Broadband lesson: Power's in the plumbing

Today, most of these highfliers have closed their doors or sought shelter from angry creditors in bankruptcy court. Investors have watched tens of billions of dollars in market value simply evaporate.

Unfortunately for employees, investors and jilted customers wondering what happened, an autopsy reveals that the signs of distress should have been obvious even during the heyday, but there was probably little that could have been done to prevent it.

The messy collapse of Excite@Home marks the end of an ambitious era, in which start-ups believed they could make millions of dollars on Internet service without actually owning the most critical pieces of the infrastructure themselves.

Excite@Home's implosion follows a string of bankruptcies and closures of virtually every prominent start-up hoping to offer high-speed Internet service. NorthPoint Communications and Rhythms NetConnections, two other high-profile wholesalers, have vanished into bankruptcy court. Many smaller broadband companies have disappeared more quietly.

A string of bad luck and bad business decisions certainly contributed to the financial carnage, industry analysts say. But the almost perfect record of failure has a common thread: The profit is in the pipes, and these upstarts did not own them.

"There is profit to be made here," said Rob Lancaster, an analyst with telecommunications research firm The Yankee Group. "But the right business model comes when you own the actual infrastructure."

Putting down roots
The early days of the broadband business are reminiscent of the early cable industry. Those video pioneers had huge capital start-up costs and few early subscribers to help recover costs. But owning the cable lines that ran into people's homes--as well as 20- to 30-year franchise agreements to keep competitors at bay--put them in the powerful role as gatekeepers. As a result, most franchises eventually became money machines and were gobbled up by larger companies.

Excite@Home's demise powerfully illustrates why the power is in the pipes--the conduits into people's homes that it did not control.

At the time of its initial public offering in mid-1997, the company said it had spent more than $57 million to build its technological and corporate infrastructure. At the same time, it had just 5,000 paying customers. The high up-front costs and relatively meager subscription revenue mirrored the development of the cable industry in the 1970s, but that's where the similarities diverge.

Though the cable industry spent billions of dollars building the infrastructure, often incurring heavy loads, at the end of the day the companies owned hugely valuable connections directly to customers' homes.

Rather than commit to a lengthy and expensive project to lay their own lines in hundreds of cities, the high-speed Net businesses partnered with the telephone and cable companies.

Although there were strains, particularly between digital subscriber line (DSL) providers and telcos, the alliances held appeal for both sides. The broadband companies got quick access to customers, and the pipe owners received lucrative per-subscriber fees at little cost.

But the marriages of convenience did not necessarily produce equal partnerships. In the case of DSL, the phone companies were consistently accused of dragging their heels in signing up new customers. (The reason: New DSL customers meant lost revenue for their own high-speed services, such as ISDN.)

As for Excite@Home and the cable companies, the imbalance of the partnerships was reflected in the revenue-sharing agreements. Excite@Home received about one-third of the average $40 monthly bill, while the cable companies pocketed two-thirds.

In Excite@Home's case, the balance of power shifted further from the company as the cable industry gradually developed the know-how to build its own networks.

"Ultimately, (the broadband companies) didn't bring enough value to justify the relationships," said Jupiter Research analyst Joe Laszlo.

Paying the pipe owner
On the cable side, Excite@Home and Road Runner were formed as joint ventures with the cable companies. On the telephone side, NorthPoint, Covad Communications and Rhythms NetConnections built their own DSL networks that piggybacked atop the local phone companies' lines, although their relationships with the dominant "Baby Bell" phone companies was dicey at best.

These strategies showed signs of financial strain from the start.

@Home's arrangement with its cable company partners gave it only about a third of the subscriber revenue that the high-speed Net service brought in, ultimately about $13 to $15 per customer. The company originally believed that its costs would drop to the point where this would be a feasible business model. But this proved an elusive goal.

After the company finally went to bankruptcy court in late September, its attorneys negotiated a temporary arrangement with cable companies that would give it $20 a month per subscriber. It took this much to cover its costs, attorneys said--but until it reached bankruptcy court, and threatened to shut off service to millions of subscribers, Excite@Home never had the leverage to negotiate this kind of contract.

DSL companies such as Covad, Rhythms and NorthPoint faced a different set of issues. Their high-speed Net services had to compete directly with those offered by SBC Communications, Verizon Communications and others, even as they used the same wires.

That made for an uncomfortable relationship. The small companies frequently went to court or state or federal regulators arguing that the big phone companies weren't living up to their legal duties, dragging their heels in "provisioning" or setting up the lines that the start-ups needed to offer service.

Even when the lines were set up, for years the small companies found it difficult to make a profit. The telephone companies dropped the price of their own DSL service to compete with Excite@Home and the cable companies, but were charging wholesale prices for access to telephone lines that ran as much as $15 to $20 a month.

"It was eating us alive," recalled Abhi Ingle, vice president of marketing for Covad, the only remaining major provider of DSL access.

Ultimately, regulatory decisions made that cost fall, however. Covad now pays just $5 a line to the telephone companies, easing financial pressures significantly.

Hubris, Wall Street and debt
The simple cost for access to the "last mile" connections to customers' homes wasn't the only thing that decimated the first generation of broadband companies. Created in the Net's boom times, the companies expanded their networks and their businesses far too rapidly, incurring debt that they couldn't support in leaner times.

Excite@Home's case was perhaps the most egregious, as the @Home broadband business bought Web portal Excite for $6.7 billion, Blue Mountain Arts for nearly $1 billion, and other content companies in its attempt to get additional revenue from its subscribers.

Those myriad acquisitions later became financial millstones for the company as Internet advertising fell, and neither side of the company was strong enough to prop up the other side.

On the DSL side, start-ups found they had easy access to capital and that Wall Street was valuing them on the basis of how fast their networks were growing.

Covad, typical of the three leaders, spent $500 million in 2000 alone to build its network, incurring close to $1.4 billion in debt.

"Wall Street was saying, 'Don't worry, just get big quickly,'" Ingle said. "We fell into that trap."

When capital dried up, the companies found they didn't have enough money to support their operational and debt costs. The troubles snowballed as their Internet service provider clients, who bought DSL at wholesale rates, went out of business.

NorthPoint, which never took full advantage of the regulatory decisions that dropped the costs of phone lines, was the first to go under. Rhythms followed not long afterward, and Covad went into bankruptcy protection to restructure its debts.

Last ones standing
As the smoke clears from the broadband collapse, only a few players are likely to be standing.

The most obvious ones will be the big telephone companies and the cable companies, which own the infrastructure. The giants have spent the past few years slowly climbing the Internet learning curve and are now able to run the networks themselves.

Covad, the last of the big independent DSL companies, may wind up a survivor. The company says it will emerge from bankruptcy protection later this month, debt-free, and that its costs have been reduced to the point where it can break even by 2003. A recent round of funding from SBC Communications will help it survive until then.

Also left standing will be a few big ISPs, which have the leverage to negotiate sustainable wholesale rates from the network companies, analysts say. America Online and MSN are in this camp. EarthLink has a chance as it grows larger. Others are less certain.

But the past few years will certainly serve as textbook examples of how not to create businesses, analysts say. New companies now have detailed maps of the minefields that surround entry into the Net access business.

"All the relationships and agreements in the broadband world were made up," The Yankee Group's Lancaster said. In that kind of environment, he added, "you're going to have failures before you have successes."