Smooth Sailing: Best Practices for Implementing an

Smooth Sailing: Best Practices for Implementing an
IP Telephony
Contact Centers
Smooth Sailing: Best Practices for
Implementing an IP Contact Center
March, 2005
Table of Contents
Section 1: Opening Statements ........................................................................................ 1
Section 2: Executive Summary ......................................................................................... 1
Section 3: IP Contact Center: Opportunity and Risk ..................................................... 2
Section 4: Issues and Corresponding Best Practices .................................................... 2
Section 4.1: Capabilities essential to the contact center ........................................... 2
Section 4.2: Deployment success begins with assessment ....................................... 3
Section 4.3: Quality of service vital for voice traffic ................................................. 4
Section 4.4: Reliability takes on new meaning ......................................................... 4
Section 4.5: Security on many levels ...................................................................... 5
Section 4.6: Manageability issues .......................................................................... 6
Section 4.7: Scalability concerns ........................................................................... 7
Section 4.8: Evolution vs. revolution....................................................................... 7
Section 5: Avaya Takes IP Contact Centers Seriously ................................................... 8
Section 1: Opening Statements
Among the most successful 16th-century mariners competing to explore the New World was the renowned
Captain Drake. Many countries were rushing expeditions into this uncharted territory. Drake, however, had an
advantage: from foreign ships he had kidnapped navigators who knew the coastline from previous voyages.
With their help, Drake navigated the dangerous waters and went on to establish routes that others would
follow for centuries to come.
When you run a contact center, the alluring territory of internet protocol (IP) telephony can seem uncharted
and perilous. Fortunately, Avaya knows these waters well. Our experience can guide you through a successful
journey. And you don’t have to abduct Avaya experts to get that experience. In this paper we have compiled
the lessons of our long history of IP contact center deployments. Think of this paper as your Niño daSilva.1
Section 2: Executive Summary
Over the years, conventional circuit-switched technologies have earned the loyalty of IT managers. But
these technologies have limitations, making it complicated and costly to deploy today’s advanced solutions
throughout the enterprise.
IP telephony, meanwhile, is proving itself as an architecture that is flexible and cost effective. Savings and
efficiencies can be gained by migrating to IP telephony — and lost by stumbling into common pitfalls.
Adhering to best practices during your planning and implementation will help to avoid the snags and ensure
success. This paper outlines the hazards and best practices as collected from an Avaya history with contact
centers and IP telephony.
Conventional contact center solutions include a rich set of advanced features, the loss of which can harm
the contact center’s performance metrics. Avoid a backward step by preserving that full feature set, including
third-party applications, in the transition to IP.
The tests of IP telephony are call quality and reliability. A detailed network assessment is essential to
identifying weaknesses. Define your organization’s service level agreements (SLAs) and standards for QoS up
front, to avoid shifting expectations.
As the voice and data networks converge, voice over IP (VoIP) communication must be protected. Think in
terms of the “security trinity” in your network design. Implement measures that will protect your assets,
detect a breach, and enable you to respond quickly.
Convergence raises questions regarding the responsibility for managing integrated systems. Think beyond
appointing roles for the usual functions. Concentrate instead on who will be in charge of monitoring,
managing, and reporting.
Be prepared to support your company’s plans for growth. Deploy a system that has proven itself handling
thousands of lines. Verify that the solution you choose will support your plans to pool contact centers and
interflow calls between them.
Making the jump to pure IP telephony is not the best path for every contact center. Uprooting existing private
branch exchange (PBX) infrastructure and integrating new voice servers with the data network could be
disruptive and costly. If yours is an established center with a digital PBX, consider IP enabling the existing
PBX to protect your existing investments.
Niño daSilva was one of Drake’s most willing and valuable Portuguese navigators, having been to South America on previous journeys.
Avaya recognizes that the financial commitment of a VoIP infrastructure can be significant. You must choose
the right path for migration. Avaya lets you take each step at your own pace. Choosing Avaya reflects the
choices you make about technology directions, protecting investments and, above all, serving customers.
