Large Customer Billing and Beyond

Large customer billing and management is not the conjured media image of families chit-chatting. It is not about 30-second prime time television spots, which play more to consumer and medium business markets, but it is the prime time for carriers who contend for the business of the large customer segment.

Whether a carrier segments its large customers based on monthly revenues or another way, large customer billing and management is about accommodation through specialized support. At the very high-end are found those national and multinational customers, in the U.S. or abroad, who are handheld by carriers’ dedicated teams from the sale (new and incremental), to order processing, to network operations, to billing, and to account maintenance. These customers do not wait in ACD queues, nor do they navigate box-canyon IVR/VRUs.

High-end billing and management, for the largest of the large customers, even stretches the mega-carriers from a resource and infrastructure standpoint because each customer’s requirements are unique. But for the carriers that have the where with all to cater to the high-end customer segment, and not all do, it is about getting as much of the customer’s voice and data business as possible, thus ensuring huge chunks of the revenue stream.

Large customers’ requirements may not differ so much contextually from other market segments, but they do in practice. A key driver that has been, and will be, a consideration when customers evaluate carriers is aggregated cost per minute, which is satisfied chiefly through billing. Because of the complexities of managing a large organization, changeable hierarchy invoicing and reporting is a must. Carriers must be able to quickly respond to customers’ needs for hierarchy and rate structure changes to accommodate their business activities, such as mergers and acquisitions. The bill must be right.

As data delivery, tools for analysis, and self-management meld together for large customers, the move is away from such as canned paper billing reports. The longer after a bill cycle management reports are produced, the less the value to the customer in today’s fluid market conditions. Data requirements are on-demand, with immediate delivery, fueled by the instant gratification of the Internet and the lightweight, web-based interface tools springing up for “self-management” in the consumer markets.

While customers are exercising greater management and control of their own data networks, they also expect this insight into the public network as an extension of their own network in terms of performance monitoring and configuration management. Or, they expect greater compliance to service level agreements (SLAs) so that they do not have to be concerned with the network detail. Outsourcing of customers’ internal network management to capable carriers is in evidence, publicized in relationships between the likes of IBM and AT&T.

Large customers expect to have future requirements met, and to be heard on an on-going basis. Whether this directly or indirectly impacts a carrier’s product development priorities depends on the carrier, and the approach used to understand the large customer segment.

Segmentation and Support

Large customer segmentation differs from carrier to carrier, but generally falls around $10,000 in monthly revenues. Sprint segments its large customers by $10,000 per month or by a customer that has 100 or more employees. But the very high-end is usually segmented even further, and is usually based on a carrier’s 50 or 100 largest customers. AT&T’s high-end segment is uniquely supported through its global services division, and Sprint and MCI have always applied exclusive focus under different divisional names for their high-end base.

Most carriers use dedicated teams so that the contact with large customers has continuity and accountability. Infrastructure support groups, such as billing, are sometimes dedicated, sometimes virtual. In the everyday milieu, these folks are responsible for understanding and fulfilling customer SLAs and requirements.

“Focus groups have been a good mechanism for us to understand our customers,” says Marilyn Moore, vice president of customer management at ICG, “in addition to the everyday knowledge we acquire.” ICG does not hold a variety of routine focus groups, which places a greater reliance on its monthly customer satisfaction surveys and on timely feedback from the account teams to the supporting infrastructure groups.

“We use very specific focus groups,” says Ron Sieve, group manager for invoicing and reporting at Sprint, “such as our billing forum held three times a year by our invoice and reporting development group.” Focus groups are intended to cover any and all topics - issues, desires, the way information is communicated on the bill, and the kinds of reports.

The Billing Costs to Compete

High-end customer billing and management for the likes of Sears and GE are the bailiwick of the mega-carriers -- the Big 3 IXCs and the largest LECs. Many large customers -- universities, major hospitals, local governments -- are supportable by carriers such as ICG, with limitations based on geographic footprint of the customer, carrier serving area, call volumes, and service mix.

