The road to success

This year, some Wall Street analysts believe that Apple could sell as many as 70 million iPhones and a further 30 million iPads. And if breaking through that magic 100 million sales figure doesn’t set the pulse racing, then consider this: Atlantic-ACM, the specialist telecoms consultancy, estimates that on average, each phone owner will consume around 500Mb of capacity a month and each tablet owner around 1Gb.

Apple’s vision for a future in which its mobile devices dominate our every waking moment might still be some way off, but the demands that the gadget giant’s ambitions are placing on the US bandwidth infrastructure are very real indeed. AT&T already concedes that data usage among US consumers has surged 5,000% since the launch of the iPhone. Now that the iPad is exerting its dominance in schools, hospitals and factories across the US, demand for bandwidth capacity will continue to explode.

Of course, Apple is by no means the only supplier lighting fires under the market: Verizon’s rolling launch of 4G, already in around 30 National Football League (NFL) cities and now extending out to Tier 2 cities, is equally bandwidththirsty. And as more applications are devised to take full advantage of LTE technology, capacity demand will surge yet further.

If the challenge is to absorb all this traffic and carry it back to the network backbone, it is one that Ethernet technologies stand ready to meet full on: they are more flexible and scalable and, as such, are spearheading a paradigm shift in the way networks are designed, throwing the backhaul market wide open to nimblefooted metro providers in the process.

“The Ethernet revolution is completely redefining the competitive playing field in backhaul services,” explains Chris McReynolds, senior director of product management at Level 3. “Everyone who is an asset-based provider has a shot at this market – it’s not like the private line networks that were concentrated in the hands of a very small number of legacy players.”

McReynolds, who oversees Level 3’s involvement in more than 100 separate metro markets, sees no let up in the pace of growth in Ethernet backhaul: “There are very interesting business models out there and a lot of mechanisms to expand your network, build, partner with one or more wireless providers or leverage off other people’s last mile networks. Just like Ethernet is reshaping the enterprise market, so it is also reshaping the wireless backhaul market. It’s a completely different ball game to the private line voice world and it’s all to play for.”

Ron Mudry, chairman and CEO of Tower Cloud agrees: “Right now, there’s probably a two-year window of opportunity in which to establish yourself as a major player in wireless backhaul. It goes without saying that whoever does that well, will be incredibly successful.”

For so long, the talk in the industry was of the transition from T1s to Ethernet – and for so long, the reality never quite matched up to the hype. “Well, last year, we made that crucial jump and by the end of this year, the focus will be pretty much exclusively on Ethernet,” explains Mudry, a founding father of the metro markets with more than 27 years experience. Now, he says, the race is on to make it count. “Once the build-out phase has run its course, would-be providers are going to find it much harder to break in – partly because many customers will already be signed up to long-term contracts and partly because the growth trajectory flattens out.”

At FiberLight, the Georgia-based network provider which serves more than 21 metro markets, the flip is almost complete: “I would say that the ratio of our business is now roughly 80:20 in favour of Ethernet over dark fibre and that’s a complete reversal from where we were in, say, 2007,” says Mike Miller, founder and CEO. “Crucially, we were not particularly encumbered with a lot of legacy infrastructure so we were able to roll something out pretty much from day one.” The greater flexibility inherent in Ethernet architecture is proving just as crucial: “It’s extremely easy to bolt on your own additional features and services, so, for example, for the first time in FiberLight’s history, we are launching our own dedicated IP platform to businesses and the government on our own infrastructure. That’s something we are very excited about,” Miller adds.

The platform, which went live in January, took just six months to design and build and is capable of delivering internet services at speeds of up to 10Gbps.

But life looks set to get tougher for metro fibre providers. As Verizon moves out its 4G deployment to smaller cities and even rural areas, the challenges become far greater. “It’s not like you are going to find yourself in a dense metro area where there is a lot of fibre to take advantage of,” explains Mudry. “For one thing, it’s a lot harder to get fibre out to the towers in the first place and for another, the tower sites are a lot further apart. This is a real break from the past for us.”

