Information Systems Projects in Contact Centers - Enterprise Information Systems Design, Implementation and Management: Organizational Applications - page 383

IVR consists of computer automats that allow
users to “self-serve” while communicating their
needs. Users interacting with an IVR use their
telephone keypads or, with speech-recognition
technology, voice commands to provide informa-
tion, such as account numbers or indications of
the desired type of service. IVR uses synthesized
voice to report information, such as bank balances
or departure times of planes. IVR can also be used
to provide simple services, such as the transfer of
funds among bank accounts.
CTI technology is used to integrate more
closely the telephone and information systems.
CTI allows a binding between a telephone call
and the context associated to that call, including
user's personal data, desired service and the trans-
actions done during the call (Cardoso, 2000). CTI
can be used to automatically display the caller's
user record on the agent's workstation screen. It
eliminates the need for the agent to ask the caller
personal information saving, in this way, agent's
time and reducing calls' duration. It can also reduce
variability among service times, thus improving
the standardization of call handling procedures.
Agents' skills can be used to route calls. This
routing, based on agents' skills is called skills
based routing (Avramidis & L'Ecuyer, 2005;
Koole & Mandelbaum, 2002). Agents can be
organized upon their skills, for instance the abil-
ity to speak languages. If one call arrives and the
system detects its origin, then it can be delivered
to agents speaking the right language, providing
a better service to the user.
CTI are IVR help to integrate a special informa-
tion system into a call center's operations, the Cus-
tomer Relationship Management system (CRM).
CRM systems track user's records and allow them
to be used in operating decisions. CRM allows an
efficient and personalized relationship between the
organization and each user, independently of the
number of the users (Cardoso, 2000; Gans & Zhou,
2003). “Call centers” became “Contact Centers”
with the use of interaction channels like e-mail,
fax, sms, chat or web. With these new channels

and business approaches, complexity increased
and, agent's skills had to improve to do the work
with new technologies and services.
Three main possible architectures and as-
sociated technologies can be identified. These
architectures, centralized, distributed, and hosted
may be used in conjugation generating hybrid ones
suiting the organization's specific requirements.
Furthermore, several different technologies may
be used in the architectures. There is any specific
match between architecture and technology.
A centralized Contact Centre, has a techno-
logical infrastructure based on a single server or
multiple servers, from the physical point of view,
but a single one from the logical point of view,
and is usually supported on a local network. The
interactions, when they reach the Contact Centre,
are placed in one single queue regardless of the
channel (Demaria, 2005). The system handles each
interaction according to a configurable priority.
For example, it is usual to treat synchronous in-
teractions like voice and Web collaboration in real
time. E-mail and fax interactions, for example, are
asynchronous interactions, and may be handled
offline (Koole & Mandelbaum, 2002).
Distributed Contact Centre represents an
evolution of centralized contact centre, enabling
to overcome the geographical and time zone
barriers. With the Distributed Contact Centre an
organization may operate in multiple geographical
distinct places, sometimes with very different time
zones, and thus offering a service 24 hours over
24 hours. It also makes possible the load balanc-
ing (Whitt, 2005). The basic idea of distributed
contact centre is to have a set of centralized contact
centres communicating and working as if they are
a single one. For example, if a user tries to access
the contact centre in her/his geographical area and
if the centre is already closed or not responding
due to an overload, then the interaction is routed
to another contact centre that is available. With
this architecture, the interaction can be processed
by the agent with all the necessary information
(Whitt, 2005). In the end, the original Contact