While the high-end segment is a seductive acquisition target for most carriers even if they espouse a different target market, the cost to compete is high. No matter how badly the mega-carriers would like to sweep it under the rug, much of the high-end customer billing is still supported with manual effort, or automation done on a customer specific basis. End-to-end support is often a hybrid of the two. These carriers still seek to automate the billing and related processes because they engage hundreds of people to support the high-end. While full automation has been elusive, they have cut costs through process efficiencies and partial automation.

Large customers negotiate price very aggressively - driving to the least aggregated cost per minute on voice services. Margins on the high-end have been notoriously difficult to nail down because of the massive support costs and the desire to win business regardless. Now carriers such as Sprint are doing a better job assessing whether it makes financial sense to ink a contract with a high-end customer.

According to Lynn Miller, manager for customer service at Sprint, “Various infrastructure groups work with sales and pricing in the up front negotiation process to ensure the contract conforms to systems capabilities, and that it meets margin requirements.”

The big IXCs are with little question the most adept at handling the high-end and special customer arrangements. Because they have had to compete fiercely, and longer, their systems infrastructure and processes are generally more refined than those of the LECs are, and able to handle processing capacities and nuances new entrants can’t.

“The LECs still struggle with summary billing,” says Christopher Finn, principal, Communications Sector Consulting Practice at IBM Global Services, “and continue to seek post-process approaches so they can perform cross-service discounting beyond the basic summary billing.” According to Finn, the LECs try to replicate what the big IXCs have done in regards to bundling for large customers, cognizant or not of how expensive it has been to the IXCs.

“The ‘Holy Grail’ to all carriers, regardless of vertical market segment,” says Brian Hayward, fund manager of Invesco’s Worldwide Communications Fund, “is to offer bundled services, even if the concept of one stop shopping is tired and has gotten too much air time.” Within limitations, and given the obstacles dealing with the ILECs, it is something the CLECs are doing with reasonable success.

“Regardless of customer segment, we are able to use the same billing capabilities,” says ICG’s Moore. The CLECs and new entrants do not have to deal with the post-process bundling agony of the ILECs, or the up-front manual or customer specific automation of the IXCs. Thus, support costs are lower than at the IXCs and the ILECs, thanks in part to software packages and no legacy systems environment to hobble them. Conversely, this does not address the high-end of the large customer segment - in which they are not yet ubiquitous players.

Billing the Key

“Accurate billing just never seems to get off the large customer’s radar scope,” says IBM’s Finn, “especially at the very high-end where a lot of manual processing occurs.” Michelle Davis, telecom manager at Fluor Corporation, a global engineering construction company, reports that staff members spend time spot-checking the accuracy of bills.

In recent years, large customers and government agencies have used auditors to discover overcharges. Presumably, this has put a strain on the carrier to customer relationship, especially when the over-billing has been for an extended period of time, and in the thousands of dollars per month. However, customers don’t want to have a lot of staff dedicated to managing the telecom relationship, especially the billing component. Customers want carriers to implicitly help them reduce/redirect budget.

“Little has changed in this regard,” says Invesco’s Hayward, “so the information must be accurate, timely and intelligible, and customers only want information they need.” When a carrier says they can give a customer a report to download from the web each month, or at least on CD-ROM, as opposed to a customer doing it manually or plowing through stacks of paper, it strikes a resonant chord, adds IBM’s Finn.

One-off contracts (special customer arrangements, special pricing arrangements, bulk service agreements) are widely used to achieve a negotiated cost per minute aggregated across multiple services and customer locations. The big IXCs have been best suited to do this in the past with true hierarchy-based billing and contributory and eligible (C&E) discounting capabilities, the lifeblood of cross-service discounting. For the high-end they still are best suited. One-off contracts with customer specific pricing are satisfied in billing chiefly through means such as Sprint’s plan types. Plan types set an indicator specific to that customer which is detected as early as call rating, and subsequently threads its way through a myriad of billing tables performing unique look-ups in order to deliver the pricing the customer signed up for.

