Transforming Your Call
Center: Living Happily Ever After In The Contact Center
BY IKE MITCHELL, COMPUTER SCIENCES
CORPORATION
Most companies are experiencing an increased demand for customer services through multiple media sources and are finding it necessary to expand their operations in ways they never imagined. How do you take a customer care operation from nothing (or almost nothing) and grow it to meet the needs of your business? How do you avoid throwaway investments? How can you have a champagne customer care operation on a beer budget? Careful planning to capture customer care requirements, using all the features of the technology you have and spending your strategic technology investments on the right technology the first time, are the hallmarks of a successful customer care organization.
Because one of the greatest complaints about e-commerce is an appalling
lack of customer service, this article will concentrate on e-business
startups, but the information that follows is applicable to all
businesses. E-business startups share common customer service profiles
across their industry. Customer service at this point in the industry life
cycle is either nonexistent or is an unorganized hodgepodge of resolution
processes. Even well-established companies venturing into e-business face
the startup's unique customer service challenge. Unlike traditional
brick-and-mortar businesses in which high levels of service are the price
to "play the game," in the new e-business markets, quality
customer service still provides a competitive advantage.
In today's world, much of the service provided by a business is
accomplished on the Web page itself. At some point, however, all companies
face the need to provide human-based service through a contact center.
Companies that are building their customer interaction center from scratch
typically go through three distinct phases of development:
- Basic services,
- Traditional call center services, and
- Enterprise customer service.
Through each of these phases, the services to be provided, the quantity
and quality of the staff to provide these services, the technology
required to deliver these services and the processes used to resolve
customer issues develop and change. It is possible to have a roadmap to
ensure investments, both dollar and human resources, at each phase can
evolve to support the next phase. Additionally, at the speed of
e-business, these investments must be implemented quickly, thus this
roadmap can eliminate missteps that delay the deployment of customer
services and smoothly move the operation to a point of excellence. Quickly
moving through the first two phases enables the e-business to focus on
those more advanced e-customer services required to distinguish a company
in its niche.
Basic Services
The focus of provisioning basic customer services is on getting calls
answered and resolving the issues. Operational excellence and
"wow" technology cannot be addressed at this point without
significantly delaying the introduction of the customer service function.
Services. During the early days of operating a basic customer
service function, the type of calls expected can only be estimated. These
guesses are best made by focus groups trying to navigate the Web site.
Those readily identified services should be documented, resolutions built
into the site itself if possible, and resolution procedures produced for
customer service representatives. This is only half of the solution to
identifying services to be provided by the center. Once the center opens,
a method for agents to capture call types is a must. Soon after opening,
frequent meetings with customer service representatives to flesh out
additional service requirements are required. By the end of phase one, a
solid list of service requests and resolution procedures should have been
developed.
Staff. It is during this first phase of providing basic services
that the technical skills of the personnel required to service callers are
at the highest. Because the type of service request is not clearly
defined, generalists with broad skills, who are fairly knowledgeable in
the company's product or service, are required. In addition to servicing
callers, these people have very specific tasks to develop support tools
for future reps, reducing the technical skills required by customer
service staff in future phases. Through the knowledge acquired in Phase I,
customer service staff develop deep knowledge of the company's product or
service and often become valuable employees, moving on to other jobs
within the organization. It is essential to capture and institutionalize
the knowledge obtained by these people. As a company moves to more
advanced levels of customer service, efforts should be made to retain them
as subject matter experts and they should be made available to
second-generation customer service personnel. They will make excellent
supervisory personnel for the next phases of the customer care operation
development.
Because the initial workload for a new customer service application is
small, the initial group of service reps should split their time across
multiple job functions to maintain their productivity. Staff with
back-office responsibility typically fills this need. These individuals
develop a fairly deep understanding of the "behind the scenes"
activity required to support the business and often become knowledge
powerhouses, capable of moving into any area of the business.
Technology. Now that we have an understanding of the services
this basic call center will offer and who will handle the calls, the focus
then turns to technology questions. How do we get the call to the right
group of people and what tools are needed to support them?
Many of the issues surrounding the opening of a basic services call
center focus around telephony issues. Although the center is small, there
is a certain base level of telephony capability required of a telephone
system. These are:
- Routing calls directly to the customer service group, bypassing the
receptionist,
- Queueing callers should all customer service representatives be
busy, and
- Playing specialized announcements/ music while callers are on hold.
Most companies look to their existing phone system for solutions.