Section 3: IP Contact Center: Opportunity and Risk
There is a certain warm comfort that comes with the tried and true. Conventional circuit-switched PBXs
and automatic call distribution (ACD) have been around for a long time. They are secure, reliable, and have
sophisticated capabilities for contact centers. Over the decades, they have earned the loyalty of conservative
managers who have spent years fine-tuning their contact centers to most efficiently use their agents.
The winds of change, once a distant rustling, now are pushing the status quo toward a new course.
Distributed contact centers are increasingly important to global organizations. This, in turn, strengthens the
interest in centralized administration, especially where sophisticated, industry-specific telephony applications
are involved. Cost control and efficiency are the battle hymn of almost every industry.
These trends in contact centers are filling the sails of IP telephony. Circuit-switched contact center technologies
have their limitations, making it complicated and costly to deploy the next level of advanced capabilities throughout
your enterprise. By contrast, an IP-based architecture is flexible and cost effective. Analysts agree that it is not a
matter of whether IP telephony will become mainstream, but when — and the projected timeframe is short.
While IP telephony can be justified on cost savings and efficiency, one major deployment problem could wipe
out the savings. And if the new solution does not provide better efficiency with the same functionality, the
cost savings could be consumed by additional staffing costs.
Fortunately, the switch to IP does not have to be risky. This paper outlines the hazards and best practices as
collected from a long history with contact centers and IP telephony.
Section 4: Issues and Corresponding Best Practices
The following eight sections describe the themes that may arise during the transition to IP telephony, and the
best practices for avoiding or overcoming issues.
Section 4.1: Capabilities essential to the contact center
Conventional contact center systems include a rich set of advanced features that work together to improve
customer service and agent efficiency. Since the largest cost component in a contact center is staffing,2 the
loss of any of these features can harm the contact center’s performance metrics.
Basic call operations, such as transfer and conference, are standard features in any IP PBX. Advanced contact
center features are much more complex and should not be assumed.
Call routing, for example, which gives callers faster service, bases its decisions on expected wait times that are
derived from internal data such as the number of agents staffed, number of calls in queue, call priorities, and
average handle times. Features using sophisticated predictive algorithms based on current staffing and past
behavior are providing even more flexibility for contact routing. Less sophisticated routing can result in extended
wait times and frustrated callers. Adding staff to compensate may wipe out any cost savings achieved.
According to CIO Magazine, personnel accounts for about 70 percent of each call center’s costs; the rest goes to rent, phone bills, and computer
and telephone equipment. (“Working Smart: Airborne Freight’s Call Center Management System,” April 15, 1999 [
Best practices for preserving productivity
The move to IP telephony should expand the capabilities of your contact center, not diminish them. Work with
your vendor to ensure absolutely that your center will continue to enjoy all of the features it depends on today. Your
current staffing levels (and therefore costs) are based on having the rich, mature applications you use now. Avoid a
backward step by preserving that full feature set, including third-party applications, in the transition to IP telephony.
A robust IP contact center architecture separates the application layer from the transport layer. Contact center
features are the function of software applications, while the transport — whether TDM or IP — is hardware
dependent. When applications operate independently of their transport layer, they can continue uninterrupted
when the transport is migrated from TDM to IP.
One of the promises of IP telephony is that it enables ubiquitous access to applications from every enterprise
location. Whether the agent is headquarters based, in a branch office, or a home agent, they can and should
enjoy the same feature set. Validate each vendor’s complete support of your set of devices, with standardized
capabilities and appropriate user interfaces.
Continued performance improvement in your contact center will depend in part on future enhancements
to your applications. Avoid committing the enterprise to a proprietary environment that would limit rapid
enhancements or become a dead-end technology. Evaluate any IP telephony offering on its open architecture
and broad support from the development community.
