonged period so they won’t attract notice from competitors. Companies want time to attack a problem and get acceptance in the market for their products before they emerge more fully onto the scene.

I’ve thought a lot about the varying fates of Ariba Inc. (www.ariba.com), which creates software that helps companies use the Internet for procuring supplies, and Agile Software Corp. (www.agilesoft.com), a software maker that helps businesses share their product designs and plans with suppliers over the Web.

Agile stayed under the radar for a long time and built a strong customer base in the market for contract manufacturing of electronics. When Agile finally hit competitors’ radar screens, it had an unshakable hold on that first industry. From there, Agile had a very strong position to go after a second and third industry.

Ariba, on the other hand, burst onto the scene, playing for first-mover advantage. It played out its hand so early that companies such as SAP and I2 Technologies (www.i2.com), a supply-chain management software developer, could co-opt its idea.

FENTON: One of the biggest mistakes I see in the application software business is that companies think they need to immediately have a presence in three or four industries because they need to be first in all of them.

Agile is a great example that shows success in one attractive industry is sufficient to achieve scale for a young software company. Datasweep Inc. (www.datasweep.com), a software company that allows customers to keep track of what’s happening in the manufacturing process, is another emerging example of how truly owning one industry from top to bottom is much more valuable than owning a little bit of many.

Datasweep targeted high-tech manufacturing companies and has been exceptionally successful. It can serve both contract manufacturers and their customers, and their interest feeds on each others’, because Datasweep gives them a way to share information.

Some may call this first-mover advantage, but it’s really the ability to be the first to achieve scale in an important segment. Great companies then move into adjacent segments, but very rarely do companies own a little of a lot of segments.

MOORE: A book called Profit From the Core: Growth Strategy in an Era of Turbulence, by Chris Zook, proposes the same notion: building a power position in one area and then extending it into adjacent industries. At Chasm Group LLC (www.chasmgroup.com), we used to call it the bowling pin theory, because knocking down one pin can knock down others.

The advantage of a company initially staying focused in one industry is that the players know each other and they talk. They’ll hear your company’s name from three or four different people. Pragmatic people make decisions by relying heavily on what they see others doing, so they will be much more predisposed to adopt your product.

This kind of deep indoctrination is permanently sustainable. Look at Apple. It has had so many struggles, yet, to this day, it still has a strong market share in the graphics and publishing markets. Once people vote with their feet, they stay put mostly because of the costs of switching.

Scoring an early success in an industry can also help with the investment community, where there is a specific pecking order. The first company to reach critical mass in a market is valued far more than the second one and so on down the line with the later players.

FENTON: Yes, being first can matter. Look at network equipment maker Redback Networks Inc. (www.redback.com), which dominated the other companies that came right after it even though their products probably had equivalent features. Redback won some contracts while it had the market to itself and already was riding a curve—based on customers’ experiences and the company’s growing scale—so the others were never going to catch up. Redback also understood the importance of being first to educate a customer base and "own" the wave that it was riding.

The advantage isn’t always permanent. Juniper Networks Inc. (www.juniper.net) has been able to catch up to Cisco Systems Inc. (www.cisco.com) in the router market, even though Cisco had become the standard in the market and Juniper started much later. The reason is Juniper’s products’ speed and the number of features offered. But Cisco still has the advantage in its distribution channel and customer relationships. I don’t think that race has been called yet. I think we’ll continue to see a lot of jockeying for position between those two businesses.

MOORE: I want to look at another first-mover example. I put Amazon in the camp of those that have enjoyed a first-mover advantage, and I believe its advantage will be sustainable.

As far as I can tell, it will own the book-distribution business and some of the compact-disc distribution business forever. When I poll audiences—of varying degrees of sophistication—I ask who has bought a book from Amazon. Virtually every hand in the room goes up. That’s a phenomenal brand statement.

Amazon founder Jeff Bezos has achieved a Martha Stewart-like brand in book reselling. The problem may be that to reach the valuation for his company that he aspires to, Amazon will need to be much more than a bookseller. I am a bit skeptical as to how much further Amazon will scale. Still, I am astounded at how powerful the allegiance is in its initial beachhead market.

FENTON: I agree. When companies can establish a brand early on, the brand lasts virtually forever. If you go back and look at an industry like catalog retailing for example, there’s Sears, Roebuck & Co. (www.sears.com), whose brand has survived since the very beginning despite numerous setbacks. Neither Amazon nor Sears was literally first to market in its category; however, they were the first to provide a defining customer experience.

It happens in other business and consumer markets, too. The experience is defined by a particular brand, and that brand survives. We don’t fully appreciate the power there.

It is worthwhile to consider another great counterexample to this first-mover theory—Siebel Systems. I remember hearing from folks at Siebel’s competitors that Tom Siebel had come into the market for selling software that automates front-office operations such as sales, customer service, and marketing. "He’s so late," they said. "The competitive landscape is already littered with failures, and there are three or four emerging leaders." But Tom managed to change the rules. He decided he wanted to sell to a higher level in organizations and line his company up more directly with consulting firms, which would help install the complex systems. Even though he was a late entrant, Siebel emerged as the category-defining leader. In the process, he changed the category. It came to be more about customer-relationship management than sales-force automation.

MOORE: Right now, in the market for software applications sold to business, the three gorillas are Siebel, SAP, and I2. None of them was a first mover. SAP didn’t emerge on the radar in the U.S. until at least 1993. And Manugistics Group Inc. (www.manugistics.com), which also makes supply-chain management software, was ahead of I2 by at least 10 years.

FENTON: Another fascinating feature of the application software market is that speed to market has mattered much less to success than the quality of the founding chief executive. For instance, Tom Siebel’s supreme optimism and efficacy as a sales and marketing person have really defined his company. Sanjiv Sidhu’s passion and domain experience carried I2 through eight years before it achieved a dominant scale.

I challenge you to name a big, multibillion-dollar application software company where the founder isn’t the chief executive or the active chairman.

By the way, the inverse is true in the market for communications equipment. Rarely, if ever, is the founder the chief executive in the companies with multibillion-dollar successes.

Moore can be reached at gmoore@chasmgroup.com. Fenton can be reached at pfenton@accel.com.