Lessons learned: Requirements for the profitable last mile

The key lesson of the CLECs is that an economically viable last-mile service provider business requires ownership of the access facility. Regardless of the Federal Communications Commission's recent decision regarding unbundled network element regulation, at its core the CLEC reseller model has a fatal flaw: Service providers must rely on the cooperation of their primary--and more powerful--competition. As the experiences of NorthPoint, XO and many other CLECs demonstrate, this requirement results in an unsustainable business model.

Deployment of new wired infrastructure--whether copper, cable or fiber--is costly and slow. This is true for incumbent local service providers as well as their CLEC brethren. Digging up streets, negotiating rights of way and wiring entire coverage areas before signing up the first subscriber can be cost- and time-prohibitive.

The answer? The wireless last mile. Wireless offers the most potential as a cost-effective, facilities-based last-mile option.

Yet earlier attempts by Sprint, AT&T, WorldCom and others at last-mile wireless access have had disappointing results. In many respects, the fundamental problems of these first-generation wireless systems have been economic, not technical. Most of these systems worked--that is, they delivered high-speed data connections to their residential subscribers. But they didn't provide the operator with a workable business case.

Because these truly "fixed" wireless systems required clear line-of-sight access from the basestation to the home being served, their capital and operating costs exceeded the revenue that could be generated. The line-of-sight requirement meant that in many cases at least half of the subscriber base could not receive service, and that one truck roll would be required just to determine whether a particular subscriber could get service. Second and sometimes third or fourth truck rolls involving skilled technicians would be required to install an antenna on the exterior of the home and properly point it to the appropriate basestation tower. Despite these significant costs, service providers had to price their broadband services competitive with DSL and cable to sign up and retain customers. The result, not surprisingly, is that none of these first-generation systems have succeeded on a large scale.

The shortcomings of these earlier wireless access systems, however, are very instructive for the next-generation of wireless broadband. The fundamental problems Sprint, AT&T and the others faced, along with the challenges local exchange, cable and other wireline companies are experiencing with their broadband rollouts, help illuminate the key requirements for a profitable wireless broadband business case: reducing the initial capital investment required to get into business and allowing for a "pay-as-you-grow" model; minimizing the operating costs necessary to run the business; maximizing the revenue opportunities supported by the network; and keeping the consumer happy to reduce churn and its associated costs.

Reduce initial capital outlay and time to profit

The initial, and perhaps most important, question that our customers ask is, "How much will it cost to simply get into the business of offering this service?" In today's harsh economic climate, service providers seek rapid payback with minimal debt and risk. Even for large, established carriers, the capital markets have not been kind to those who seek to put in a large, expensive network infrastructure where the capital investment far precedes the first signs of any revenue generation.

For a wireless broadband system, this means two things:  The system must support large cells and it must have a scalable, cellular architecture. Large cells enable coverage of a particular geographic area with a minimum capital investment. The beauty of a wireless cellular architecture is that it offers a solution that can scale with subscriber demand. Only a system that meets both requirements will result in a profitable business model.

Although some first-generation line-of-sight systems meet the first requirement in that they supported supercells, they cannot easily scale. If an operator puts in new basestations to handle increased service demand, costly truck rolls are required to re-point antennas located at existing--not just new--subscriber homes to support the new configuration. Moreover, supercell systems are not scalable because they run out of capacity.

Wired options present even greater scalability problems. A network that can support an entire neighborhood or community must be put in (or upgraded) despite uncertain subscriber demand and before even the first subscriber can be serviced. This requires a massive upfront capital investment, regardless of the eventual subscriber take-up.

Next-generation technologies can offer both a non-line-of-sight cellular architecture and large cells for a truly scalable, "pay-as-you-grow" business model. Non-line-of-sight cellular systems allow service providers to begin with an initial capital investment that ensures coverage over a particular geographic area, and then add network infrastructure to handle additional capacity as subscriber demand merits, once the system has begun generating revenue.

Although a scalable, non-line-of-sight system is important, that system must also support cell radiuses of five and even ten miles. With a macrocellular system, service providers can minimize the number of cell sites needed to deploy in a given area--therefore, minimizing the capital investment requirement to get into business. Even what might seem to be a small difference in cell radius--say, the difference between one and two miles--can have a tremendous difference in the business case, as the area covered by a single basestation is an exponential function of the radius (remember junior high geometry: the area of a circle = ïr2). It would take four basestations with a one-mile radius to cover the same territory as a basestation with a two-mile radius. The difference grows significantly as the cell radius increases: A basestation with a ten-mile radius can cover the same area as 100 one-mile-radius basestations. 

