VoIP's Quantum Leap
VoIP providers face a shift from "How?" to "How many?"
By Gary Kim
Fat Pipe Online
December 15, 2004
There comes a time in the development of every truly significant mass market product when a quantum shift occurs, when a series of small quantitative developments will result in a sudden, dramatic qualitative transformation. For voice over Internet protocol (VoIP), that time
has arrived.
The temperature of water can rise or fall, imperceptibly, for quite sometime before anything happens. But suddenly, there is a quantum change, and water changes to steam or ice. In the same way, VoIP has undergone a quantum shift. Where many small technological developments have lead us to this point, leadership now shifts in new directions.
Give credit where it's due. Thousands of people and hundreds of companies have labored mightily to prepare the way. And they've succeeded. "There's been more success with VoIP in the last year and a half than in all the last 10 years," notes Michael Khalilian, International Packet Communications Consortium CEO.
Still, there comes a time when the ability to turn up millions of accounts, very rapidly, is the key to ultimate success. "It's all about distribution now," says Christopher Wood, AT&T VoIP product manager.
It may not be entirely true that "we've been our own worst enemy, confusing the marketplace, struggling with interoperability issues and high equipment costs," as Stan Little, Sylantro vice president, says. "Frankly, we weren't ready for prime time."
Still, "as long as we continue to require a customer to go buy a Linksys router to get service, we don't drive customers to buy," Little notes. "We sell to soccer moms, and I shudder to think what Joe Six Pack thinks when he sees a recommended wiring diagram connecting four boxes."
But that phase is drawing to a close. "The issues we face now aren't about technology," Little says. "Cell phones took off despite significant quality-of-service issues, because people could figure out how it improved their lives."
Shifting Stage
Right now, the stage is shifting. "How do you take building blocks and create services and offers?" asks Dan Dearing, NexTone Communications vice president. "How do I build an infrastructure to compete at $25 a month?" asks Simon McIver, Net2Phone director. "Channels are an issue, once you figure that out."
In the final analysis, "telecom is about economics," McIver notes. And much of telecom economics is about scale, automation and process refinement. "Sales volume requires scaling processes, even though everybody starts out manually," says Rich Grange, New Global Telecom CEO. "Can you replicate all your processes in new markets?"
And make no mistake: VoIP now is becoming a challenge even for the largest tier one service providers. "It's the first phone service we've ever launched that requires new networks, new devices and new services all at once," says Wood.
"Turning up 10,000 customers is one thing, but one million customers is quite another matter," Wood notes. Taking nothing away from Vonage and the other early pioneers, Ron Harden, Volo Communications senior vice president, notes that "Vonage spent $100 million and has 230,000 subs. Some of the service providers we're talking to already have five million customers and will quickly convert a million."
If "30 percent of broadband residences are using VoIP in three years," as Grange believes, and if broadband accounts continue to grow at anywhere near current rates, that will equate to a need to activate and support something on the order of 20 million consumer accounts.
And if, in four years, "72 percent of the U.S. workforce is remote or mobile, and 33 million Americans already work at home," as Avaya vice president Susan Bailey argues, scores of millions of VoIP lines serving businesses and consumers will need to be provisioned. Any way one looks at the matter, "in the near term, we have to scale it," says Chuck Rutlege, Quintum vice president.
So credit Vonage with a lot of effective missionary work. Still, says Little, "I think AT&T is really carrying the water for the whole market, right now, helping buyers develop a sense of need."
And, to some extent, service providers shouldn't worry so much about what the "killer app" is. "The last people you should ask, regarding any killer application, are service providers," says Wood. "No consumer focus group ever said they wanted instant messaging."
Indeed, as the history of many wildly successful innovations shows, customers cannot express a need for something truly new that they've never seen. "Customers will decide what the killer app is," Wood says. Are there problems? Sure.
Still, "at this early stage, we have learned from the growth of the Internet networks," says Little. "Over time, network and interoperability issues get worked out."
OSS S.O.S.
Of course, service providers may not have the luxury of time or money. "Providers are feeling pain today," says Kevin Farrell, Telegea vice president. "Vonage is turning up 25,000 to 30,000 customers a month, so time to market is key." And "customers often substantially underestimate the amount of effort required to turn up customers," Farrell notes.
In fact, "VoIP is much more complex than TDM [time division multiplex]," says Kaynam Hedayat, Brix Networks director. "There are many more chances of failure, because the availability and reliability of each network element is the issue, not simply the technical parameters of jitter and latency."
