The Basics of Bottom Line Profitability

Cost management and revenue management are critical, as success is now determined by margins and how Wall Street analysts view profit models—most of which now emphasize cost management. Traditional talk of revenue and subscriber numbers is now giving way to the basics of bottom line profitability.

Profitability will be greatly affected by the way carriers, service providers, content providers, ISPs and ASPs position new services, bill for them, and manage relationships for reconciliation for those services. With the use of MPLS, XML and middleware technologies, the industry is hoping to improve the bottom line.

Preserving Revenues

“Carriers need to preserve revenues across the service spectrum,” says Jim Guillet, assistant vice president of product marketing for Alcatel’s switch data networking business unit. “As IP and ATM converge with MPLS,” he says, “there increasingly should be a focus on topline revenue growth.”

In multiprotocol label switching (MPLS) networks, an edge router labels incoming packets. The packets are forwarded along a label switch path (LSP), where each label switch router (LSR) makes forwarding decisions based solely on the contents of the label. At each hop, the LSR strips off the existing label and applies a new label which tells the next hop how to forward the packet.

LSPs are established to guarantee a certain level of performance, to route around network congestion, or to create IP tunnels for network-based virtual private networks.

MPLS will enable network optimization, because the technology allows a carrier to pinpoint over loaded and under loaded parts of networks. MPLS-based tools can achieve this by load-balancing networks. By connecting to networks and downloading LSPs, Alcatel, for example, can define link locations and run algorithms to describe what is available and running on the network. “[The tool can] then propose an optimized scenario and load-balance the network, so no one link is over- or under-utilized,” explains Vinay Rathore, director of strategic marketing for routed data networks for Alcatel.

Guillet claims that in trials using new tools for traffic engineering in MPLS networks, networks gain 30 percent additional capacity.

“Whether it’s ATM MPLS or IP MPLS, …it’s not just about connectionless service, but preserving what investment you’ve made into Frame Relay, private line and voice networks,” he adds.

Without MPLS “the level of intelligence is low,” says André Kopostynski, director of marketing at EHPT USA. “Routers only know how to send packets to neighboring networks, but the intelligence to dictate how much bandwidth to reserve for something like a videoconference over IP isn’t there. No standard exists to ensure consistent quality as packets are passed among devices. … Even with email it can take 2 seconds or 30 minutes.”

However, such intelligence will be necessary for setting QoS expectations and interconnect compensation, which will be based on packet splits and data routes. “With voice, you could say a 911 call is more important than a chat with a friend,” Kopostynski says, “yet you couldn’t charge accordingly because of privacy issues, right?” But with data, operators have the opportunity to set the parameters. “They just need to figure out how to get their arms around interconnect so they can charge appropriately,” he says. “Once QoS parameters are set, billing and charging models will dictate that real-time data feeds to a stock broker will cost more than someone checking email routinely from the road once a day.”

Until standards emerge, some operators, such as NTT DoCoMo, are teaming up with content, application and entertainment providers to set dedicated paths to servers and ensure consistent quality.

Interconnect Reconciliation

Once providers take the plunge into IP services, one of the most significant cost management exercises will be that of interconnect reconciliation for reselling arrangements among multiple carriers, ASPs, ISPs and content providers.

As service providers move to offer data services, interconnect reconciliation will become an essential tool for carriers routing traffic through scores of different carriers to hundreds, if not thousands, of destinations each month.

Real-time management of multiple content providers, transport providers, local and long-distance carriers, ISPs and ASPs will be a major headache, since each party must manage tariff plans, service plans, switching clocks, mediation gateways and multiple billing platforms. These are just some of the variables that have to be integrated coherently and consistently across each player’s platform. Inconsistencies in tables or rules will lead to discontinuity in configurations and inaccurate CDRs—a major peril for carriers.

“With the networks getting more and more complex, and with the introduction of content routing and the multiplication of content and application providers, the food chain is getting longer between the provider and the user,” says Martin Demers, senior vice president of marketing and sales at Ace-Comm.

