December 1997

BY TIM DAVIS, DAVIS SOFTWARE ENGINEERING, INC.

Today’s call center typically has many roles and performs many functions. Hardware and software are available that make it possible to identify the caller, route the call to an appropriate agent, and display the caller’s complete contact history. In addition, the agent is called upon to accurately answer product questions, fulfill orders, schedule appointments, and assign help desk tickets. To effectively fulfill all of these duties, the call center manager must invest in four or five different software applications — at the very least! This reliance on applications underscores a move away from the switch-centric call center to an application-centric approach.

Computer-telephony integration (CTI) has allowed technological advances in telephony enabling and information collecting such as Dialed Number Identification Service, (DNIS), Automatic Number Identification (ANI), Automatic Call Distributor (ACD), and Integrated Voice Response (IVR). Add to that the proliferation of call center applications — contact management, order entry, appointment scheduling, sales management, and help desk support — and what we end up with is a personal computer or workstation overflowing with the best of today’s call center technology. Unfortunately, many of these wonders were not designed or tested to work together, so integrating them has posed many problems. In many cases where integration is attempted, the solutions don’t work together as well as one would hope or even expect, despite all of the marketing hype to the contrary.

WINDOWS
Thanks to Microsoft, we now have Windows everywhere. Certainly, Windows-based applications allow the agent to have as many tasks as necessary open and running at once. However, most of the paper-filled, cluttered desktops have now been replaced with cluttered screens. Sure, we can now have 10 different applications running at the same time, but how cohesive will the data underneath it be? Will the contact manager application be able to transfer data into the order entry application? Will the fulfillment application data show up in the scheduling application? Chances are, if they are all individual applications, they will not work well together. The expectations for call center software needs to move forward. We need to leave the telepho-nyenabling state of CTI and progress to the next level, which is the enabling of the entire call center and its many responsibilities.

ENABLING THE CALL CENTER
The majority of the products on the market today center quite heavily on scripting of unsolicited outbound calling activities. For some call centers, this type of product is more than sufficient since they are usually outbound service bureaus. But an everincreasing number of today’s call centers, many of which are in-house call centers, are tasked with more and more duties. The CTI technology needs to be expanded to meet these needs.

An oft-repeated complaint from call center managers is that in order for their call centers to do what they need to do, as many as six different applications are required. Let’s not forget the integration of these applications, which easily could run in the $100,000 to $200,000 range. These call center managers are showing an increased reluctance to settle for these slapdash solutions. One complete, expanded, call center application should effectively handle all the needs of a typical call center. Unfortunately, the current line of thinking is to simply accept the current situation: a loosely cobbled together assort-ment of CTI solutions as the norm. Switch manufacturers only compound the problem by proliferating the feature creep scenario which is pervasive in this industry. Most of these companies still take the approach that the switch is the central theme of the call center. This is a little like the proverbial tail wagging the dog. Although the switch is a vital aspect of the call center, the CTI software — or more accurately, the desktop application — is the real core component of today’s call center. Because of this, the desktop application, not the switch, is what needs to be expanded.

The switch has improved dramatically in recent years. It has infinitely more capability to hold a call in a queue, play messages or music, collect digits, and route a call. It does its job of delivering the live call to the appropriate agent very well. But it does not have the capabilities to move CTI applications where they need to go. It’s time for the software application (rather than the $1,200 phone) to be the focal point of the call center. It’s time for the dog to wag its own tail.

This is why we need to change our expectation of what call center applications are supposed to do. A switch does not have the level of sophistication to manage the entire CTI application — from the time a call becomes live in the center, to all the data presentation and collection, to the tracking and reporting that is done after the call has finished — but an expanded software application does. No matter how many programmable buttons are placed on the phone or how many “display lines” are available on these units, the desktop application will always prevail. The desktop application needs to control the switch, not the other way around.

ENHANCING THE CALL CENTER PROCESS
To better understand call center CTI applications, we need to describe the process transactions of the call center. These transactions have three very distinct phases. Any call center CTI application implementation must address each one in order for the call center to effectively communicate with and serve their customers. These three phases include:

• Connection of the outside contact to an internal agent.
• Knowledge transfer.
• Tracking and reporting.

The key to a successful call center transaction is that all three phases are working together to achieve the best communication — combining voice and data — that technology can provide. Each phase is somewhat dependent on the others. Phase two effects phase three. Phase one is dependent on what is gathered in phase three, and so on.

PHASE 1 — CONNECT THE CONTACT
Currently, 85–90 percent of the interest generated regarding CTI is centered around this phase. Although this is definitely an important phase, it is not the only one, and despite popular opinion, it is by no means the most important. The reason for all the focus on this phase of the transaction is due to the fact that it involves most of today’s CTI hardware, including switches, predictive dialers, IVRs, and ACDs.

