Module 2 - Determining how to make measurements in your call center
The Call Center Learning Center is proud to present a tutorial series
focused on performance measurement systems. This three-part tutorial
series will pull from Centerserve's research-based eToolkits and best
practices reports.
Module 1
addressed how to determine what to measure in your center. This
module will discuss how to make these measurements. Module 3 will
focus on benchmarking your center's performance.
Key questions for your performance measurement program
Designing and implementing a call center measurement program
requires you to
answer three important questions.
If you have a measurement program in place, these questions will
help you make your center more efficient and improve the
effectiveness of the program.
-
What will you measure?
-
How will you make these measurements?
-
Who will you benchmark your performance against?
|
Module 1
covered how to establish appropriate measures for your call center
environment. This module will discuss how you will go about making
the measurements.
How will you make your measurements?
Once you have defined your metrics, you will need to decide how to
measure them. As we have found with our research at the Call
Center Learning Center, there are many different ways to measure call
center performance. You should be as specific as possible about
what each of your measurements entail and how it is being made.
There are three important guidelines to consider:
-
Special considerations for certain
measurements
For example, cost metrics can be calculated by including a variety
of different "overhead" costs. Measurements related to answer
time can be difficult, since there are many points to choose from
(for example, when do you start to record call time?). Figure
1 outlines some of the special considerations you should consider
for common measurements in your center.
Measurement
|
Special Considerations |
|
Abandon rate |
At what point during the call process do you begin to
measure abandons? Do technology options (VRU, e-mail,
Web chat) cause customers to abandon before reaching an
agent? How does this positively affect abandon
rate?
|
|
Availability and agent utilization |
Does your system differentiate time spent on separate
work tasks?
|
|
Average speed of answer |
What is the exact point in the call process when a call
should be considered "answered?" How will you balance
this measurement so as not to encourage quick
disconnects?
|
|
Contact volume |
Where in the call process should the measurement be made
(this will influence abandon rate, service level, cost
per contact, and average speed of answer)?
|
|
Cost metrics |
What specific costs are included in your calculations?
|
Customer
satisfaction |
What technology options are available to help collect
customer service data (Web surveys, automate VRU
surveys)?
|
|
Handle time |
Will you use a short timeframe or long timeframe for
this measurement? Should agents be evaluated on an
individual or group basis, or incorporate both?
|
First contact
resolution rate |
How will you incorporate this measurement across all
customer contact options?
|
|
Service level |
Can you separate this measurement by media channel?
|
|
Span of control |
What situations will cause you to measure this metric?
|
|
Turnover rate |
For what timeframe will you measure turnover?
Weekly? Monthly? Quarterly? Yearly?
|
Figure 1 - Special considerations for call center measurements
A valuable resource to guide
you through the specific "how" and special considerations for each
call center performance metric is the
Call Center Measurement eToolkit.
It provides a complete, alphabetized guide to performance
measurement that is easy to understand and easy to implement.
Each measurement area includes complete definitions, common mis-uses
of terms, how to compute each measurement area of your call center,
goal-setting tips and recommendations for improving performance in
each area.
-
Impact of multi-media
Multi-media contacts have created many issues around the "how" of
call center measurement. One of the prime examples currently is
e-mail. Many centers are finding it difficult to get accurate
measurements for metrics like cost per e-mail and first contact
resolution. In other cases, such as abandon rate, multi-media
contacts make the measurement less important. In defining the
"how," be sure to develop strategies and systems for handling the
growing multi-media contacts your center will receive.
-
Consistency for
benchmarking
Consistency in measurement is very important. Ensure that
everyone is working the same way to secure common call handling
procedures for internal benchmarking comparisons between different
agents, shifts and timeframes.
Although every call center is different, with unique goals and
challenges, it is useful to be aware of what other call centers are
doing. If you choose to measure differently than other call
centers, your benchmarking process will not be productive since you will
not be comparing "apples to apples." Remember to strive for
consistency in all procedures and processes to create measurements that
are useful to your center.
More on the benchmarking component of
the call center performance measurement system will be covered in Module
3. To obtain call center best practices and benchmarking
information right away, take a look at the
Call Center Best Practices - Operations Edition.
The key to successful measurement processes is to define your
measurement methods to serve your center's unique characteristics and
needs. The three steps listed in this tutorial provide guidance
for defining the "how" of performance measurement in your call center,
but ultimately, your call center leadership will have to determine the
specifics of how to collect each measurement. The
Call Center Measurement eToolkit provides in-depth information and
data for 13 popular call center measurements, including specific details
on how to conduct each measurement and methods to improve your
performance.
Stay tuned for Module 3 of this series to learn more about
benchmarking your call center performance internally and externally.
|