Add to that the fact that, in many cases, traffic moves "over two to three networks, and that makes troubleshooting lots more complex," says Netcracker Technology vice president Julie Wingerter. "Who's responsible, where's the outage? A device can fail, but it might be three layers below where you're looking."
"The supply chain is more complex than it used to be," says Todd Benjamin, Rodopi Software CEO. "The new relationships among trading partners have to be reflected in the system."
Also, provisioning tasks are quite different, since service providers must create mesh data networks, not simply administer calling plans. "That means IP-VPNs, configuration of the multimedia networks, site management, server provisioning and then end-to-end application delivery to each desktop," says MetaSolv Software senior manager Rajiv Tankhar.
Complicating matters is the fact that services often are delivered over wide geographic areas, over networks owned by several different providers, with whom any single service provider has no direct business relationship. "In that case, the customer is the integration point," says Benjamin.
When something breaks, the customer has to manage the process, and the issue is, "Where do they start?" Benjamin notes. "There are new points of failure, such as name servers that fail to point properly at other feature servers."
"A hard disk problem can cause a new order to be rejected," says Sameh Yamany, Trendium chief technology officer. So can temporary network congestion. "In cases where a customer terminal adapter doesn't work, it often is a signaling issue," Yamany says.
In other cases, "maybe the signaling gets through, but the real time protocol stream ("bearer traffic" in a TDM network) doesn't," says Sansay CEO Andy Voss.
"When push comes to shove, time to scale hangs on your operations support systems," says Rob Kunzler, Telution director of marketing. "It isn't time to market, but time to scale, that's the issue.
"You can come to market with a duct tape solution at first," says Kunzler. "But hundreds of thousands to millions of customers will be quite different." At that point, "it's all about reducing the cost of sale and delivery, time to revenue and the ability to upsell vertical services," says Kunzler.
"It's the single most important element for your business plan," Kunzler notes. "Manual processes can be crippling."
As an example, Kunzler points to one cable TV operator using a manual provisioning process. "Entry of a single order took 48 hours, just to get an order into the system," he says. "There was no device activation at that point. It was almost like you needed one customer service representative for each order."
"Right now, people are testing whether a VoIP service works by making a call from every phone," says Michael Smith, Clarus Systems vice president. "You can't test manually, in volume."
If, as most of us expect, VoIP retail pricing continues to decline, there could be real pain for any provider that has not automated processes. That's especially important if "measured rate calling is dead but service providers still have to pay per-minute costs to terminate traffic," says McIver.
"We are in a pricing death spiral," he warns.
Part of the solution is a greater reliance on more efficient distribution channels for mass market customers. Retail distribution is part of the answer, AT&T believes. Of course, "you have to close the sale," says Wood. "It isn't as simple as just putting the boxes out there." Nor is the process of automating and simplifying all the processes required to activate and deliver service.
"You have to activate at the point of the sale, fast," says Kunzler.
"One thing we see is that price competition will be like long distance seven to eight years ago," says Kunzler. "Prices simply will commoditize, so channels will differentiate." To the extent that service providers have "feet on the street," the support systems must have the ability to rapidly produce a quote and then fulfill, in volume, Kunzler says.
To gain speed, lots of providers will build VoIP systems separately from their legacy platforms. That helps, but at some point those customers still need to be ported back over to the legacy systems. That's another challenge.
The incipient VoIP industry is about to enter a new and different phase of development, where the skills required for success are radically different. Among those skills is the ability to rapidly and repeatedly add large numbers of new customers. In other words, unless scale is built into a service provider's core operations, success will choke it.
In such cases, the first often is last, while the last is first. FAT
About NetCracker Technology
NetCracker Technology is a leading provider of inventory-based OSS and IT Infrastructure Management solutions for telecommunications service providers and telecom-intensive enterprises. Built on an architecturally open, 100-percent web-based J2EE platform, NetCracker's pre-integrated modular solutions manage telecom assets and telecom service delivery from start to finish. NetCracker provides customers with an intuitive GUI and the flexibility to implement out-of-the-box or custom-tailored solutions. Some of NetCracker's OSS modules include Network Inventory Management, Outside Plant and Asset Management. Founded in 1993, NetCracker Technology is headquartered in Waltham, MA and can be reached at 781.419.3300 or www.netcracker.com.
Back to the news page 