Section 4.2: Deployment success begins with assessment
The supreme test of IP telephony in the contact center is call quality and reliability. These topics are so
critical that we have included sections on each in the pages that follow. Overall voice performance is affected
by the performance of each network component, so the burning questions are: “Can the IP network handle
voice?” and “How will we know in advance?”
Of course, the network eventually will handle voice, but Gartner estimates that 85 percent of today’s large
enterprise LANs will require hardware or software upgrades before they are ready to support IP telephony. In
other words, the first weeks and months on IP telephony can be rough ones, if issues are not identified and
resolved beforehand. During a difficult deployment, the community of concerned collaborators can expand
considerably. Expectations can become a moving target, amidst unwelcome attention to the project. Key
personnel can spend untold hours tinkering with the network and managing perceptions.
Best practices for identifying network weaknesses
To avert a difficult deployment, involve the right people early in the planning process. Contact center management
and IT should collaborate especially closely to actively address each of the issues outlined in this paper.
An objective, detailed network assessment is essential to identifying weaknesses. Assessment requires asking the
right questions pertaining to quality, reliability, and the prioritization of voice on the IP network. The assessment
should endeavor to anticipate growth, planned restructuring, acquisitions and divestitures, to the degree possible.
Assessments require specific technical capabilities and tools. It is often advantageous to engage an outside
service provider for this stage of the process. They will be able to locate assessment hosts within the network
in such a way as to accurately measure end-to-end performance. The tools they use will be specifically
designed for IP telephony environments.
Once the assessment is complete, proceed with the indicated network upgrades and changes. Quality and
reliability depend on it, and are worthwhile exploring in more depth now.
Section 4.3: Quality of service vital for voice traffic
The quality of service (QoS) requirements for voice are different than for data. Some people remember the
experiences of the early years of VoIP. Their first question is whether the voice quality will be acceptable.
The main QoS issues to consider are those that might result in the perception that a caller has a bad
connection. Voice over IP is digitized and broken down into data packets, each containing a fraction of a
second of sound. Voice is a real-time function, and quality can be adversely affected by delay or packet loss
caused by any underperforming component on the network.
Delay introduces an unnatural audible effect to the conversation. It occurs when there is a delay as voice is
being compressed, packetized, transmitted, reassembled and decompressed — a formidable job that must be
performed in an imperceptible length of time, every time. “Latency” and “jitter” are terms applied to describe
the source of delay problems more precisely.
Packet loss happens when a high percentage of packets are not successfully routed from end to end. Nominal
packet loss goes unnoticed; high packet loss causes poor sound quality or a static-like noise. With data, lost
packets can be re-sent; but re-sent voice packets would arrive too late to be delivered in the correct sequence.
The voice contained in the lost packets, however minute, is discarded.
Best practices for ensuring QoS
The definition of adequate QoS varies from company to company. The key to achieving QoS goals is for
stakeholders to reach agreement early on the definition of QoS and the assumptions that affect it. Define your
organization’s standards for QoS up front, to avoid the moving-target conundrum of expectations.
QoS deserves special attention in WAN infrastructures, or where Real-time Transport Protocol will be used, for
example, to deliver videoconferencing.
QoS is closely related to other areas — reliability, manageability and scalability — where potential issues will
be uncovered in the network assessment. Service levels will depend on defining the standards and heeding
the assessment results.
Section 4.4: Reliability takes on new meaning
Is IP technology reliable enough to send customer calls over it? In terms of “uptime,” we already know that IP
data networks can achieve 99.999 percent (five nines) reliability.
As long as an IP network is delivering data, it is considered to be “up.” When the IP network becomes the
voice network, reliability standards are more stringent. Voice network downtime is unacceptable for most
contact centers. The new measure of reliability is a combination of availability and call continuity.
Reliability in data networks is achieved by thinking in duplicates — redundant servers, redundant routers,
redundant power supplies — to make interruptions go almost unnoticed. Redundancy alone does not
guarantee call continuity. Even if almost all downtime is eliminated, calls can be dropped in the split-second
switch from primary to failover systems. Call continuity requires maintaining active calls during a failover
event. Otherwise, a brief power failure could disconnect all conversations in progress.