Macrocells reduce not only the initial capital expense required, but also the secondary costs associated with each cell site--for example, finding site locations, securing backhaul, providing air-conditioning and other maintenance costs. These costs can drive down a carrier's ultimate profitability, even once a large number of subscribers have been signed up. 

Minimize operating costs

As the first-generation fixed wireless providers quickly learned, installation and customer service costs can cripple a last-mile business. A costly feature of DSL, cable and first-generation wireless systems is the need for a visit by one or more professional technicians to install equipment at the subscriber's home.

With non-line-of-sight capabilities and no need for externally mounted or professionally installed antennas, second-generation wireless broadband systems can substantially reduce the requisite subscriber acquisition costs. These true "plug-and-play" subscriber units, which can be pulled out of the box, connected to a computer and phone using standard connections, and gain a signal instantly, can be as simple to set up as a toaster or coffee maker. Gone are the days of the infamous--and costly--truck roll, where a skilled technician is required to install equipment at every subscriber's premise.

In terms of lowering operating costs, eliminating the truck roll is just the start. As CLECs and ILECs alike have realized, customer service call centers are another driver of high operating costs.  To fully reap the benefits of a truly self-installing subscriber device, a last-mile solution must also simplify the provisioning process. Subscribers should be able to activate service quickly and painlessly, without waiting on hold for the next customer service representative, who then inputs the subscriber's order into multiple billing and service databases. Using a clean-sheet design, second-generation wireless broadband equipment providers can integrate fully automated provisioning systems into their solutions, so that with a click of a mouse subscribers can sign up for service, input their billing information, and add or delete new service features.

A related way to significantly reduce customer service costs is to design a system that enables remote diagnostics and upgrades. With earlier systems, service providers usually have to send out a professional technician--again, the dreaded truck roll--whenever a subscriber complains of a problem. That's because it is difficult for a customer service rep to diagnose and troubleshoot problems over the phone.

What we're learned from the computing world is that a distributed, software-based architecture with intelligence spread throughout the network can result in major operational benefits for the service provider. In particular, with powerful, intelligent subscriber units, service providers can have visibility from the core network into each and every home. This is in contrast to the subscriber units for most DSL and cable systems, as well as for most wireless broadband systems, which use simple modems. Network visibility enables customer service reps to diagnose and troubleshoot problems remotely, reducing the need for technician visits. It also enables remote and cost-effective software upgrades as new product releases are developed--something impossible to do in traditional hardware-based and mainframe-architected systems, where system upgrades require costly replacement of hardware.

If designed with the business case in mind, second-generation wireless broadband systems have the potential to substantially reduce operating costs.

Increase revenue through multiple service capabilities

Service providers want to leverage the same infrastructure for multiple services--and therefore multiple revenue streams. If they are going to make the capital investment to put in a new infrastructure, why would they choose one that supports a single revenue-generating service only? Using the latest technology innovations in IP protocols and quality of service, wireless modem and RF, distributed computing, and advanced telecommunications, new wireless broadband systems have the potential to deliver much more than high-speed data and, in the process, enable the service provider to achieve high average revenue per user. 

• True broadband. The primary service driving most service providers--whether ILECs or CLECs--to consider new networks is broadband. This means the capability to deliver multi-megabit throughput rates that can compete with DSL and cable--and for which service providers can charge DSL and cable prices for "all-you-can-eat" service. Some of the first-generation wireless systems failed to deliver megabit rates, and even many DSL providers are unable to provide more than 100 to 200 kb/s, depending on the subscriber's location and the condition of the copper line. Although not all subscribers will want multi-megabit rates, it is important that service providers be able to offer those speeds--for a premium--to their most demanding users. 

