May 14, 2001 12:00 PM,

Field of nightmares

Donny Jackson

On the surface, supplying high-speed access to businesses seems like a simple, sure-fire way to make money. After all, companies are increasingly dependent on the Internet.

Too many mouths

Probably the highest-profile mistakes have come from independent DSL providers, which appear to be facing imminent extinction.

DSL providers NorthPoint Communications, Rhythms NetConnections and Covad Communications once were darlings of Wall Street. Today, NorthPoint is out of business, while Rhythms and Covad have seen their stock prices dip below $1 per share.

One key problem for the DSL providers has been their pursuit of a wholesale model, says Gary Kim, former president of NxGen Data Research. When wholesaling, DSL providers must share revenue with ISP resellers, which provide customers and incumbent carriers, from which copper lines and co-location space must be leased.

“The important thing is getting out of the wholesale model and selling retail direct, which takes an entire step out of it and makes it easier to be profitable,” says Kim. “Under the old [wholesale] model, three people had to be fed; with retail direct, only two people have to be fed.”

Indeed, the model is so cumbersome that AT&T decided not to pick up NorthPoint's customers when it bought most of the bankrupt DSL provider's assets.

“I think a lot of the problem there is that [the customers] came with contracts,” says Bryan Long, vice president of marketing for Copper Mountain Networks, a DSL vendor that counted NorthPoint as its largest customer. “[If I'm AT&T,] I don't want to deal with [ISPs like] Megapath. Maybe I want that customer, but I want that customer. I don't want to have to feed Megapath or Telocity or whoever.”

And the wholesale model looks worse when the struggles of ISPs are considered. In November, Covad announced a restructuring and 800 layoffs, measures it claimed were necessary because many of its ISP resellers were not making payments. Since then, Covad has terminated many of the poor ISP relationships and solicited customers to its Covad.net offering.

But it may not be enough. Covad reported in early April that 93% of its lines were sold via resellers.

Access is not enough

Even without the burden of ISPs, many question whether DSL providers can make money under the current price structure. For instance, Kim notes that Rhythms' customer base is “about 80%” retail direct, but the DSL provider still may run out of money. Rhythms has announced it is considering financial options, including the sale of the company.

Not only must competitors lease lines from incumbents, they must endure the logistical problems inherent in depending on the incumbent — a DSL competitor — to prepare lines for service. Hence, not even AT&T should threaten ILECs in the DSL market, says Drake Johnstone, an analyst at Davenport & Co.

“No one besides the RBOCs have had any sustained success selling DSL,” Johnstone says. “RBOCs have not proved easy to deal with for others trying to access their networks. I don't see any reason AT&T would be any different.”

Long agrees that competitors cannot compete with ILECs if Internet access is the only service offered, especially in the residential market. However, he believes businesses want symmetrical DSL, which RBOCs do not offer. More important, because the DSL gear deployed by RBOCs is older, RBOCs cannot add services cost-effectively.

As a result, competitors deploying newer DSL equipment can grab market share by offering added voice lines, storage and multimedia capabilities — services that can generate enough revenue to more than offset the cost of leasing the line from the incumbent, Long says.

“There is a challenge to the industry to make money on DSL services,” he says. “It doesn't matter whether you're an ILEC or a CLEC; the ILECs are in just as bad shape, except they have a lot deeper pockets. Just providing Internet services doesn't work; you can't do it.”

The money isn't to be made in a DSL pipe, he continues. “It is, after all, just a modem technology. The money is to be made in providing the services. The access providers are looking at it and saying, ‘We need to start adding some value-added services to this that we can start selling for a few extra dollars per month.’”

Don't call us BLECs

Another group learning that Internet access alone does not generate enough revenue are providers of in-building solutions for businesses. Once known as building local exchange carriers (BLECs), these competitors roundly criticize the BLEC label, emphasizing they do more than simply provide a broadband pipe into a multitenant unit — the business plan of many failed BLECs.

Today, competitive providers serving MTUs focus on providing services beyond access to tenants — even serving as a IT consultant to smaller businesses, says Allied Riser's Morgan.

“Instead of selling them a fast pipe, we're really trying to sell them a better way to do their business, and that's proving very popular for us,” Morgan says. “So we've seen that the bundle of service really isn't the voice-data bundle necessarily but the full bundle of data services — Web hosting, e-mail, firewall services.”

One new service that some providers will pursue at Supercomm is in-building wireless networking, which is becoming a hot topic among building owners.

“Our building owners have certainly come to us and said, “I'd like to get something in my building to run Bluetooth devices… and get better coverage for my cell phones, etc., etc.,” Morgan says. “To make the wireless environment work inside the building, you actually need more fiber to put distributed antenna systems and things within buildings.”

Meanwhile, many believe introducing voice into the revenue picture is critical to making the economics of in-building solutions work. For CLEC Eureka GGN, the key reason it selected a solution from start-up Kenetec is that the system allowed it to generate voice revenue from a converged network, says Steve Camas, Eureka GGN's chief technology officer.

With revenues from voice and data, the investment in Kenetec's solution should be paid back in less than 14 months, Camas says. A data-only revenue stream would require 55 months for payback, says Nathan Kalowski, Kenetec's vice president of marketing.

“That's why data-only plans are a flawed business model,” Kalowski says. “Voice is a key revenue generator… If you get the revenue, the payback comes. If you don't get the revenue, it doesn't matter how cheap your system is.”

No field of dreams

Of course, the price of the solution is a huge issue, for the service provider and the building owner. In a slumping economy, determining who assumes the capital risk is especially important.