There are powerful alternatives in reliable data networks. For example, Avaya and Extreme Networks have a
strategic alliance for best-of-breed converged networking solutions. Extreme Networks provides the robust data
solutions to meet the rigorous security, availability, and performance demands of converged networks, while
Avaya brings to the table advanced voice IP applications. Together, this powerful alliance provides enterprise
customers advanced converged networking solutions and services not available anywhere else in the market,
i.e., truly best-of-breed solutions.
Voice network availability is a factor of uptime and sufficient voice quality. A data application might respond
slowly during a minor network problem, but the same network problem can cripple a voice application. If the
connection quality is so degraded that the participants terminate the call, then there has been a perceived
episode of unavailability.
Best practices for preserving availability and call continuity
Network failures can take many forms, from brief slowdowns to complete outages. Define service level
agreements (SLAs) and then contrast them to a spectrum of failure scenarios. Using the network assessment
as a guide, judge the ability of the LAN or WAN to meet the SLAs. Use the results of this analysis as the basis
for any necessary upgrades in response to the network assessment.
Confirm that your vendor can deliver high availability and call continuity. This level of reliability requires sophisticated software design that allows for call state information to be shared between primary and failover components.
Build redundancy into the network with the goal that users and callers should never perceive an interruption
of service. There are several approaches that should be considered during the assessment:
• Duplicate network paths supported by multiple service providers, WAN access points and switches
• Duplicate servers for applications and core IP network services
• Duplicate IP interfaces so that, if one fails, IP endpoints can be redirected to another interface without interruption
• Virtual router redundancy protocol (VRRP) on large networks with layer 3 devices
Global enterprises with multiple contact centers can, with the right technology, choose to treat all centers
across the enterprise as one pool of agents. This enables a country n +1 strategy of redundancy, where a
failure in one center is accommodated by sending overflow to another. Consider this possibility and build it
into your requirements as needed.
Section 4.5: Security on many levels
Toll fraud was once the main security concern of contact center managers when adopting new technologies.
IP telephony sprouts its own crop of potential security risks:
• Eavesdropping on unencrypted voice communications
• Denial-of-service attacks
• Viruses and malicious code
• Hacking and data theft
• Unauthorized inside access via the LAN
• Abuse of external access provided for remote or virtual agents
None of these risks is new to data networks, but the conventional voice network has not been previously
subjected to them. Voice communication is an important function that must be protected as the voice and
data networks converge. Shortcomings in products deployed, as well as shortcuts and overlooked security
risks, create "holes" that are convenient for attackers.
Best practices for securing VoIP communications
Security stems from design as much as from implementation. A network assessment will uncover security risks,
and those should be dealt with at the outset of your deployment. Think in terms of the “security trinity” —
incorporate prevention, detection and response in your network design.
Involve all of your technology and business stakeholders to collectively anticipate potential threats. Identify what
you are protecting, and from whom or what you are protecting it. Prioritize the most likely threats. Then implement
cost-effective measures that will protect your assets, detect a breach, and enable you to respond quickly.
The best security policies are those that can be implemented through system administration and published
guidelines. In the former case they are enforceable with security tools; in the latter, with consequences for
violation. The Internet Engineering Task Force has identified the components of a good security policy,3
ranging from technology purchasing and maintenance to access and authentication.
Consider these broad categories of security measures:
• Data security — passwords, encryption
• Code security — firewalls, virus protection
• Physical security — locks on doors, alarm systems
• Corporate security — written policies, regulatory compliance
• Assigned responsibility for security
That final point is important, and not only for the resulting accountability. Security and compliance are a
changing landscape. A person or role must be responsible for continuously reviewing and improving your
security measures.
Section 4.6: Manageability issues
Who will manage the converged network and deal with issues that arise? Voice and data networking were
previously managed by separate functions, using their respective sets of tools. Convergence raises questions
regarding the responsibility for managing the integrated systems.
