SLAs for Service Bureaus—The Outsourcing Safety Net

Billing sometimes may feel like a walk on a tightrope—a fine line between your revenue stream and all the possible mishaps. Taking billing outside the company could be even more intimidating, if service level agreements didn’t create a safety net in the billing service bureau agreements. A reliable net can be woven with thoroughly deliberated contracts.

To achieve this, all of a service provider’s business processes must first document business processes to the letter. And in a highly competitive environment, those processes change rapidly. It must then predict which outsourcing provider most closely matches its business plans—today and next year. Today’s service bureaus must not only adapt to the quick-change model, but design contracts to create adequate compensation and realistic development schedules. The tug-of-war between contract flexibility and very specific service level agreements challenges even the most seasoned negotiators.

Joe Farrell, vice president of sales and marketing for EUR Datacenter, says that new entrants and established carriers bring very different challenges to a negotiating table. Entrants in the CLEC marketplace often have muddied expectations of a service bureau, he says. Decision-makers often don’t take the time nor have the wherewithal to investigate what the business really needs or what it will actually receive from the service bureau. “They have an impression that this particular service bureau is doing work for 10 other CLECs, therefore anything they ever need to be done is going to be done, without really looking to see,” Farrell says. “The biggest problem area, from a service bureau standpoint, in working with new entrants, is that they don’t have expertise.”

Even though the contract makes clear provisions, Farrell says, because of enthusiasm, excitement and pressure to get to market, new entrants may overlook specifics. “We had a situation with a client,” he says, “where we very specifically told the management group, ‘What you see is what you’re going to get.’ Well, as soon as the contracts were signed, all of a sudden they realized that they had not spent the proper amount of time and energy to review it.” Because EUR was working with several other CLECs, one thing the carrier had expected was an interface into the ILEC for order entry purposes. Another was an automatic interface into a calling card provider for the plastic, which provides complications, Farrell says, because there are many different calling card suppliers—all with different formats, methods and procedures for data transmission for calling card plastic requests.

Definitions present another challenge. In a service provider’s eyes, customer care may incorporate handling return mail, correct address inquiries on returned correspondence and delinquent account collection; but in reality, the service bureau might only process outgoing mail, provide address verification within its account entry screens and generate dunning letters. These kinds of expectations, not clearly communicated, make life difficult for both parties.

“In contrast, an IXC or an RBOC, or any other carrier that’s been in business for a couple of years will take a totally different approach to their contractual relationship with a service bureau,” Farrell says. “Typically they are much more practical and realistic, because they have experience in the area.” Where new entrants might want a five-hour turnaround time, experienced carriers expect bill calculation and production to require several days. They know that data feeds from the switch—as well as messages from connecting companies, transactions from the directory assistance handler and updated taxation information from a third-party provider—slow the assembly process and create a greater need for editing.

Manufacturing the safety net

“The billing service provider doesn’t have a finger on the pulse, it has a hand on the switch. So the performance of the billing service, and billing, customer care and OSS provider is critical because it has such a direct and drastic impact on the company’s revenues,” says attorney Jim Kearns, a partner at Bryan Cave, LLP.

Roy Heggland, senior vice president and legal counsel for Convergys, says that because of the size and complexity of its clients, the contract negotiation process usually takes more than six months. A typical Convergys contract extends for 5 to 7 years and runs 150 pages, with schedules, SLAs and descriptions of the operating environment. Other service bureaus report less—3- to 20-page contracts, with negotiations sometimes taking less than a month. Contract templates and e-mail facilitate the process.

Mark Rein, IT director at Utilicom Networks, a Saville Systems customer, reports that his company’s service bureau contract negotiation with Saville took only three days. Utilicom had signed a letter of intent and given money to a competing vendor, but pulled out of the deal when the vendor’s counsel changed the wording of the entire contract at the last minute. “I went to Saville on Monday, told [the salesman] how much we wanted to pay per bill and what our expectations were, and signed the contract on Wednesday,” he says.

Farrell says such abbreviated negotiations are uncommon because of granular SLAs. With outsourcing, he says, a company will want very, very tight service level agreements that in many cases would be totally unrealistic if the process were internal. The agreements establish longevity and history in the relationship and provide consistency through management turnover, reducing upheaval when one person thinks five days is reasonable and his successor thinks it should be done in five hours. Once the contract is finalized, frequently a legalistic or numeric relationship moves to more of a partnership arrangement. “It’s good to have the DMoQs [demonstrated measures of quality] so we all know what we’re shooting for and then we can do periodic reports. SLA review makes a great report card template,” Farrell says. SLAs and DMoQs are reviewed for monthly or quarterly report cards or annual audits, but contract specifics are buried somewhere under event rating and bill cycling in day-to-day procedures.