Unfortunately, most Internet startups have basic key/hybrid phone systems
with limited capability to support even a basic call center operation,
much less a full-function customer center of excellence. At best, the
existing system may have basic automatic call distribution (ACD) or
uniform call distribution (UCD) capability. Additionally, access to the
public switched telephone network (PSTN, or the telephone lines) must be
configured to identify callers needing customer service. The PSTN access
and the premise telephone system must work together to get callers to the
customer service group with minimal intermediate steps such as a
receptionist or automated attendant ("press 1 for sales, press 2 for
service"). Obviously, if a toll-free number (area code 800, 888 or
877) is to be used, this must be obtained and directed to the appropriate
termination to accomplish this routing. (If using a toll-free number, you
may find it difficult to obtain a suitable vanity number such as
1-800-FLOWERS. In this case, try for an easy-to-dial, and
easy-to-remember, number. These easy-dial numbers are especially
appropriate if your business is international, as not all phones have the
alphabet on the dial buttons.) Coordinating these efforts can be difficult
and time consuming because there are hundreds of options depending on the
capability of your telephone system, and there are at least three vendors
involved in setting this up (the premise telephone system, the local
telephone company and the long-distance/800 service provider). The key
point to remember is to use the capabilities of your existing system. This
phase may justify small investments (<$10,000) to obtain the necessary
capability.
Processes. From an information systems point of view, many
startup customer care operations begin with nothing more than #2 pencils,
yellow pads and a folder of FAQs. Whether paper-based or automated in some
way, basic services information systems have to accomplish the following
objectives in addition to any transaction requirement you may have:
- Call volume stroke count,
- Call type stroke count,
- FAQ capture system, and
- Escalation tracking.
These basic statistics form the basis for sizing and justifying the
next steps in developing the customer care operation. They enable the
anticipation of service requirements rather than simply reacting to them.
Many companies make moderate investments in existing customer contact
systems, such as ACT or GoldMine, to create the core call data processing
capabilities. These systems can often be customized to provide basic
customer relationship management (CRM) capability. While a good place to
start, the primary limitation of such systems is that they are trying to
do CRM functions with contact management tools. As such, you are always
developing processes around the limitations of the technology rather than
building technology to support the required processes. At this point, it
is important not to make large capital or human resource investments into
such tools because in Phase II it may be necessary to discard them and
begin again.
Traditional Call Center Services
Phase II is characterized by the movement from "make do" systems
to true CRM tools and the impact this migration has on people and
processes. Moving to this phase is where the most money is wasted if the
effort is not well planned and smart decisions are not made. Decisions
here are the most complex and will set the stage for how easy or difficult
it will be to move into the final, "enterprise" phase of
customer care.
Unfortunately, many business managers are unaware of the
interrelationships of the many parts required to build even the most basic
contact center. Office managers, responsible for basic office product
technology such as copiers and faxes, are often called upon to make these
decisions. As such, decisions are commonly made on a technology by
technology basis, without regard to how they will work together. As a
result, the office manager looks to vendors for help. This is a very
dangerous approach. A vendor may look you square in the eye and say,
"Yes, our product is CTI compatible and has standard
interfaces." While superficially true, this statement is not unlike
saying it writes to an alphanumeric file. Sure it is standard, sure
everyone can read it. It does not necessarily mean that the information
sent over the interface is complete or useful, or that other vendors you
are working with have developed the interfaces to use it. The CRM
marketplace is market driven. As such, if you invest in a platform with a
small installed base, you will continually face interoperability problems.
The major customer care system vendors develop interfaces largely for
other major players in the marketplace.
Services. If you followed the advice above when developing your
basic services contact center, you should now have a wealth of information
on which to build your services transaction set. Your call volume and call
type stroke counts have identified opportunities to move simple service
issues to the Web page. After implementing these, continued volume and
type stroke counts will identify Web design problems and perhaps a need
for interactive voice response (IVR) capability. You have also done a
prato analysis and understand the 20 percent of call types that make up 80
percent of your calls and you have created exhaustive FAQs and detailed
solutions/ processes for those 20 percent of transactions. When you enter
this phase of customer care, your service transaction set should be well
understood and in a form to automate in a knowledge base as part of a
basic CRM implementation.
Staff. As this basic service transaction set develops over time,
your staff requirements will change. At some point, your robust FAQs will
reduce the need for analytical and technical skills and increase the need
for organizational skills. Very often, a two-tier staffing strategy,
called "tier triage," develops. Your original group of agents
now serve as escalation points for the 20 percent of calls the new reps
cannot handle. Initially, in this phase, the original reps serve both the
lead/escalation rep and supervisor's role. This may be the case ongoing,
or if the center grows, there may be enough volume to justify a full-time
escalation/technical agent group.