Monitoring network performance will be essential to the contact center’s performance. Tools designed for
data network monitoring will not be adequate for monitoring voice on the network. Likewise, the tools used to
manage the voice network were never intended for the rigors of data network management.
Best practices for network management
When assigning responsibility for management of integrated system, think beyond appointing roles for the
usual move-add-change functions. Identify who will be in charge of monitoring, managing, and reporting.
IP telephony enables new levels of monitoring for call quality — even monitoring every call, if desired.
Incorporate the monitoring level that is right for your contact center. In some cases, real-time continuous
monitoring, or even remote monitoring, may be needed. With adequate monitoring, you or your remote
monitoring provider can often fix a problem before agents know it exists.
Equip network managers with an integrated tool set that is designed for IP telephony. Ensure that the tools are
capable of managing a multi-vendor environment and integrating with your existing technologies and processes.
IETF “Security Handbook,” RFC 2196 pp. 8-9,, 1997.
Outline your strategy for handling network issues after deployment. Assess what problem scenarios could arise,
how they will be detected, and how they will be dealt with. You will have averted most of them by following the
recommendations in this paper; nevertheless, the unexpected can happen and it need not be disruptive.
Section 4.7: Scalability concerns
How will your IP network scale as your contact center expands? This question is important for two reasons. First, you
must be prepared to support your company’s plans for growth, or risk experiencing very visible growing pains at each
of your contact centers. Second, IP telephony enables you to treat multiple contact centers like one center, so your
scalability requirements likely will comprise your total number of combined lines for all centers.
Best practices for handling growth
Scalability is addressed simply: Deploy a system that has proven itself handling thousands of lines. Verify that
the solution you choose will support your plans to pool contact centers and interflow calls between them.
During periods of exceedingly high call volumes, interflowing calls over IP can stress any network. Consider
configuring your network to overflow some calls to the PSTN when the data channel is too busy to maintain
your standards of quality.
Section 4.8: Evolution vs. revolution
Your company’s PBX equipment represents an investment in TDM infrastructure. What will happen to that
investment if you move to an IP-based voice network? Justifying IP telephony is a reasonable concern,
especially if the PBX is a recent upgrade to the current generation of TDM technology.
Uprooting existing PBX infrastructure and integrating new voice servers with the data network can be
disruptive and costly. There is the software required to deliver the full capabilities you need. There is the
hardware necessary for redundancy. If you choose to deploy IP end-to-end, there is the task of replacing
digital and analog desk sets with IP phones. Then there are the LAN or WAN upgrades that may be deemed
necessary in the network assessment.
Best practices for protecting your investment
Making the leap to pure IP telephony is not always the best path for every existing contact center. If yours
is an established center with a digital PBX, consider IP enabling the existing PBX. You can retain your
investment in existing equipment, as well as familiarity with the current system’s features and user interface,
and yet take advantage of using IP where and when it makes sense for your business.
This factor is most evident in the decision to deploy IP phones. A flexible IP environment will allow agents to
keep their familiar TDM phones, unless there is a business rationale to switch to IP hard phones. Agents who
work some days from home (e.g., for workforce flexibility or business continuity reasons) may prefer IP soft
phones, so their interfaces will be the same in both locations. And for new centers or expansions, IP phones
generally make the most sense.
Make new investments that will endure through to a pure IP stage. If your company is planning to expand, or
if you are currently evaluating a new or upgraded PBX platform or contact center system, it is important that
the wiring and contact center platforms under consideration support the direction of your company toward IP.
Evaluate IP telephony vendors based on all of the best practices outlined in this paper. Consider their ability to
provide assessment, hardware, software and services, as well as the best tools for security and management.
Take time to work through the details of your implementation plans with your vendor. Make sure their
application interfaces use an IP connection, so no change to the interfaces will be required during the move
to an IP infrastructure. Announcements and music on hold should be sourced where it makes sense for the
business, regardless of whether the infrastructure is IP. Likewise, IP and SIP trunks can be used within IP
contact centers if there is a business reason to use them.
