I n the past 10 years, the call center industry has made tremendous techno-logical leaps with “interactive voice response” (IVR) systems. With a myriad of revolutionary telephony-based applica-tions that have penetrated the call center marketplace, the IVR products seeing the greatest growth and potential have been automated attendant, call routing, and information retrieval applications. This new generation of IVR systems allows companies to provide a 24 hour self-serve line by giving callers rapid voice-activated access to information. For example, if you want to call a company to find out the sta-tus of an order you placed, you need only pick up the phone and speak your name. Gone are the days of having to dial by name, entering phone commands, fum-bling at your numeric keypad, and sitting on hold for unacceptable lengths of time.
With increasing pressure from con-sumers demanding better customer service, and the exponential growth of catalog and online ordering, the new generation of IVR systems provides an enormous improve-ment to the call center industry. However, speech technology is still in its infancy and has many more obstacles to overcome. Most IVR systems are fixed to “listen” for specific bits of information during a call. Once these key words are heard, the sys-tem processes the request, searches the database, and attempts to retrieve the cor-rect information and produce the desired result (i.e., providing information about the order that Jon Andersson placed). Clas-sic IVR systems only need to identify a small amount of information before for-mulating an “educated guess” and supply-ing the caller with information. But, what happens when a database has one person named Jon Andersson and three people named John Anderson? Would the com-puter be able to recognize which name the caller requested? Unfortunately, most sys-tems will find the first close fit and provide the corresponding information. As a result, this leads to an unacceptable percentage of inaccuracy. One of the best ways to com-bat this problem is by utilizing a new speech technology that conducts a “con-versational style” search.

The Conversational Search
For callers, a conversational style search is the most natural method for conducting search-and-retrieval functions in voice-driven auto-attendant, call routing, and information retrieval functions. This revo-lutionary technique allows callers to engage in a natural dialogue with the IVR system, much as they would with a live operator. The conversational process takes the caller through a series of dynamic questions to ensure easy and accurate access to people or valuable desired information.
The ideal conversational style search product should offer certain features to ensure accuracy. These features should combine sophisticated speech recognition techniques with the ability to search large databases of information on the basis of spoken input, all while conducting an ongoing, intelligent conversation with the caller. Let’s use the example of having two people named John Anderson in a direc-tory. The typical IVR system would most likely provide the information correspond-ing to the first John Anderson it found. By contrast, a conversational IVR system would be intelligent enough to recognize that it has two matches, and would ask a question to further qualify and ascertain the correct match. The following example demonstrates this sophisticated system in action:

Conversational System Operator: Thank you for calling the Main Street Catalogue Company. For information about an order that you placed, please state your full name or say “operator” to talk with a live customer service representative.

Caller: John Anderson
CSO: Did you say John Anderson?
Caller: Yes.
CSO: Are you John Anderson in Cambridge, MA? Or are you John Anderson in Providence, RI?
Caller: Cambridge, MA.
CSO: Thank You. The order you placed on June 6 was delivered on June 7 via FedEx. Thank you for using the Main Street Catalogue Order Status Information Line.

The system saw that there was more than one John Anderson in the database. By simply asking a few more questions, the system was able to identify the correct John Anderson and provide the correct information. Because every call center is configured differently, the company has the ability to determine which types of questions it wants the system to use in “qualifying” the caller. The company could use name, state, city, account number, order number, or any combination of these discriminators.
Conversational IVR systems have a unique built-in spelling feature that fur-ther enhances the probability of finding a successful match. What is the first ques-tion a live operator asks when they are having trouble understanding you? Don’t they say (and don’t you expect) “How do you spell that?” This conversational tech-nique solves real call center problems, such as two customers whose names sound exactly the same (such as Jon Andersson and John Anderson). The spell-by-name feature and the operator-like conversation allow the system to find the correct cus-tomer and provide the requested informa-tion. A typical IVR product would be stumped, and would either transfer the caller to a live operator or give a choice of two names that sound identical.
Another conversational feature that a typical IVR system could not provide is the ability to tell when a name is not in the database (also known as “garbage rejec-tion” or “false positives”). Conversational voice-driven IVR systems have fine-tuned their search technique to a point where they can conduct a conversation with the caller to “not” find someone who does not exist in the database. Why would you want to do this? This could indicate a much larger problem, like why isn’t the person in the database. Perhaps their order was never entered, or it was entered improperly? This allows a customer to be immediately trans-ferred to a live call center representative to determine the root of their problem, which is beyond the scope of the IVR system. Speech-based IVR systems relying on traditional technology will keep guess-ing at the name until the caller gets frus-trated and either hangs up or transfers to the live operator, to whom the caller will vent his or her anger at the “system’s” performance.