• Carrier-grade voice. Even though high-speed data is the service driving most service providers to consider deploying wireless broadband systems, it's not enough. High-speed Internet access has the strong potential to become a commodity--differentiated from provider to provider by price. As a result, the ability to offer other revenue-generating services, particularly telephony, is very important. But it can't be scratchy cell phone or VoIP-quality voice service; it must be toll-quality voice, as clear and reliable as any landline. The value of true voice service is tremendous, as the additional costs of offering it are minimal, and therefore the additional revenue generated is nearly all profit for the service provider. Even if only a small percentage of the subscriber base adds a phone line to their broadband service, it can substantially affect the service provider's bottom line.
  It is important to note that the ability to provide toll-quality voice over a fully converged wireless broadband network must be integrated from the beginning. Because it requires multi-layer quality of service, sophisticated radio resource management, packet header compression, and a number of other technical optimizations, it cannot easily be added to an already existing data-only system. For voice to be a real revenue generator, second-generation wireless broadband equipment providers must design their systems to support voice from Day 1.

• Value-added applications. The lesson of the i-mode phenomenon in Japan is that new services have the potential to create many new revenue streams to bolster the service provider business case. By taking advantage of the lessons learned in the personal computer and Internet worlds, second-generation wireless broadband systems have the potential to make the creation of new services and features that subscribers want, and for which service providers can charge a fee. 
  For example, an intelligent subscriber unit can house a firewall or parental filters under the control of the service provider. Subscribers don't have to deal with the hassle of setting those up on their own computers and are willing to pay a small fee for them. In addition to home gateway features, because they are fully converged on IP, these systems have the potential to offer multimedia services--such as call transcript, smart address books and streaming video--as new revenue streams but with minimal development expense. 

Decrease churn by enhancing the subscriber experience

Finally, service providers are concerned about the quality of the subscriber experience, which ultimately affects their business case. Unhappy subscribers typically result in high churn rates, which both increases costs and decreases revenues.

Many of the features mentioned above--those that decrease costs and boost revenues--have the added benefit of also making the service easier to use and therefore increasing subscriber satisfaction.

• Ease of Installation and Provisioning. Most subscribers dislike the inconvenience of waiting for the phone company or cable technician to come to their house--sometimes more than once--to install the necessary equipment and activate service, and that hassle is often the reason they decline to sign up for broadband service. But when no truck roll is necessary and they can hook up the subscriber device in minutes, subscribers are more likely to try the service and, ultimately, enjoy the experience. The absence of outdoor antennas is also an important benefit for subscriber concerned about the aesthetics of earlier fixed wireless products.
  Just as they dislike truck rolls, subscribers dislike having to call customer service centers, often being put on hold for long periods of time, to set up or change service. Moreover, there is often a delay between the time the call is made and the time the service is activated or changed. The ability to make changes easily is likely to strengthen the subscriber's loyalty to the service provider, even in the face of competition with lower rates.

• Existing equipment and standard connections. Subscribers want to be able to plug in their existing computers and phones without the need for any additional connectors, equipment (like VoIP phones), or software. If a system is easy to use, subscribers are much more likely to choose--and stay with--it over other systems that require additional hardware or software. 

• "Sticky" applications. In addition to basic phone and broadband service, subscribers enjoy enhanced applications and additional functionality that make their lives easier. When service providers are able to offer "sticky" applications tied to the network, whether they be firewalls or smart address books, subscribers will have a stronger tie to a particular service provider and be less likely to switch providers solely based on the latest pricing special.  

• Portability. Wired broadband systems such as DSL and cable, as well as line-of-sight wireless networks, are inherently fixed. The subscriber can use the service only where there is an established connection. One of the greatest benefits of non-line-of-sight wireless systems is that the service is inherently portable. The subscriber can move the subscriber device to any location within the service provider's coverage area and receive service. No longer tethered by wires to a certain location, many subscribers will value the portability offered by a wireless broadband system as a true differentiator from wireline options, offering them enhanced functionality and reducing their incentive to switch providers.

Lessons revisited

Today's wireless broadband equipment providers have the benefit of learning from the shortcomings of earlier systems. These lessons, along with the major technological advances that have taken place over the last few years, offer wireless broadband companies a unique opportunity to create systems that address the core requirements of the service provider's business needs--systems that minimize the initial capital outlay, reduce operating costs in a variety of different ways, offer many new and exciting revenue streams, and are easy to use.

Both ILECs and CLECs have real and immediate needs for these systems. The challenge before us is to capitalize on this opportunity, learn from the past, and create a new communications future.

Greg Caltabiano is Chief Operating Officer of SOMA Networks.

Visit SOMA Networks online.