Section 5: Avaya Takes IP Contact Centers Seriously
Avaya recognizes that the financial commitment of moving contact center traffic to a VoIP infrastructure
can be significant, and that you will base your decision on the unique requirements of your business. Each
company must evaluate IP telephony for their unique situation and choose the right path for migration. A
contact center must take each step at its own pace, without rushing into unnecessary forklift upgrades. Your
choice of vendors reflects the choices you make about technology directions, protecting investments and,
above all, serving customers.
Whether you choose the pure-IP path, or to IP-enable your existing PBX, Avaya offers both approaches to
convergence. With Avaya, you have several alternatives for introducing IP telephony into your contact center.
You can add IP to your existing Avaya contact center, or you can implement LAN-based solutions. Additionally,
Avaya offers options to deploy a network-based, or hosted, contact center solution that allows you to access
the same contact center functionality, but you can pay for usage on a monthly basis and avoid upfront costs.
Alternatively, you can choose Avaya servers and gateways that allow your endpoints to be flexible. For outside
connectivity, you can continue to select from the public network, tie lines, ISDN, an ATM network, or an IP
broadband network provided by an Avaya certified Network Service Provider. At the agent desk, you may want
to keep your digital or analog phones, use IP hard phones or IP soft phones, or have a mix of these.
In addition, Avaya knows that alliances help individual companies focus on their strengths and build solutions
that are more comprehensive, faster to implement, and more cost effective for their customers. While some
organizations are successfully finding their own way to the advantages of IP telephony, the complexity of this
challenge is significant. IP telephony technology and systems have reached a level of maturity where they
must be treated in a holistic manner in terms of overall network performance, security and reliability, and the
business activities they support. Alliances, such as Avaya and IBM for example, can help maximize the
benefits of complex technologies like IP telephony and help companies to minimize risk and reduce the time
it takes to implement a solution. This relationship, and other alliances, ensure that vendors focus on the
development of business solutions and minimize the inter-vendor conflicts that might otherwise develop. In
this way, Avaya can help individual companies focus on their strengths and build solutions that are more
comprehensive, faster to implement, and more cost effective for our mutual customers.
Avaya offers companion white papers that address the business value of moving to an IP contact center and
that help you map these suggestions to Avaya solutions and services. For more information on how Avaya
can help you safely navigate your way to IP telephony, contact your Avaya Client Executive or Authorized
Avaya BusinessPartner, or visit us at (go to Research By/Resource Type/White Papers).
Some examples are “Three Best Practices for Today’s Profitable Contact Centers“, “IP Telephony Migrations:
Alliances Make Solutions Work” and “The Decision to Deploy a Hosted Contact Center”.
About Avaya
Avaya enables businesses to achieve superior
results by designing, building and managing their
communications infrastructure and solutions. For
over one million businesses worldwide, including
more than 90 percent of the FORTUNE 500®, Avaya’s
embedded solutions help businesses enhance
value, improve productivity and create competitive
advantage by allowing people to be more productive
and create more intelligent processes that satisfy
For businesses large and small, Avaya is a world
leader in secure, reliable IP telephony systems,
communications applications and full life-cycle
services. Driving the convergence of embedded
voice and data communications with business
applications, Avaya is distinguished by its
combination of comprehensive, world-class
products and services. Avaya helps customers
across the globe leverage existing and new
networks to achieve superior business results.
© 2005 Avaya Inc.
All Rights Reserved. Avaya and the Avaya Logo are trademarks of Avaya Inc. and may be registered in
certain jurisdictions. All trademarks identified by the ®, SM or TM are registered trademarks, service marks
or trademarks, respectively, of Avaya Inc., with the exception of FORTUNE 500 which is a registered
trademark of Time Inc. All other trademarks are the property of their respective owners.
Printed in the U.S.A.
04/05 • EF-GCC2691
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF