September/October 2001

  
The Network-Based Call Center Opportunity

BY ERIK J. LAURENCE
 

Industry analysts are bullish on the NCC and expect tremendous and unprecedented growth in NCCs. Ovum, for example, predicts that by 2005, fully 35 percent of all agent seats will be served by NCCs versus less than two percent today. With more than six million agent seats worldwide today and a total of nearly 12 million agent seats forecast for 2005, that translates into growth of network-based seats from 120,000 today to over four million by 2005. The best way to explain what an NCC is may be to begin by explaining what it is not: It is not a premise-based call center and it is not an outsourced call center.

PREMISE-BASED CALL CENTER
A premise-based call center typically consists of a room full of agents. Each agent has a desk, a chair, a computer terminal, and a phone. Down the hall or in the back room -- or at least somewhere on the premises, there is a room full of equipment. Prominent in that room is the automatic call distributor (ACD) -- essentially a telephone switch that queues and distributes incoming calls. Also in the room is an interactive voice response (IVR) system that allows callers to key in and retrieve basic information via their phones' DTMF keypads in lieu of speaking to, or while waiting to speak to, an agent. Nearby would be a customer relationship management (CRM) system enabling the agents to pull up records on the callers and/or create new ones about them. Some newer centers also have systems that support Web chat, e-mail distribution, and maybe even voice and video over IP (to enable live voice and video communications through your computer).

Down the hall from that room sits the MIS and telecom department -- the wizards who keep it all together and running. Unseen are the teams of highly paid consultants and systems integrators who managed to get all of these systems talking to each other using computer-telephony integration (CTI) -- a complex technology required to integrate the disparate worlds of computers with telephone systems. Unfortunately, the complexity and cost of CTI has often meant incomplete integration between call centers. We've all had some experience with the frustration it causes: We're often asked to key in our account number while waiting for an agent, presumably to "ensure speedier service." When we are finally connected, the agent asks again for the same information. More often than not, this frequent occurrence is caused by a poorly implemented CTI project, resulting in blank screen pops at the agent's desktop.

OUTSOURCED CALL CENTER
An outsourced call center is very similar to a premise-based call center. In fact, it essentially is a premise-based call center, complete with agents, ACDs, IVRs, CRM, and the rest, only at someone else's premises. Call center outsourcers effectively run entire call centers on behalf of their clients, terminating their clients' toll-free and other call traffic on their own premises. The centers are typically staffed with agents trained to represent the client's product or service as if they were employed directly by the client (which usually they are not).

THE RISE OF THE NETWORK-BASED CALL CENTER
At the heart of the network-based call center is an ACD, only this ACD lives in a service provider's network rather than on a company's premises. The trick behind the NCC is that it's not based on legacy circuit-switched technology -- instead, it's based on all-IP open computing systems. By removing the circuits from the equation, the only thing between the service provider and the agents is a managed IP network. This makes hosting an all-IP NCC about as straightforward as hosting a company's Web site. It also means that the agents can essentially sit anywhere. Thanks to the inherent switching of the all-IP environment, agents no longer need to all sit at a central location -- they can be distributed at multiple regional locations or even work from home with DSL or cable modem connections.

Best of all, an all-IP call center becomes a unified contact center, handling not only telephone calls and IVR, but also Web chat, browser collaboration, VoIP, video, voice mail, e-mail, and more. The motivation for enterprises to go the route of the NCC instead of deploying premise-based call center equipment is considerable. The benefits include:

Capital expenditure avoidance. Instead of large upfront purchases of capital equipment, enterprises are able to sign up on a month-by-month basis, paying either by the agent or by usage (more on this later) for the services they need.

Focus on core business, not complex telecom systems. Deployment and maintenance of call center equipment is complex and labor intensive, and large telecom and IT departments are required for the task. With today's trend toward an ever-sharper focus on core competencies, businesses should focus on what they do best and not have to be in the business of building and running their own "phone company."

Call centers becoming multi-channel "contact centers." There is dramatic change underfoot in the way customers want to contact the companies they do business with. Telephone communications are being augmented by Web-based chat, e-mail, and even voice and video over IP. This means that call centers in enterprises all over the world are rethinking the way they handle customer contacts, and they're all thinking about how they're going to enhance existing equipment and when they should replace legacy systems. In short, they need to make major changes anyway, so why not consider NCC?

Capacity management. One of the strongest benefits of NCC is in the area of capacity management, especially as it applies to these new contact methods. Enterprises simply don't know what the new mix of contact methods will be -- to a certain extent no one knows what it will be. Phone calls vs. Web chat vs. e-mail vs. VoIP vs. video vs. voice mail -- who can figure out the right amount of capital equipment for enterprises to buy in each of these areas? That's where the NCC comes in. With NCC, customers pay for the capacity they need as they go.

Multi-site call centers. NCCs can provide a single queue for an entire enterprise independent of the number of sites or the location of agents. This capability maximizes agent efficiency while significantly reducing the technological complexity of multiple sites with layered, independent queues.

Call centers for small businesses. Call centers are becoming more essential to every business, and for small businesses this is no exception. While operating a call center is complex for large businesses, it especially strains the resources of smaller enterprises. This is an area where the benefits of NCC are truly compelling.

IP delivery of phone calls. One of the highest costs in a typical call center is the cost of the calls themselves. A domestic toll-free circuit-switched 800 call in the U.S., typically delivered over T1 or E1 lines, can cost anywhere from $0.07 to $0.30 per minute. Outside of the U.S., the costs can be higher. The same toll-free 800 calls, routed via VoIP gateways, can be delivered to NCC agents in the IP domain for much less -- in some cases even as little as $0.02 to $0.04 per minute. But even if it's only a penny or two less per minute, the total cost savings can be very substantial.

The NCC is also a compelling opportunity for service providers. For telcos, the NCC represents an opportunity to create a natural complement to the toll-free service they already offer to call centers. In fact, given that call centers currently spend more than $20 billion dollars annually on toll-free service, it is strongly in the telco's interest to find a value-added, "sticky" service that can enhance loyalty among their existing call center customers. And for all service providers, the NCC offers a new revenue stream with a strong business case. One analysis shows an ROI on NCC investment to the service provider of higher than 90 percent.

NCCs present a tremendous opportunity for tomorrow's call centers. Companies benefit by focusing on their core businesses while enjoying expanded functionality at a low cost in their contact centers. Service providers benefit not only by enjoying a new, high growth revenue stream, but also by playing a leadership role in this rapidly expanding market.

Erik J. Laurence is vice president, marketing and business development for CosmoCom. CosmoCom offers a new generation of IP-based call center platforms that are fully unified in supporting standard voice telephony and Internet-based multimedia communications.

[ Return To The September/October 2001 Table Of Contents ]

BY ERIK J. LAURENCE

Given that a very significant portion of all call center agent seats in 2000 were provided by outsourcers, we can safely assume that the outsourcing model provides a sufficiently beneficial business case to enterprises. Therefore, we will approach the business case to the enterprise of the network-based call center (NCC) by starting with the market dynamics of the outsourced call center.

A typical outsourced call center generates revenues of roughly $50,000 per agent seat per year, according to analyst reports of public call center outsourcing companies. This equates to about $4,200 per agent seat per month. For this amount, an outsourced call center provides a complete package: Personnel (i.e., the agents themselves), real estate (i.e., the building where the agents sit), automatic call distribution (ACD), basic interactive voice response (IVR), basic customer relationship management (CRM), and telephony.

An NCC would typically provide everything in the above package except for the agents and the building where they sit. An NCC service provider may or may not bundle telephony with its service, so for the sake of a clear comparison, let's look at a scenario where telephony is not included. By calculating the monthly costs of these items and subtracting them from $4,200, we arrive at a figure of approximately $1,250 per agent per month. $1,250 is the maximum amount that, all else being equal vis--vis an outsourced solution, the enterprise should be willing to pay an NCC service provider for ACD, basic IVR, and basic CRM. Our calculations are shown in Figure 1. Note that in all cases, we have tried to make conservative estimates; for example, while we use a telephony rate of 10 cents per minute, many outsourcers pay as little as seven cents or eight cents per minute -- at those rates, the picture looks even better.

CORROBORATION FROM PREMISE-BASED VIEWPOINT
Industry consultant Primary Matters calculates $1,300 as the monthly amount an enterprise would be willing to pay per agent for an NCC comprising ACD, IVR and CRM functions. Their calculation was based on a comparison of the costs an enterprise would have to pay to build and maintain its own call center. As Primary Matters approached this estimate from a viewpoint entirely different from our own, the result serves to reinforce and corroborate our $1,250 figure.