It is during this phase that command and control processes are put into
place. Center and agent performance metrics should be implemented and
compensation should be, in part, tied to key performance metrics.
Workforce management, system agent reporting, quality monitoring,
service-level goals, progressive discipline and training -- all of the
traditional call center command and control processes should be deployed
over the span of this phase and should be working reasonably well before
investing serious dollars to reach the "enterprise" phase.
Simply put, you get a big bang for the buck in this phase. You can have
the greatest CRM package in the world and it won't do you a bit of good if
you don't have any agents to handle the call or if the agents you do have
are rude or poorly informed. If you have no way of finding this out
through command and control methods, your CRM exercise will surely fail.
Technology and processes. One of the most difficult decisions to
be made in this phase of customer care is how to invest in technology.
Specifically, there are big gaps in both the telephony and CRM markets
between mid-market and top-tier vendors. These decisions require a
pull-no-punches evaluation of your business prospects. Investments in
mid-tier technology will push you quite a ways into the
"enterprise" phase of customer care, but you will have limits on
what you can do. Growth and enhanced services will be more difficult and
costly or, in some cases, impossible. Examples abound:
- Workforce management packages collect data via telephone
system-generated reports. Many mid-market telephone switches are not
supported.
- Many mid-market CRM packages only support Microsoft NT environments.
You are out of luck if you have (or acquire) any Unix systems.
- It is not uncommon for mid-market telephone systems to grow by
simply tying multiple systems together, resulting in limitations on
how big a customer care group can grow because of congestion (busy
signals) on the lines tying the systems together.
- There are fewer off-the-shelf, shrink-wrapped applications for both
mid-market telephone and CRM systems and those that do exist are less
robust.
- There are fewer people skilled in mid-market systems. As such,
custom application development is more difficult and the resulting
application has a very low market value because your particular
technology combination is not widely used.
The list goes on. Most fearful is the old adage, "You don't know
what you don't know." The frequency of surprises (especially
interoperability) is higher in these mid-market systems.
So, does this mean you have to spend millions on top-tier technology?
Not necessarily. If your business plan is modest, you have very few
customers and a stable, known technology future, mid-market systems are
great. There is also the throwaway approach. Invest in mid-tier systems
and focus your efforts into process development rather than technology
development. Processes are portable from one platform to another. When
corporate funds allow or business environments force the migration to a
top-tier system, throw out all the technology and migrate the processes to
the new platform. The key to the throwaway approach is to do it with your
eyes wide open. Make senior management fully aware that there will be a
significant investment in the future.
Of course, there is the "all-out" approach. In a perfect
world, you will have the funds and resources to go to an all top-tier
environment with all of the bells, whistles and gadgets. The most common
situation emerging companies find themselves in, however, is the desire to
go top-tier with a mid-tier budget. The trick is scalability and scope.
Technology. The watchword for telephone systems is scalability.
How does a system grow? Look for a vendor that allows you to grow without
replacing systems. Better yet, look for a vendor who uses the same card
set in all sizes of systems. This means that cards purchased for a small
system are reusable in larger systems. Make sure your system grows along
an expandable backplane. This means that additional cards plug into the
same system bus as existing cards rather than tying multiple systems
together. Don't get wrapped up in features. The trend is to move most
features and services onto the CRM platform. This will not happen on day
one, but over time.
After scalability, the next and most difficult part in system
evaluation is interoperability. Make sure you understand in detail the CTI
interface of the switch. This is absolutely imperative to developing a
comprehensive CRM tool.
In terms of price, expect to pay a 25 to 40 percent premium over
mid-market telephone systems. This is where scope comes into play. When
trying to obtain top-tier systems on a limited budget, start small. Make
do with standard features of the system. Just like the throwaway scenario,
focus on processes. Unlike a throwaway scenario, with a top-tier switch,
expanding and enhancing will not break the budget.
Processes. When looking at CRM systems, it is useful to
understand what they are and what they do. A CRM package primarily exists
to do four things:
- Provide a single, unified, intuitive agent interface,
- Process management,
- Middleware (multiple database access through a single GUI
interface), and
- Provide a unified customer view, to see the entire customer
relationship rather than an account view.
The best metaphor for this is the classic spreadsheet application. A
CRM package is like a spreadsheet package in that when you turn either one
on, they really don't do anything by themselves. Like a spreadsheet into
which you enter formulas, text and tables to create useful tools, a CRM
package allows the user to enter processes, interfaces and databases to
create useful tools. Like a spreadsheet for which you can buy pre-prepared
pages to do engineering or accounting, with a CRM package you can buy
pre-prepared modules to do help desk, field sales or customer service. In
both cases, these systems are largely what you make of them. This
understanding helps frame a scalable approach to CRM package deployment.