Kearns says that most service bureau contracts also address scope changes during or after implementation. “The notion that business plans will change is considered in the negotiation process,” he says. “Most service bureaus want to be a partner and work with [clients] to accommodate enhancement processes, instead of amending the contract each time.”

Clause renegotiation procedures and customization timetables are built into contracts, Kearns says, although as a practical matter people generally respond sooner than required. Sometimes development work starts with a letter of intent while the business processes are still being ironed out, although it is very rare that the initial implementation begins during contract negotiation.

When something goes wrong…

A universal point of contention in service bureau-carrier relationships remains liability for incorrect bills. Billing is not a high-return operation—usually some miniscule fraction of a cent per bill produced—which limits the service bureau’s ability to absorb the risk of bill run results. The liability issue shows up in contract indemnification clauses, Kearns says. While no telecom company expects its service bureau to assume the risk of bill collectability, some have incredible expectations that service bureaus will guarantee lost income due to bill errors. “Some of the smaller service bureaus will do that, basically on the grounds that ‘we want the business, and if we had a catastrophe, since we are in a mission-critical position, the client is not likely to put us in a position of bankruptcy over this issue,’ ” he says. “What carriers want, as a practical matter, is the bills done right. It can become something of an esoteric argument, to which they consent on paper because they don’t think anything will ever come of it. Others don’t want to sign something that might conceivably happen. It comes down to how seriously people take their contracts.”

There are some middle-ground solutions. One is a dollar cap, in which a service bureau assumes liability up to a fixed monetary amount. In another scenario, the service bureau assumes responsibility for damages up to one month’s service bureau fees. The arrangement Kearns favors, though, requires the service bureau to rebill correctly, no matter what the costs. However, both parties generally have extensive bill verification procedures to avoid a problem.

DCA Services, an Oklahoma City-based service bureau, has a 72-point checklist for bill verification. Evolving from 12 years of billing experience, the checklist provides line items to check and what to compare each point against. It includes such aesthetic things as addresses fitting properly in envelope windows, sequential page numbers, correct verbiage above and below perforation, and postal net and scan line bar codes. It also requires signoff on rate calculation, discount and finance charge computation, and credit and debit notation. Increasingly, companies must focus on automatic transaction summaries listed on bills as businesses support more types of transactions.

Rick Nagel, DCA’s vice president for sales and marketing, says there are some issues that the service bureau should not negotiate. “We draft our contracts to not assume liability for tax computation. The tax table providers don’t assume liability, why should we?” he says. And before a service bureau agrees to pay anything, clear liability provisions and responsibilities should be outlined.

System failures and business process breakdowns must be detected and resolved as quickly as possible. Many companies contract third-party auditing firms to aid this process. Leo Montgomery, partner at Ernst & Young, says that to avoid being inundated by each carrier’s auditors, each service bureau has its own auditor file an SAS-70 report for its clients, which both outlines and tests the controls and security in its operations.

A typical audits examines what controls exist—and if they are being run effectively. Calls that are being accumulated on tape are checked for both accuracy and inclusiveness. It tests bill verification controls, system access security, change order controls (whether the system or data can be modified, and by whom), error management and transaction integrity. The auditor also analyzes physical access to the system, environmental and preventative controls, disaster recovery capabilities and other service level agreement provisions.

Third-party audits and clear control procedures reduce fuzziness in the line between service bureau and service provider responsibilities and indemnify the service bureau for errors or lack of control on the providers’ part. Sometimes, the audit review can signal the service provider that it has overlooked an internal control procedure nobody had even considered.

Who owns the net?

Contract negotiations should address other issues of ownership as well. Functionality customizations are one area of concern. If the customer owns them, questions about developing the same features for other clients needs to be resolved. If the service provider pays for the customization, a compensation schedule for functionality rollout to the service bureau’s other clients should be established. Kearns thinks it’s advantageous for the service bureau to own and the enhancements so that they are available to each of its clients. That approach, however, lessens the advantage for competing providers that use the same service bureau.

Customer information and data records prompt additional ownership discussions. The companies need to determine responsibility for the data integrity, security and backups, as well as outline proprietary and confidentiality measures.