To understand how much an enterprise would be willing to pay for an ACD and basic IVR solution only, we must remove the CRM function from our $1,250 figure. If we assume $300 per agent per month for this piece, a generous number given that it represents only the most basic of CRM functions, we come up with roughly $950 per agent per month. This is the maximum amount that, all else being equal, the enterprise should be willing to pay an NCC for just the ACD and basic IVR function. That means that at any price lower than $950 per agent per month, the enterprise saves money by choosing NCC over outsourcing. This calculation is also shown in Figure 2.

But all else is not equal. In fact, up until now, we have described in a fair amount of detail why NCC offers far more value to enterprises than either premise-based call centers or (especially) outsourced call centers. The $950 per agent seat comparison price is for an outsourced call center offering the most basic of functions, and does not include multimedia multi-contact capabilities, sophisticated IVR, unified queues across multiple sites, remote agents, CCI-based integration to sophisticated CRM, and so forth. So it becomes pretty easy to argue that NCC, even at prices considerably above $950 per agent seat per month, offers the enterprise significantly more value than outsourcing.

BUSINESS CASE FOR THE NCC SERVICE PROVIDER
The section above shows the market for the NCC will support roughly $950 per agent seat per month for a minimally functional agent. A fully featured seat (i.e., one that would include inbound and outbound telephone capabilities, Web chat, voice and video over IP, e-mail and voice mail, all of course with unified queuing, skills-based routing, and many other state-of-the-art new generation call center capabilities), could therefore command considerably more, possibly as much as $1,500 per agent seat per month.

While we believe a more conservative price point for an NCC offering would be somewhere between $500 and $700 per agent per month -- and this would clearly make a strong business case from the enterprise customer's point of view -- we are going to demonstrate in the following section that even at $400 per agent per month, the service provider enjoys a strong business case with a healthy profit.

Of course, we're assuming a fully featured agent -- one that includes inbound and outbound telephone capabilities, Web chat, voice and video over IP, e-mail, and voice mail -- all of course with unified queuing, skills-based routing, and many other state-of-the-art next-generation call center capabilities. Excluded from the $400 price would be the agent's terminal equipment (typically a Windows PC with a USB headset) and the cost of the IP network between the enterprise and the service provider.

RETURN ON INVESTMENT
As shown in the business case model in Figure 2, an initial purchase of NCC infrastructure has a very attractive ROI for service providers. The model makes the following assumptions:

• All-in-one hardware plus software cost for NCC infrastructure of $5,000 per agent on a peak concurrent users basis;
   
• Monthly revenue of $400 per agent;
   
• Licenses for 1,000 peak concurrent users (this will be defined in the next section);
   
• A 2:1 ratio between named agents and peak concurrent users;
   
• A 12-month ramp-up period to sign the initial 2,000 agents;
   
• Nineteen percent annual fees for system support and maintenance; and
   
• Incremental general overhead expenses of 20 percent annually (as a function of revenue).

An initial software and equipment investment of $5 million breaks even early in year two and continues to generate a nice cash flow thereafter, with an internal rate of return (IRR) that is commensurate with the rapid market growth forecast for the NCC market.

Net present value (NPV) has been calculated using several discount rates. Whether NCC is being contemplated by an established telco that has a low cost of capital of 11 percent, an independent service provider with a cost of capital of 13 percent, or even a venture funded startup with a higher 16 percent cost of capital, the net present value is attractive in all cases and suggests the undertaking of such a project.

PRICE STRUCTURE ALTERNATIVES
There are several ways to structure the $400 per agent per month amount. The first is the most straightforward and it's called "Named Seat Pricing." It is essentially the full $400 charged as a fixed price per each physical or "named" agent per month. While this amount would be for a fully functional seat, service providers could offer discounts for partially functional seats (e.g., telephony only or Internet only). The main benefit of Named Seat Pricing is that the fixed fee approach gives enterprises the ability to easily anticipate and budget expenditures.