While the CLECs may be more nimble than even the IXCs, they still struggle from volume issues on UNIX and NT platforms, and AS/400, regardless of the claims of the representative software and hardware vendors. CLECs are prudent to closely examine the benchmark claims made by various vendors because these are generally performed under pristine lab conditions, and the composition of the call records does not fluctuate widely between call types, call rates, call jurisdictions, call lengths, and other processing and rating determinants.

“No,” says IBM’s Finn, “the CLECs aren’t ready to handle Fortune 100 customers because they are working from software packages set up for less complex accounts, and it also takes a while to put together methods and procedures to handle those customers, especially when the processing becomes partially manual. CLECs are not geared up to take on some of the costs and processes required to support those customers yet.” CLECs may serve universities, individual and local government agencies, some airline business, etc., but not national accounts with thousand of locations (and hierarchy nodes) such as Wal-Mart. This is based in part on regional CLEC coverage, but also on billing limitations. It is hard to replicate how large customers are handled in IXC hierarchy billing, let alone in LEC summary billing. It would be hard for a customer to migrate in large part to a CLEC because the CLEC would have difficulty replicating all the special pricing and nuances of existing contracts. An apples-to-apples comparison would be very hard for a CLEC.

Are the CLECs behind on the eight ball entirely in the high-end market? No. Where they are faring better is in the wholesale market leasing access and capacity, which simplifies pricing and clips the need to process a network worth of CDRs. For ICG this includes marketing network capacity and services to national ISPs and IXCs, such as special access, in addition to commercial telecom.

Of Hierarchies and Discounting

“We feel IPS is still a customer key driver,” say Sprint’s Sieve. “It can bill all of our voice products, and frame relay will be in the first release of Sprint ION - which we are testing now.” IPS was built to support large customers through its complex, 6-level hierarchy structure, discounting capability and plan types. It supports multiple entities under one umbrella, multiple services, locations, and customers on a single bill. Frame relay and other data products will benefit from cross-service C&E capability.

C&E structures allow the grouping of pricing events based on attributes, and the summarization of charges and/or quantities in order to determine the discount percentage/amount and then apply the discount. Within a C&E structure, a capable carrier can support the ability to apply different discounts to eligible usage based on any product attribute, customer attribute, hierarchical grouping or specific charge.

MCI (on the MCI side of the MCI WorldCom business) and AT&T have variously used anywhere from 3-level to 9-level billing hierarchies for large customers and the high-end. All of the big IXC’s hierarchies are both invoicing and reporting capable, but of the three, according to IBM’s Finn, Sprint’s IPS does it the cleanest with little or no post-processing. In a two thousand-node hierarchy configuration there can be numerous invoice points, but it can be overlaid with a reporting hierarchy to ensure all the sub-tended nodes receive billing data.

The LECs continue to be locked into summary billing, according to IBM’s Finn. “While some LECs have tried to do things in a green field way out of territory, it is hard to separate the network business and be a non-facilities based CLEC trying to buy elements. It has not proved to be very feasible.” Part of the limitation is working with base systems with a post-process consolidation step trying to do more cross-service billing than summary, or non-geographic, billing allows.

Characteristically a full infrastructure upgrade is too painful for the big LECs. They have achieved limited successes only, regardless of claims about investment and how big are the successes. “When the large LECs try to make the big bang upgrade it is like trying to boil the ocean,” says IBM’s Finn.

No LEC mega-carriers are currently trying to bite off full legacy system replacements, but while trying to bring more into the architecture, they are manipulating core to middleware infrastructure to limit the impact in interfacing to legacy applications. But the fact remains that all the inputs needed to do LEC consolidated billing are horrendous -- undocumented, or unknown.