Under the old BLEC model, the competitive carrier almost always paid for installation. With capital plentiful and accountability scarce, the late 1990s represented a “Field of Dreams” for competitors, most of which adopted a “build-it-and-they-will-come” attitude.

Unfortunately for many BLECs, the number of customers that came often didn't exceed the number of competitors that piggybacked on their efforts to supply broadband to a building.

“If you take a look at the space, you see building-centric providers getting good bandwidth from lit fiber companies, and those lit fiber companies are connected to the metro core,” says Joseph Varello, executive vice president of Everest Broadband. “And, as they connect to the metro core, when they [provide fiber to]… a building for a building-centric provider, they bring in a slew of competitors within the building as well,” he says. “You wind up relegating the building-centric provider to an inside-wire company… and that's not the business we want to relegate ourselves.”

To protect its investment, Everest Broadband is building networks using its Metro Ethernet Services Architecture. This network design features a point of presence gateway used to distribute services to buildings in the area. While creating efficiencies in service aggregation and deployment, it also lets Everest wholesale access to its buildings to competitors.

“What we want to do is make sure that we focus on our primary market segment without having to give every other carrier in the market an option to be in the building,” Varello says. “We do that by trying to create, first, a relationship with the retail customer and then ordering enough bandwidth between the building and the POP to…become a metropolitan provider of services for those other carriers that also want to sell to tenants in those buildings.”

Too much, too fast

Of course, BLECs' biggest problem was that many service providers spent too much money installing infrastructure with little or no assurance of return.

“The BLECs had a poor business model — they were taking all the capital risks” says Shah Talukder, Cisco's director of marketing. “They could not take advantage of the opportunity because of the capital crunch.”

Many BLECs certainly overextended their funding on buildouts, but finding someone to share in the investment risk is not easy in, urban areas, where landlords have multiple broadband options.

With this in mind, PhatPipe is pursuing partnerships with owners of industrial parks outside the reach of other broadband options.

And the key to these partnerships is that the property owner, not PhatPipe, pays for the fiber buildout, says PhatPipe President and CEO Don Ankeny.

“The real-estate owners we're working with are very large, very well capitalized entities that have relatively low cost to capital, certainly relative telecom and broadband companies,” Ankeny says. “So it makes sense for that capital investment to be made by that lower-cost-of-capital entity, and then we install, operate, service and maintain and share that revenue with them. When you think about some of the high-rise buildings, our investment is less than the amount they spend on trees.”

While the logic may make sense to PhatPipe, it was a tougher sell a year ago when property owners could hope that an aggressive BLEC would wire their buildings for free. With capital spending slowing, PhatPipe's solution lets property owners ensure that their buildings have the broadband access needed to attract and retain tenants in a tough marketplace.

“[Owners of industrial parks] are more receptive to the idea of writing a check to pay for the infrastructure today than they would have been a year ago when they could have just passively let somebody come in and do it for them,” Ankeny says. “It gives them control over their own destiny to get their properties wired ahead of their competitors.”

The arrangement calls for the landlord to receive as much as 65% of the telecom revenue generated from the property. With property owners wanting to get as much return as possible on their broadband investment, the landlord usually introduces new tenants to a PhatPipe representative personally, helping PhatPipe achieve higher penetration rates, according to Mike Harman, PhatPipe's senior vice president of operations.

With no equipment costs, PhatPipe believes its business model can withstand tough economic times, Harman says.

“We're not spending a lot of capital going into a city and building a point of presence with the hope that we're going to get customers to support that platform at some point and time,” Harman says. “As we create critical mass, we use fatter pipes in providing services for those customers as we have them, as opposed to what some of other BLECs have done, which is to build infrastructure and hope they get the customer base to support it.

“I think we all know what the score is in that game.”

New “old” entrants

The score may be known, but it appears the players will keep changing — not only business models, but names. Long a domain dominated by CLECs, some analysts believe cable companies and ILECs will impact in-building offerings.

Traditionally, cable players have steered clear of business offerings, because their networks were built to serve residential markets and long lacked two-way interactivity.

“Once you go two-way, you can offer broadband services and voice services, and that's what you need to become a CLEC, basically,” says Jeff Moore, senior analyst for network services at Current Analysis.

Comcast has done just that in the mid-Atlantic region, Moore says. While cable companies may be limited to regional coverage at best, the fact that they do not have to lease lines is a big advantage, he says.

“The biggest point of differentiation for cable broadband players is the fact that they have direct access to your business or home,” Moore says. “Lots of players have their own switches, lots of players have their own leased fiber network, lots of players have fixed-wireless possibilities, but there are precious few companies that actually have a line into your home or business, and the cable players are one of the few.”

Of course, the ILECs also fit this description, but incumbents have not been as aggressive in the MTU market, which is fraught with complications regarding the infrastructure in each building. But Cisco hopes to give ILECs an option other than selling high-margin T-1 lines in the market via its Long-Range Ethernet solution, which delivers symmetrical speeds of 5 Mb/s over any type of wiring, according to Cisco's Talukder.

With LRE, ILECs can standardize their installation training and allow their sales force to pursue all buildings because the solution works everywhere, Talukder says. In fact, he believes hotels — often overlooked when broadband deployments are considered — can be an especially lucrative business for ILECs.

“There's always been a last-mile problem,” Talukder says. “But there's never been the right solution set at the right price points with the right deployment space. I think we have hit on something.”

Allied Riser's Morgan disagrees that ILECs will make a big mark on the in-building market, noting that high-quality customer service — including serving as a IT consultant for smaller companies — is critical to success in the space.

“Very frankly, that's just something the ILEC is never going to do.”