An alternative would be to charge on a per minute usage basis. Rather than paying a fixed fee per agent per month, enterprises would pay an amount that would vary depending on actual usage. Although this may make it more difficult for the enterprise to forecast expenditures, it may enable enterprises to better match their expenditures to the natural cycles of their businesses. With usage-based pricing, service providers could bill according to the following usage measures:

• Agent login minutes (total minutes all agents logged in);
   
• In-call minutes (total minutes agents logged in and in a call, be it telephone, chat, VoIP, video, or messaging).

Assuming eight hours of login time per agent per day, with 60 percent of that time actually spent in call, and assuming 1.5 IVR minutes for every agent in-call minute, a pricing model roughly equivalent to the above named seat model would call for $0.02 per agent login minute plus $0.02 per agent in-call minute plus $0.01 per IVR minute. This pricing scheme is shown in Figure 3.

While the bottom line still comes out to roughly $400 per agent per month, a benefit of such a scheme is that it allows a service provider to price high value-added NCC services in the same manner that toll-free 800 access service is priced -- by the minute. This greatly simplifies the offer while dramatically enabling the service provider to move up the value chain. A sample offering could be as simple as something like "five cents per minute for toll-free service plus two cents per minute for agent login, two cents per minute for agent in call and one cent per minute for IVR in call." Or even just 10 cents per minute for everything, where the number of concurrent agents determines a minimum number of minutes.

In fact, bundling call center services with 800-service minutes is a great way for a telco not only to differentiate a toll-free service, but also create a high degree of loyalty or "stickiness" for the enterprise to buy all of their 800 minutes from the service provider, rather than split the traffic among multiple service providers as many large enterprises do today.

Another way to structure pricing would be to have a basic monthly fee per named agent, plus a usage component (e.g., $200 per agent per month plus half of the per minute usage charges indicated above). There are several additional revenue opportunities for NCC service providers. These include initial setup charges and ongoing administration charges for such events as scripting updates.

Although the amounts for these items can vary significantly, we believe that charging for these items is not in the best interests of the NCC service provider. Initial setup charges serve primarily as a barrier to entry. With many enterprises seriously considering NCC, eliminating startup fees will help them choose in favor of making the change to a hosted call center. We also believe charges for administrative activities such as adds, moves and deletes, call flow scripting changes, and the like are ill-advised. One of the primary complaints against Centrex ACD as provided by the phone companies has been in the area of charges for scripting events and other configuration and reconfiguration activities.

It is important for service providers who are considering becoming providers of NCCs to choose a platform designed to support multi-tenancy, including a robust set of features enabling tenants control of agent configuration and grouping, call and Web scripting, and so forth without service provider intervention. Tenant self-administration is a win-win scenario: Tenants enjoy the convenience and control it provides while service providers save money and effort by providing it.

THE HIGH-WATER MARK ECONOMY OF SCALE
Service providers can leverage their NCC platform asset to benefit from economies of scale by ensuring they purchase the platform from a vendor willing to license it on a peak concurrent users or "high-water mark" basis. It works like this: The service provider sells monthly licenses on a named agent basis to the enterprise -- that means one license for each of the enterprise's physical agents. Now, not all agents will be working all the time, especially in a 24/7 call center, but also in call centers that are open only nine or 12 hours per day (there are coffee breaks, lunches, days off, training sessions, meetings, etc.). The service provider can use this to its advantage, as it only needs to buy a system from the vendor big enough to support the estimated maximum number of agents that would be logged in at the same time. The ratio between named agents and the high-water mark will vary, but the higher it is, the more the service provider benefits. Figure 4 shows a simple example.

Let's assume a service provider serves an enterprise with a 1,000-agent call center. Each agent works eight hours per day, but due to meetings, breaks, time off, etc., on average spends only six hours logged in per day. This means 6,000 hours of agent login time per day (the assumption here is no weekend work -- a M-F-only call center). Assuming the call center operates only 12 hours per day, we would have 500 agents logged in during a given hour (for the sake of simplicity, we're assuming an even distribution of call volume throughout the day). This puts the ratio of named agents to the high-water mark at exactly two -- meaning in this case the service provider has sold 1,000 named agent licenses, but only purchased 500 maximum concurrent logins from the vendor.

The business case model in Figure 4 assumes a named agent to high-water mark ratio of two. However, the above example is for illustrative purposes only and this ratio can be, and often is, much higher. For example, a 24/7 center with 67 percent login time yields a ratio as high as eight, which would dramatically improve the already good business case.