In the main, providers will have to draw heavily on microwave solutions to meet the challenge, although this in itself could throw up problems where disparate technologies need integrating: “I think microwave will be the only viable alternative in many instances,” Mudry says. “Microwave has played a pivotal role in extending wireless backhaul networks out to the suburbs of some of the larger Tier 1 cities, so carriers have shown that they are comfortable with it when a fibre-based solution is not on the table,” he adds.

Nevertheless, the payback in smaller markets is equally attractive. Not least, the networks should typically be cheaper to roll out, so providers will look more closely at taking on the entire build themselves, from the dark fibre backbone, through to the laterals and even down to the microwave equipment. And because the markets are small, it is not unreasonable to expect a single provider to capture 80% to 90% of customer sites, thereby emerging as a dominant provider and increasing the chances of securing multiple customers to single towers.

As one wireless backhaul provider, which declined to be named suggests, that can have a significant impact on the longer-term profit potential of a tower. It can cost anything from $25,000 to $125,000 to hook up a tower in a rural location, so the return on investment must add up: “In dense metro markets, our co-location rates average between 1 – 1.25. When you drill down to the smaller Tier 3 cities, we will be targeting somewhere north of 2.5.” That might still sound low, but there may only be three or four carriers to target, and at least one of those is likely to be an ILEC capable of satisfying its own backhaul requirements. “So even if getting fibre to the tower looks to be more expensive, the economics might well stack up longer term if you can tickle up the co-location rate.”

Clearly, the proliferation of smartphones has done much to focus the minds of metro network providers on the backhaul market, but growth opportunities abound elsewhere. “International carriers, internet companies, content companies, cable companies – they are all pushing for more extensive fibre networks,” comments Dan Caruso, co-founder and CEO of Zayo Group. “There are also boundless opportunities to extend and deepen networks organically on the back of customer deals.”

The logic is simple: as you roll out fibre to an existing customer, you pass many more who can be hooked up to your network quickly and cost effectively. Those new customers then typically require new fibre-to-satellite offices, which opens up yet more potential new business. And so the virtuous circle continues.

It might be a simple model, but it works – and works well, as Mike Hurley, vice president of sales and marketing at Fibertech explains: “We build anywhere from 600 to 1,200 route miles of fibre every year. Think how many new customers we pass right by with an alternative infrastructure in doing that. As long as you continue to deliver great service to your existing customer base – and, in doing so, win the mandate to supply their new bandwidth requirements – your opportunity set continues to grow.”

Geographically, many metro providers view central and western Virginia as underserved markets ripe for further expansion, partly because of the swathe of government agencies requiring mission critical connectivity and partly because of the number of co-location facilities in the area. In the last four years, for example, FiberLight has built more than 500 route miles of fibre in the region and it plans to double that figure again by 2014, when new routes to Charlottesville and Winchester are completed.

Others look not so much for regional opportunities, but strategic ones. Some believe that mid-market enterprise is crying out for fibre connectivity; others hold that those companies in the Fortune 500 require the most bandwidth capacity and therefore offer the biggest returns.

The latest round of stimulus funding is also pertinent. To some extent, the government-backed networks will become a competitive threat to existing infrastructure providers. They typically rely on a major backer such as a state university or a consortium which may be serving health, education and government needs over the new network, and that takes out of the equation most of the commercial rationale for laying the fibre in the first place. Indeed, it is questionable whether there is enough customer bandwidth in some of these rural areas to build much of a business at all.

Says Hurley: “The stimulus programme is working, in that it is encouraging fibre deployment out to anchor institutions in underserved areas where there is no business case to otherwise support the capital deployment required to get it there. That said, there will be some overbuilding of existing networks back into metro areas and that may create some competitive pressure to existing providers in some areas.”

But the future, Hurley says, looks rosy. “I don’t see any major interruption to the expansion of the metro markets any time within the next decade – the opportunity to connect more buildings and serve more customers is just there and the bandwidth requirements will continue to grow when we get there.”