We're all hoping that the right metrics we collect in our call center will
deliver insight to productivity, resource needs, and member satisfaction. But
many face the problem of not understanding what data is relevant and are missing
the right metrics to actually deliver truly useful information. Let's look at
a real-world example:
In an effort to centralize operations, a member service center manager was
asked to take on the loan department's call volume. To plan for the increased
number of calls, the manager asked the loan department about their call volume.
"We only handle about 150 calls a day," said the department team leader.
However, on the first day of taking on the new call volume, the member services
center received a total of 740 calls!
After some investigation, the member service center manager discovered that,
in actuality, the loan department only handled 150 calls each day, while the
other 590 calls were being abandoned before they got to a live service representative.
It is easy to assume that this is a one-off situation and perhaps an extreme
illustration, but the important point is how you measure is equally critical
as what you measure.
Let's take a look at a common metric used to gauge performance - hold time.
A member calls into the contact center to inquire about a loan. First, she waits
for her call to be answered, typically a wait time of 30 seconds. The Member
Service Representative (MSR) taking the call must then transfer the member to
the loan specialist, who is not immediately available, and another 60 seconds
are spent on hold. The loan specialist takes the call, but must speak to a supervisor
to complete the transaction. The member spends another 90 seconds on hold, then
the call is finally concluded.
In this imaginary example, the hold time for the call was 180 seconds. However,
the member service center manager argues that hold time was only 30 seconds,
the amount of time the member was originally on hold because "that is what
the report said." The problem with reporting on call centers is that unless
every piece of the center's infrastructure is linked to the same database, the
result is - at best - a partial view of what's happening, and in the worst case,
something completely inaccurate.
So why is it so difficult to obtain accurate metrics? Think about the number
of independent systems in a call center: the phone switch or PBX, the Automatic
Call Distributor (ACD), the Interactive Voice Response unit (IVR or auto-attendant),
the Core Processing system, the e-mail system, CTI (computer telephony integration),
and the list goes on. Typically each system has its own database and administrative
interface, so if you want to create an abandoned call report, segmented by member
category, you may have to combine data from several databases to determine the
true number of abandoned calls.
Even if you were able to get the raw data for abandonment, you still need to
find out why people are abandoning. For instance, you may know that you have
an eight percent abandonment rate, but where did they abandon? If they are all
abandoning in the IVR, then maybe there's an issue with the menu, and you must
be able to take a look at exactly what point in the IVR they're abandoning.
If they did abandon, did they call back? And so on.
The idea of a central data repository for all applications is one that has
been adopted at many major companies as they implement Enterprise Resource Planning
(ERP) applications to run every aspect of their organization. While the benefits
to a call center may be measured in the millions or hundreds of thousands per
year rather than billions, the underlying reasons are the same. You can reduce
duplication of effort (example: a member sends in an e-mail inquiry, then calls,
requiring two MSRs to respond), administration takes seconds rather than weeks
(example: changing the IVR script to route certain calls to an MSR with a certain
skill-set could have repercussions throughout the system, and each database
would have to be updated to reflect the change) and of course, metrics.
If you want to run your contact center at the highest level, you need accurate
metrics that cover every step of the interaction, and you need to be able to
extract information from those metrics that provide a business-level understanding
(examples: cross-referencing member satisfaction against a specific MSR's performance,
understanding how the cost of voice interactions compare to another channel
or how time spent with a member to educate them on Internet bill pay is reflected
in a lower number of calls to the contact center).
The challenge that credit unions face with their contact centers is an historical
one. They typically have adopted a given technology to solve an issue, for instance
purchasing an IVR to improve self-service or an ACD to improve routing. Unfortunately,
each new technology purchased only decreases their ability to get useful metrics
from the entire system. This is where a communications management system that
integrates all inbound and outbound channels can really deliver value to your
For more information on contact center metrics and integrated communications
management systems, register for one of Apropos Technology's upcoming FREE online
seminars: "New Age Metrics for the Contact Center," or "Getting
Members to Use Online Services." To register or find out more, please visit