Ownership of the systems should be tackled in contract negotiation as well. Either party could own software and hardware components, and their respective insurance policies and maintenance. Each of these considerations further depends upon where each piece is to be located. Location clauses are even more important in the increasingly popular facilities management arrangements, which satisfy those squeamish about completely internal billing departments and fully outsourced billing.

Contractual relationship ownership for the system’s different pieces and interfaces must also be determined. This particular area, Farrell says, can become complicated. “If a customer is reselling Frontier, and using SCC for their 911 interface, a Q-Tel 7000 database for message traffic control, and Vertex for tax information, we have databases loaded into our computer that we are accessing on behalf on the client. If the client owns the relationship, the client then has to negotiate the error resolution with the supplier. If the service bureau owns the relationship, the service bureau accepts that responsibility.”

Heggland says Convergys assumes liability for most of the third-party relationships for its clients. The exceptions are the tax information vendor relationships and a handful of print vendor relationships that very large clients have negotiated on their own. In the printing cases, Convergys functionally administers the relationship the same way, but assumes no liability.

Usually, Farrell says, whether the service bureau owns the relationship or not, it takes responsibility for it. This poses no problem as long the contract’s SLAs clearly define exceptions. Errors and update delays due to third-party suppliers need to be excluded from the service bureau’s liabilities.

Using sturdy materials

The contract also should address disaster recovery in sharp detail, Farrell says. Some service providers expect a continuous operation whereby the service bureau automatically writes redundant transactions to another data center for instantaneous disaster recovery. For less cost, service providers can be assured that the service bureau has offsite tapes and will recover as quickly as possible. Between the two are a number of other possible measures, and both sides need to have very clear expectations, Farrell says.

Utilicom’s Rein says that disaster recovery and environmental security were major considerations in his service bureau decision. He recommends proof of workable disaster recovery procedures, a detailed plan and—hold your breath—a visit to the data center. Rein discovered that one company he initially considered “didn’t have a real data center. They had Compaq computers sitting on cinderblocks,” he says. “That was it. There was no fire suppression system. There were no environmental controls.”

Response times

Expected response times consume another phase in contract negotiation, Farrell says. “How long does it take to get an answer from the [system] once I’ve queried it? That is increasingly more complicated, because there are so many different system interconnections. How do you define who is responsible for what?

“For instance, a customer-care transaction touches the switch, the credit bureau, the plant systems, an MSAG database for address validation, and a marketing database. Some of these are contained within our servers, some are external with real-time data feeds. So when you’re going through an order entry process, how do you define what the response time is? That gets to be real tricky. Most people understand and appreciate that certain transactions will take longer than others.”

But system response times and customers’ expectations have to meet. As a director of IT, Rein says, “Your customers are calling and saying, ‘I want a credit for this call,’ and if you can’t look at the call, you’re out of business, because they are going to think that you’re a second rate company. Your life is running on your billing system.”

Standing on the high-rise platform, waiting and watching the safety net’s construction may take a little extra time, but knowing that it’s there, even if you don’t look down at it often, is reassuring. Yet even with the contract negotiation time, service bureaus generally get you to market faster than building your own department.

Additional Considerations for Negotiating a Service Bureau Contract

Service bureaus often offer pricing per account, per bill or per event, with a bundle of included services. However, larger carriers negotiate more element-specific pricing, says Joe Farrell of EUR Datacenter. There are a number of things to be considered in each model:
? The price for duplicate bills for an account
? The price for accounts with multiple lines
? The cost for end users requesting a more granular level of call detail
? The charges for credit verification; a regional MSAG database interface; data feeds such as LIDB and CARE
? Responsibility for costs incurred during error resolution
? The price for delinquent notice preparation, printing and mailing; and for a follow-up phone call made in addition to the dunning letter
? The charges for and the amount of information to be passed on to a collection agency regarding delinquent accounts
? The cost for differing payment processing options, including lockbox, credit card, EBPP, EFT or cash processing
? The charges and responsibilities for handling return mail and written inquiries
? The procedures and costs for mutual compensation/interconnect/CABS billing, for facilities-based service providers
? The differences in charges for wholesale and retail customers
? The amount of EBPP cost savings passed on to the service provider
? The cost of deliverables such as paper, toner and envelopes
? Any additional charges for two-sided printing
? The charges for bill handling and postal sorting
? The acceptable holidays, when considering mailing and bill handling
? The times the help desk is available; how much and what kind of help is included
? The management chain of command for problems
? Established procedures for disputes and arbitration
? And, in this high-turnover environment, agreements about swiping one another’s employees.