Other Services

Even national footprint wireless carriers AT&T and Sprint don’t have hierarchy-based billing, and are not cohesive with wireline billing. Generally, wireless at best treats large customers as affinity groups by offering a specific rate plan/promotion. Purchases are done at the individual level. Very little aggregated data delivery or analysis exists in wireless, but the pressure has not been there to provide a consolidated view to corporate customers. This could change, and certainly will when folded into mainstream billing. In the meantime, users at a large customer order individually, pay individually, expense individually, credit card bill (line item) individually. Preferred paging providers such as Mobilcom are the exception, where corporate billing is available.

In the ISP/ITSP world, pricing has been simple up to this point. But more and more it will have to be related to how the network is used because ISP/ITSPs can’t do comprehensive network management until they effectively tie pricing to the utilization of network resources. Trying to figure out how to get statistics or usage records out of the routers or other network elements has not been easy because they were not designed to produce CDR streams or anything similar, let alone to collect it for further rating and pricing. With data coming out of the network elements, a network mediation layer that can distribute the data appropriately and get it out toward the bill is part of the biggest obstacle. Although work is progressing, most billing systems developed to date may not be able to handle the capacity of a large customer, or aggregated volumes, as mentioned before. To do hierarchy and usage-based billing will require more horsepower than doing billing on NT servers for the high-end. ISPs catering to high-end customers may face a continuing struggle because they started out with low costing infrastructure.

Information Access and Interface

To consider billing further, carriers do not allow self-service, on-line billing hierarchy changes even if customers desire it. The lack of priority is carrier apprehension regarding changes in customer configurations as bill cycle cut-off approaches. Where does the traffic belong, and how should the C&E discounting prorate? Also, concerns abound that the customer may do things wrong, or that they could manipulate the hierarchy within certain pricing schemes (such as roll-ups and spreads where a customer could create “you owe me” situations at certain hierarchy nodes).

Sprint’s Miller points out that, “More hierarchy changes occur on a monthly basis as a customer implements service, then trails away as the customer becomes entrenched.” Electronic tools alleviate some of the need for making hierarchy changes since they can satisfy slice and dice reporting requirements.

Carriers will be deploying more advanced tools for analyzing data. Sprint’s FONVIEW will have a DATAVIEW counterpart for frame relay. Like FONVIEW, it will have pre-indexed and ad hoc reports. FONVIEW is undergoing updates to be more suited for distributed, networked environments (NT and W95/98). Customers from fewer than 10,000 CDRs to ones with 14,000,000 CDRs per month are FONVIEW users.

Most carriers are implementing basic e-Business capabilities so customers can make queries over the web, directly to the carrier’s datamart or data warehouse, rather than waiting to get reports as in the past. IBM’s Finn says, “This type of functionality is moving up-market, but some of the larger customers are pretty set in their ways, rendering it difficult to get them to move to some of these more mass markets capabilities, but it is evolving.” Fluor Corporation’s Davis attests to this by pointing out that Fluor’s accounting is still very paper-driven.

“A more immediate view or access to the data would be very beneficial,” says Mary Jo Combs, director of telecom at Summa Health Systems. Carriers are coming up with lighter weight ways of transferring raw data, such as packing it and doing an FTP transfer as opposed to some of the means used in the past to get customers their base data.

The proprietary applications carriers have developed and rolled out previously (such as MCI’s Customer Direct), putting code on the customers’ desktops, are now rolling towards the web. Web-enabled capability speeds time to market, is more cost efficient with better economies of scale, reduces distribution and support headaches, and eases the customer’s entry.

Implementing customer relationship management means sharing commonality between call centers, the web, and the account team. “Consider companies such as Siebel, Vantive and Corepoint who have traditional call center applications,” says IBM’s Finn. “If the screen functionality is translated to the web then you start to get some cohesion between the web and the call center, without having to replicate databases and applications for different customer channels.”