Middleware can be used to improve routing efficiencies as relationships grow in complexity. “If you have interconnections between two carriers through a bidirectional circuit group, and one carrier uses a first-available-circuit traffic distribution and the other uses a random distribution, you will end up with an uneven allocation of calls. You think you lack capacity because of the amount of congestion on specific circuits,” explains Demers. “Some of our customers … had been adding facilities on a route because they saw congestion, when in fact 90 percent of their circuits had little traffic.”

“They would never have found this problem with a 1,000-page traffic report,” he says, “because you’re looking for ‘exceptions’ in this case. Exceptions are far more readily apparent in graphical formats than by reading reams of reports. This concept is based on reporting by exceptions, so that customers can spend their time fixing the problems rather than looking for these exceptions. It’s very easy graphically.”

There are other cost issues as well. Many CLECs and emerging carriers choose to send all data to a billing outsourcer in order to save money, but that sometimes backfires. “Many pay for each record processed, even though as little as 20 percent of the data may be billable,” says Demers. He says another common problem is difficulty in distinguishing records that should be processed by the CABS system from those that should be handled by the subscriber billing system.

This difficulty is often caused by the way the traffic is routed between the carriers. “They may send all of their data to the CABS system, even if only a small fraction is CABS-oriented,” Demers says. “Or even worse, they may not be able to do any CABS billing, since they cannot differentiate between incoming traffic from a carrier and traffic they have sent out to this carrier for LNP validation which has returned on the same trunk group.”

Managing Hybrid Networks

Another problem is the hybrid network. To procure revenue, billing systems will have to evolve to accommodate volume and other demands, such as store-and-forward mechanisms that ensure message delivery—all of which will require new logic.

EHPT’s Kopostynski believes a first step is to convert duration variables in order to extend the life of existing billing systems, “at least until wholesale replacements are possible.”

That conversion from duration-based to volume-based billing will be enabled in large part through mediation systems. “Billing systems are notorious for not allowing even subtle changes in software and data,” says Imgemar Admi, technical expert in mediation at EHPT. “Modifying billing systems for even the smallest changes often affects fault management and other downstream applications.” He explains that by deploying “sniffers” to analyze and monitor data traffic on wireless networks, administrators can see how much volume is going through pipes. “The software can then convert duration into volume through algorithms and statistical modeling,” says Admi.

XML may ease the pain of making circuit-switched and IP technologies work together. Problems stem from disparate data formats and their meanings among disparate systems. XML removes these issues by providing a common syntax and a framework within which systems can describe their data formats and the meaning of data carried in those formats. Without XML-based tools, each vendor implements its own formats and data vocabularies in isolation, which requires significant custom coding to match up different data and interfaces.

“Whether wholesale minutes from long-haul carriers or digital content from Web service providers, XML models them all equally,” says Jim Culbert, CTO at MetraTech. “So part of the problem is resolved through mediation, billing and revenue-sharing applications.” He believes those that are native XML programs will ease modeling among disparate partners. In addition this will enable billing systems to consume and process data by bringing different data feeds into the billing system to rationalize and process them together.

That will be increasingly important to application infrastructure providers that need real-time revenue splits for aggregating data among multiple wholesalers, as well as to resellers and Web services companies. “Currently, most service providers do not have the flexibility to handle the broad variety of data streams coming in from their networks,” according to Culbert. “Most manually manage data dumps or use programs like Microsoft Excel.”

In fact, e-billing and e-bonding momentum is being marred by lack of standards for receiving carrier invoices and for sending orders. “Operators need to automate the carrier-to-carrier invoice workflow,” says Eric Nelson, CIO at Netifice Communications, a managed VPN services provider. He believes the biggest problem in reconciling carrier bills is inconsistency in carrier-to-carrier orders and in carrier billing formats. Currently, Nelson has to write different translation routines for each of Netifice’s partner’s bills. “If I want to examine a bill for all DSL customers we have provisioned on Rhythms,” he explains, “I have to write a different translation than I would for Covad, or Qwest or any of our other partners.”

However, that’s beginning to change, as Nelson thinks XML is proving to be ideal for describing and consolidating data. He is building an XML/BizTalk environment to enable real-time reconciliation (BizTalk is a Microsoft product designed to manage XML interfaces). “XML enables me to receive an e-bill from a carrier, and pull what I need for reconciliation without 20 people doing manual checks,” says Nelson.