The creation of the live connection is a given these days. Switch and predictive dialer manufacturers are down to splitting hairs in this process in their eternal quest to delineate themselves from one another. IVRs enhance this process and even aid in limiting the need for a live person. DNIS and ANI are also keys to this phase. ACDs use the DNIS and ANI to aid in directing a call to an agent, but with the production of switch-independent APIs, no ACD may ever again have the upper hand on an adjunct computer process. In a recent conversation with an ACD provider about the merits of using a particular product’s skills-based routing, the question was raised regarding how a switch could realistically know anything about the caller from just the DNIS and ANI. The vendor promptly replied that they maintained a list of ANIs in the switch, a list which could grow to as large as 50,000 entries. Now

let’s think about this. Is the switch really a good place to start maintaining a database of customer information? Moreover, in the general scheme of things, is 50,000 ANIs really that many?

Switches are responsible for making the voice connection, both inbound and outbound. But creating the live voice connection is only one part of the overall call center transaction, which brings us to phase two.

PHASE 2 — TRANSFER THE KNOWLEDGE
Once a live connection has been established (completion of phase 1), the real work of a call center begins. The predictive dialer may allow up to 60 minutes of “talk time” in an hour, or the ACD may serve up calls to agents with zero queue time. But, if the work accomplished during the “talk time” is sub-par, then the call center is not accomplishing its mission.

Usually, the main purpose of a call center is to facilitate some type of work effort between an agent and the customer. Doesn’t it seem only fitting that the agent should have the tools necessary to do that task? Companies spend hundreds of thousands of dollars on the latest and greatest phone equipment only to implement application technology at the desktop that is 30 years old. They may have updated the look and feel of the application with Windowsbased PCs on all the desktops, but the applications still have not changed.

Worst of all, most Windowsbased environments use Windows as a task switcher only. Although the agent doesn’t have three dumb terminals at the desktop, they still must deal with three separate applications. On top of that, we add another application just to handle the phone. This isn’t call center CTI. It’s only virtualizing the physical phone device into the one on the computer screen.

True call center CTI is enabling the agent’s desktop with a single application that is flexible enough to adapt to all types of call center transactions. Many call center customers like the GUI of Windows applications, but love the predictability of DOS applications. They want a call center application to provide consistency within a call, without the need for shrinking one application to get to another. The problem only intensifies when the need for crossapplication data presentation is required.

The information transfer phase needs to provide the agent with tools to facilitate many different activities. A robust contact management function is mandatory. In addition to basic knowledge about the caller, the entire history of communication with the customer — not just the communication that has gone on within the call center — should be available. Other functions should include: order entry, lead generation, scheduling, automatic faxing, appointment setting, and more.

Call center applications need to become the data presentation and collection tool for all underlying applications. The “talk time” of the call center transaction should be the most important phase of the process: This is what your customer will ultimately remember from the experience.

PHASE 3 — TRACK AND REPORT THE CALL
Both phase one and phase two rely on information gathered during phase three, the reporting of the call. This phase drives the database management processes of the phase two transaction. And the data collected during phase two is processed in this phase, which in turn updates the outbound call lists for outbound calling and the ANI-associated data used in the ACD inbound call routing of phase one.

Unfortunately, this is another phase that is not allowed to live up to its potential. ACD manufacturers have attempted to fill this gap by allowing the mechanics of the call to be tracked, such as the time it took the phone call to get answered, the number of calls that came in on a particular DNIS, or the average length of the call. The expanded, improved call center desktop application should be more adept at detailing the outcome of the call. By gathering and analyzing information — such as what type of products were sold to which customer, or what problems the agent encountered in the call, or recording market research data — call center managers are empowered to make educated decisions regarding the continuing operation of the call center. Gathering this information, and much more, is made possible by using an expanded CTI application containing an adequate phase three component.

CONCLUSION
Call center products should enhance the entire call center process, not just the initial connection portion of the call. We need to move past telephone enabling glorification and equip our call centers with the latest technology, effectively managing all three phases of the call center transaction. Until we raise our expectations of CTI software, the real potential of CTI will not be realized.

Tim Davis is president of Davis Software Engineering, Inc., a systems integrator specializing in design, implementation, and installation of computer-telephony integration (CTI) business solutions. The company’s leading product, TELE-SCOPE, is a PC LAN-based call center application that provides a total call center solution for telemarketing, sales, collections, service bureaus, or customer service applications. For more information, contact the company at 800-373-2668.