Self-Management and Network Control

While carriers are trying to implement packages to do billing on a more cost-effective basis, they are also looking at how the reporting, ordering and customer facing functions go on the web in a lightweight fashion. Carriers are trying to drive toward a “self-service” model on basic network management, service ordering, and statistical processing and analysis. ICG is looking at implementing capability for select customer to enter and track their trouble tickets.

Carriers are trying to figure out how customers use advanced services to determine usage-based or statistics-based pricing for these customers. This is important because carriers must start linking network management with the pricing and start linking costs with what they are charging for the advanced services. Projects are underway to better understand network costs at an aggregate level, and to understand usage at a customer level so that managing profitability is more science, less guesswork. “The industry is seeing a lot of business intelligence and data warehousing being layered on top of this in order to get those views so that pricing can be driven from a cost basis, as opposed to wetting your finger and holding it in the wind,” says Finn.

From a network management view, while some customers are pushing away from carriers, some are outsourcing more responsibility to the carrier. Some customers are looking to carriers for the piece parts - with more active control. Assume a customer manages its network via Openview or Tivoli - there’s a good chance they want to extend that management view through to the carrier’s network.

But a trend is noted that if carriers can manage the customer’s network and meet SLAs then it is pretty attractive for customers to use the carrier for network management. IBM has just done this with AT&T in all the swapping that went on. IBM has outsourced a lot of its own internal networking needs to AT&T under this arrangement. Where a customer may have asked for busy hour statistics before, now they are saying, “Just meet these service levels and omit the detail - unless things break.”

Product Development

How large customers impact a carrier’s product development priorities is likely relative to the amount of revenue generated for the carrier’s product group. High-end customers may have direct input, but much of it is indirect - communicated through sales and marketing. Other large customers, and smaller customers, are aggregated in a prioritization pool.

Sprint’s Miller says, “Our focus groups can affect the product development priorities and changes are made based on their input. Of course, though, all projects are prioritized by executives to stay in line with what is most important.” Sprint and other carriers also use metrics tracked by customer service organizations on customer queries. Carriers also receive feedback from account teams on issues and challenges high-end customers are facing - and ideally this is communicated internally to infrastructure groups such as billing development.

Future Requirements and Pipe Dreams

Customers continue to want access to more and better information and control over their telecommunications business. While the average consumer can go on a carrier’s web site today and make changes, turn off service, etc., without waiting, that ease of use has not come to the high-end customer set. They would like to have a greater ability to place queries and get information from the carrier in a fairly lightweight way, while having the information easily distributed internally.

Payment processing hasn’t seen great advances in the large customer segment either, although electronic billing and payment is expanding rapidly in the consumer through medium business segments. More than any other segment, large customers are slow payers, because of the amount of data to review for accounting and accuracy reasons, which elongates the carrier’s receivables cycle and skews the A/R and quick ratios. Any improvement, even a few days, would spell relief to the carrier CFO, and may reduce customer-annoying late fees.

The inability to link service level performance to billing must be addressed. Although service levels are linked to billing in Europe, it will take greater competition to make it happen in the U.S. As voice products become less the issue, and data and IP products become more the issue, customers will become more focused on level of service until it proves tenable and hurdles an industry wide level of acceptability. Expect service level performance to billing to be a competitive differentiator.

When it comes to the large customer segment, especially the high-end, don’t expect specialized attention to dwindle. If anything, it will get competitively more intense. Large customers are the prime time of telecommunications competition. But an army of attention is not enough - the carriers must continue to provide large customers with lightweight, self-service tools, including more advanced versions of web-based tools already springing up in the consumer markets, for queries, data delivery and reports, and network management.

Frank Slavick is a telecommunications consultant based in Denver, Colorado, specializing in product development, new business development, and billing and customer care. He can be reached at 303/554-0958, or at fslavick@earthlink.net.