The environment will also enable service provider e-bonding for service availability pre-qualification, service orders, order status and trouble tickets.

Taking these steps will improve carriers’ abilities to take part in least-cost routing, where they can quickly evaluate the least expensive routes for roaming and dynamically change call routing based on the best price and QoS they can find.

“Many of today’s carriers have in-house brokers that work to identify inefficiencies in routing and their impact on margins, so the cost of terminating traffic in other networks, as well as routing traffic within their network, is well understood,” points out John Konczal, vice president of product management at Telution, which is implementing real-time usage trending and analysis for carriers within its COMX system. “They need a billing reconciliation function that actually matches internal records to the bills received from partners. Additionally, automation of this function is imperative, due to the sheer amount of data involved.”

Determining the best price and route will only grow in importance with IP services, as operators develop the ability to tag certain data, whether email or other service, as priority, and allocate priority routing in queues. Execution of such tasks for priority delivery of text and data will warrant higher payment.

“The industry has to change its thinking a bit, as incomplete calls are just as important as completed calls in managing least-cost routing, as well as carrier-to-carrier SLAs and disputes,” according to Konczal. He says Telution’s ExecutivePortal reporting engine is designed to provide analysis tools for monitoring answer/seizure ratios (number of completed versus incomplete calls) among trunk groups. Answer/seizure ratios among partners will be a big issue, as revenues and SLAs will be based on them. “After all,” Konczal adds, “the more calls you complete, the higher the volume of data or voice calls you can price out and get revenue for.”

Billing and OSS Maturation

What is killing folks in IP-centric technology areas is lack of maturity in OSS/BSS infrastructure, according to Carl Geppert, partner and industry director with KPMG’s Americas Communications practice. It offers new margin-enhancement services focused on costs associated with IT, data analysis, testing and project control issues.

“Where processes and data might have been good for enterprise-level computing, they become real revenue drags when scaled out to the network at a carrier scale,” Geppert says. He believes carriers and service providers should be extending revenue assurance frameworks to “maximize revenue and minimize costs associated with roaming and the exchange of data messages, which will burgeon here in the next couple years.”

OSS/BSS systems will change dramatically, as they are not partitioned for carrier-type usage in IP, where services require a machine per customer per CPU. Vendors and providers alike are scrambling to upgrade enterprise systems to carrier class so they can control the fundamental cost of managing the customer.

“With newer IP services, you need Ph.D.s in computer science to provision devices; you can no longer take the same usage tracking and provisioning systems to handle costs per call and call setup,” says MetraTech’s Culbert.

Consequently, there have been huge expenditures in OSS/BSS, as providers have doled out astronomical amounts in a desperate search for high-margin services. Massive consolidation has added to the cost, as managing the sheer number of OSS/BSS, billing systems and data feeds has become almost impossible.

“Providers, it seems, are almost paralyzed, as they struggle to keep their business running,” says Verne Anton, principal analyst at Gartner Dataquest. All of them, he says, are really struggling with “future-proofing” OSS as NGN/OSS, as they branch into optical and other future technologies. “Executives are wringing their hands as they spend millions of dollars on OSS and wonder if they will have to do it again in two years,” he says.

Anton believes investment in equipment infrastructure and support systems will drop significantly as companies become conscious of their bottom lines. “Executives are still looking to improve utilization of networks through support systems, so they are spending billions of dollars, but we think that demand will decline in the short run,” he says. “There needs to be a convergence of OSS/BSS infrastructure and processes at a business level in order to have revenue assurance and offerings across product lines. Only then can you have aggregate discounting on product usage.”

It’s Elementary

“Managing cost is an elementary business process you have to master if you are going to survive,” says Netifice CIO Nelson. “The network costs are going to be the majority costs, so you have to know what you have, what you are paying for it and why.”

Nelson believes network-facing tools have evolved to assist in those responsibilities. “It is now easier to acquire information about your network than it was in switched networks, because there exists better auto-discovery,” he says. “It’s not like the old days, where you had to walk into a central office with a clipboard and start writing things on a yellow pad. Now you can see how things talk to one another.” Nelson also finds hope in the fact that there are more robust tool sets for inventory management, provisioning and activation.