Scope creep is the single largest factor in CRM project failure. You
cannot implement everything in Phase I. Emerging companies have the luxury
of setting up processes any way they want. Thus, it is a logical first
step to implement standard packages (help desk, for example) with minimal
modifications. Companies with processes in place may find that the CRM
vendor has developed its package based on best practices. It pays to look
at the standard implementation and see if it makes sense for you.
In any case, in the traditional call center phase of customer care
development, the watchwords are "keep it simple!" This will
allow you to get into a top-tier CRM package without breaking the budget.
Look for a CRM package where processes can be modified with little or no
help from the vendor or system integrator. This is not to say that you
should be able to access a new database or create an entirely new process,
but you should be able to change a credit code criterion or ask for an
activity validation without spending tens of thousands of dollars.
Remember, you are not aiming for "wow" yet. You are building
basic infrastructure, laying a solid process and technological foundation
upon which a customer service mansion can be built with few complications
and remarkable speed.
Enterprise Customer Care
Services. The deployment of customer relationship management is
really part of the development of a corporate culture, a process that
never ends. With this in mind, a successful CRM implementation to
demonstrate proof of concept and the capability of the implementation team
is paramount to any other objectives.
The goal of enterprise customer care is nothing less than "perfect
customer knowledge." You gain this through process integration, and
CRM is the tool that facilitates it. If you understand the process
management nature of CRM packages and the spreadsheet metaphor, you
understand that CRM package deployment is a never-ending process. CRM
packages can touch every aspect of the business. Every process can be
tracked and managed. CRM is not just a call center or field sales tool.
The magic of CRM is that it can manage anywhere in the business and can
force communication across the organization. The trick is to identify
those processes with high customer impact and that distinguish you from
your competitor. Now, you can go for "WOW!"
Take an example of a standard implementation of a CRM call center
module. When an order is taken, the initial implementation may result in
that order being entered into the order entry system and that's where it
ends. But think about it...a CRM package is a blank page, upon which we
can design anything we want. Why not ask the shipping department to track
the status of the order so when a customer calls or inquires on the Web
page, the agent (human or computer) has "perfect knowledge"? Why
not, like a major mail order computer manufacturer, tie into the shipping
company's system and provide perfect knowledge after the order leaves the
building? Why not update a predictive dialing or e-mail list to notify
customers of delivery dates? Why not keep records of purchases and tie
marketing systems into the system and tightly target marketing campaigns?
The fact that CRM systems make it easy (relatively) to increase and
perfect our knowledge about the customer makes it all the more imperative
that companies not be tempted to stop their CRM deployment. Don't look
back, your competitors may be catching up!
Staff. We've talked about the perfect knowledge character of
services provided in this phase. The whole knowledge and process
management aspect has a significant impact on staffing, both who you hire
and how they are organized. While you are using a CRM tool to
institutionalize knowledge, you are also formalizing and documenting
processes. Thus, "just in time" agent training is possible and
your center can be staffed with true generalists. The classic urban legend
associated with this is commercial loans in a large national bank. Before
CRM institutionalized knowledge, agents had to have specialized training
to process a more complex commercial loan. Thus, the bank had two loan
agent groups, one for consumer and one for commercial. With their CRM
package, institutionalized knowledge, formal and documented processes, the
bank was able to create a single loan group, smaller than the two
individual groups that existed before. Agents follow a work flow and
contextually appropriate information is provided as necessary as they work
through the process. While they may still need a few specialists, now more
appropriately called lead agents, the number of calls
escalated/transferred becomes smaller as CRM deployment infiltrates more
and more aspects of the business.
Technology and processes. At this point, both technology and
processes center around collecting this perfect information and delivering
it in a timely manner to human or computer agents to support customers,
often before they need support. If you have followed the advice set
herein, you already have scalable technology. You understand that CRM is
an ongoing exercise touching all aspects of the corporation. The only
direction in which to travel at this point is forward.
All in all, a common theme throughout the development of customer care
is to keep each phase small and accomplished in a timely manner. Success
breeds success. Demonstrate to senior management that you have a clear
vision of where you want to be and how you plan to get there. Commit to
making the maximum use of what you have and build a sound infrastructure
before making significant investments. Establish a trend of on-time and
on-budget customer care projects. Wise investments early on will smooth
the way for enhanced customer services.
Ike Mitchell works as a consultant in the customer care industry for
Computer Sciences Corporation (www.csc.com).
He specializes in operational excellence and maximizing the operational
value of technology investments.
[ Return
To The May 2001 Table Of Contents ]
|