SUMMARY
NCCs are a tremendous opportunity, representing a win for all parties. Enterprises benefit by focusing on their core businesses while enjoying expanded functionality in their contact centers. Service providers benefit not only by enjoying a new high growth revenue stream, but also by playing a leadership role in this rapidly expanding market. NCC platform providers benefit by deploying a new generation of all-IP technology at a time when the call center industry is reinventing itself.

Erik J. Laurence is vice president, marketing and business development for CosmoCom. CosmoCom offers a new generation of IP-based call center platforms that are fully unified in supporting standard voice telephony and Internet-based multimedia communications.

[ Return To The September/October 2001 Table Of Contents ]

BY CHUCK SYKES

Some of the largest consumer-facing companies in the world are looking to solve what we call the triple whammy of how to get the best customer support performance at the best price -- while treating their customers like the world revolves around them.

Jupiter Research is projecting that with more complex products being sold over the Web, voice interaction will be the support of choice, and e-mail will follow. Businesses are already increasingly finding that the number of customer interactions is growing, despite the era of online self-service. Customers are moving to the Web to shop, but they want a higher level of contact and right now they are turning more to the telephone. People still want to interact in a human way -- they still need to talk, so you have to find a solution to meet this expectation. Hosted and offshore facilities can help alleviate the pain in managing this rise in customer demand.

A new report from Frost & Sullivan estimates that the market for outsourcing inbound and outbound customer care, help desk, and telemarketing will be worth $60 billion by 2007. The opportunity is there for the providers -- but what are the opportunities, advantages, benefits, and potential potholes for the companies in need? For an organization that needs a contact center to service customer demands and interaction, there are three options. You first need to decide whether you can afford to locate, build, equip, staff, and operate your own contact center, whether a joint venture is feasible, or if you can outsource the entire job to a reputable company.

Here are a few thoughts to consider for each path:

• The lock, stock, and barrel route. Be prepared to commit multi-millions in capital upfront (and ongoing) to locate, build, equip, staff, and operate your own contact center. If you have the expertise, or can hire an experienced consulting organization to achieve your goals, this might be a good choice.
   
• The carpool path. Joint ventures can be a good way to share risks and rewards in an area where you do not have total expertise. Your partners need to understand the intricacies of local government, staffing, and cultural issues, as well as technical infrastructure.
   
• The hired hand approach. By outsourcing the entire job to an experienced contact center business you gain control of how you want your customers treated, and at what costs. Be sure to decide upfront whether you want the company to link up with your systems, and gauge their reputation in the country you have selected.

Once you decide to outsource to the hired hand, ask yourself the following questions to ensure the best partner in the initiative:

Has the management team run a contact center of the size and scale required to get the job done?
Hands-on experience cannot be underestimated. The number of skills required span from workforce scheduling to training, call handling, and data mining. Selecting an outsourcer that can provide skilled staff and management from a CRM background, as well as the systems and site expertise will provide security.

Does the organization have a full understanding of any local government regulation, employment, and business issues in the offshore destination?
An in-depth knowledge of local tax issues or employment laws is imperative. On the ground, a provider should also be able to source and prepare regional businesses to support a new call center location, (such as telecom, security firms, and other vendors). All contacts need to be examined by a trusted provider who understands the local culture as well as the expectations and standards of your business.

Can you get the right staff?
Customer service or technical support agents need to be carefully selected and trained. The location needs to have the raw talent available and the right mix of part and full-time employees to keep the contact center at optimum operational levels, while allowing for the fluctuations in volume and demand. Offshore facilities in developing markets offer an open and untapped labor pool, and one that can be trained and developed to a company's own standards.

Channel capabilities -- can the center handle all incoming contacts, including voice, voice over IP (VoIP), fax, and e-mail, for example?
Your customers want to be able to use the channel of communication they choose. Voice is the hardest channel to take offshore, but remains important in making your customers feel part of something local, close, and comfortable.

Companies using a hosted center want guarantees that by putting their customers in someone else's hands, service levels won't fall. As a checklist, in order to be there for your customers at every stage, what steps should a provider follow and what should you expect if you outsource?

1. State of the art technology. Some of the main ingredients of a top-tier outsourced solution would include: A software system with an open architecture to link with you and support a heterogeneous environment of carrier networks; ACD, PBX, IVR, Web, and e-mail platforms; and complementary software applications.
    