Conversation in Action
In the customer-oriented U.S. marketplace, providing first-rate customer service is a critical factor for success. Consumers are constantly looking for the easiest, fastest way to obtain order, product, and price information. With an automated system, call centers ensure that customer service is never compromised.
Such call centers may deploy speech-enabled solutions in two basic ways. They can provide assistance with simple transac-tions (providing the status of an order, for example, or quoting the price and avail-ability of specified products), or they can save precious agent time by serving as a front-end to accumulate information from the caller, and then routing the caller to the agent. Benefits associated with these features include better service, less waiting time, and reduced aggravation for the caller; greater productivity for agents who can handle more complicated customer service issues; and financial savings for the company.
Companies worry about the extra expense needed to provide the services customers are increasingly demanding. Conversational IVR systems help to bridge this gap. To be sure, a conversational style, speech-enabled search solution is a great factor in achieving increased productivity, cost savings, and exemplary customer ser-vice in the call center. As a result, this new technology is paving the speech-enabled IVR road of the future.

Corporate Software & Technologies 3333 Graham Blvd. Montreal, Quebec, Canada H3R 3L5 Tel: 514-733-8500 Fax: 514-733-8878 Toll Free: 1-800-675-2991 e-mail: info@cst.ca Web: www.cst.ca

CS&T Products
CorporateTime:
Calendaring and scheduling software

• Scales to support over 100,000 users
  
• Real-time transaction-based model
  
• Cross platform: Clients: Windows, Web, Mac, Motif Servers: Windows NT, UNIX
  
• Support for MS Outlook ® CorporateConnect: LDAP/X.500 directory connector CorporateSync: PDA synchronization
  
• Palm (available for Windows and Macintosh)
  
• Windows CE
  
• Psion CorporateTime Messaging Service for Outlook:
  
• Service provider connecting Outlook client to CS&T’s scalable CorporateTime Server (with any IMAP4 compliant e-mail server) CorporateTime Wireless Server:
  
• Receive notification of calendar events to PCS/GSM wireless phones

Customized training and consulting services

Company History, Philosophy
About CS&T
Corporate Software and Technologies Int Inc. (CS&T) is the world’s largest independent vendor of calendaring tech-nology for the corporate, Internet and wireless markets. The CorporateTime ® family of products allows millions of users to collaborate and manage activities and time. CS&T’s solutions enjoy a best-of- breed status, and are known for a range of functionality that supports access to personal and public schedules anytime, anywhere. Through its member-ship in the IETF, CS&T is committed to defining and implementing calendaring and scheduling standards. With installations of its products and services worldwide, CS&T has enjoyed steady profitable growth since its found-ing in 1988. The company is self-funded and privately held.

Mission Statement
CS&T’s calendaring solutions are designed to simplify the lives of con-sumers and communities, and increase the group productivity of corporate users. We pursue this mission through a com-pany culture that is dedicated to innova-tive design and engineering, high-quality software architecture, and professional technical support designed to ensure cus-tomer satisfaction. ADVERTORIAL—CORPORATE SOFTWARE & TECHNOLOGIES MM