2. Less investment risk. As businesses improve customer service to help them become more customer-centric, there is a risk that companies will invest considerable time and money building second-generation customer relationship management (CRM) systems and find themselves fenced in when customers start demanding third- and fourth-generation services.
    
3. Synchronized reporting and routing. The service solutions should allow you to obtain a real-time view of customer contacts, and the software needs to make contact-routing decisions for populating agent desktop applications.
    
4. Trained, customer-centric staff. A highly skilled and customer-centric workforce trained in technical skills and "soft" skills will engender a close bond between your brand and your customer.
    
5. Flexibility. As your business expands, or as you face peak customer contact periods in the customer life cycle, you need an outsourcer that can ramp-up quickly to support your growth, as well as tool-down during slower business periods.

If someone else is playing host, you should be able to sit back and reap the benefits for your customers, if you've covered these bases and found a solid partner. Any provider needs to have the expertise in understanding, using, and supporting the technology to maximize the investment in a hosted service. But only when combined with the right people and processes to manage it will this deliver a world-class service that meets the triple whammy on all sides -- giving support, at a fair price, for the best customer service.

Chuck Sykes is senior vice president and general manager, the Americas, for Sykes Enterprises. Sykes delivers customer care management solutions on an outsourced basis to industry leaders in technology, communications, and finance. The company specializes in strategic CRM, customized training, and business process redesign.

[ Return To The September/October 2001 Table Of Contents ]

BY DAVID BAEDER

The ASP model is a relatively new concept, especially for businesses still trying to get a handle on their CRM strategies. Even though it is rapidly gaining acceptance and momentum in companies of all sizes, the ASP CRM field is still subject to some persistent myths.

Myth: The Internet is too slow for effectively outsourcing CRM.
Fact: Running a CRM system from a good ASP can actually be faster than running the same software on a company network.

There are two considerations here: Most businesses today have the advantage of high-speed Internet connections, ranging from dedicated T3 lines to new DSL service over traditional copper phone lines. In addition, most home workers can connect through fast and reliable cable connections. In every instance, application processing will be extremely fast with no lag time. However, even remote users connecting over conventional 56 Kbps dial-up modems will now experience faster processing than would be possible on a company network. This is because most ASPs now use new technology such as Citrix MetaFrame, which offloads all application processing to the host server. The only data passing over the Internet are keystrokes from the client workstation and screen updates from the server. This technology only sends changes to the user's screen, rather than refreshing the entire screen like a Web browser. The low volume of moving data ensures fast application response.

Myth: The Internet isn't reliable enough for CRM.
Fact: The Internet has proven to be as dependable as corporate networks over the past few years, and is now starting to surpass them for reliability.

Although the Internet does experience occasional problems from time to time, the business world simply depends on it too much to tolerate significant outages. Corporate networks, on the other hand, are notoriously fickle. A single glitch on a solitary workstation can bring an entire network to its knees, shutting down critical applications and holding up multiple business functions. PCs lock up and have to be re-booted or repaired. Industry and government have together directed billions of dollars toward redundant public data infrastructures; the Internet is now more than ready to handle the demands of ASP-based applications such as CRM.

Myth: The Internet isn't secure enough for mission-critical CRM.
Fact: ASPs are among the most secure entities on the Net.

In truth, safeguarding applications and user data is mission-critical to the ASP business. ASPs understand user concerns and they typically put extraordinary precautions and security measures in place that can thwart even the most sophisticated hackers:

• ASP servers usually reside in ultra-secure data centers, with around-the-clock security and protection against fires and natural disasters;
   
• They encrypt all incoming and outgoing data transmissions;
   
• They keep multiple firewalls in place all the time;
   
• They run industrial-strength virus protection to guard against malicious code in transmissions and e-mail; and
   
• They back up everything regularly and even save the backups off-site.

Most users of outsourced CRM applications now find that their ASPs deploy more thorough security measures than what is in place on their own corporate networks. Their data has never been more secure.

David Baeder is president and CEO of Business Link International (BLI) of Providence, RI. BLI provides CRM solutions enhanced with Internet-based fax, broadcast fax, e-mail, and other communications options.

[ Return To The September/October 2001 Table Of Contents ]