In The Face
Of Do-Not-Call, Creating Profit In The Call Center
By Colin Shearer, SPSS, Inc.
When the do-not-call list was conceived, many thought it sounded a death
knell for telemarketing organizations. For the smart ones, however, it's a
chance to ratchet up response rates. For while the long list of names used
to be the marketer's Holy Grail, the do-not-call list changes the rules. In
the old days, response rates of one to three percent were the norm. And so
the longer the list, the more sales it generated.
What marketers have ignored are two fundamental principles: a) untargeted
pitching alienates 97 to 99 percent of their list; and b) better targeting
of marketing messages leads to increased conversion and more satisfied
customers (who do not opt-out of marketing contact).
The do-not-call list presents a unique opportunity for marketers by forcing
them to know customers a little better and specifically tailoring products
and offers to the demonstrated preferences of each individual customer. This
is a big change from the carpet-bombing approach of yesteryear, and not only
yields better conversion rates, but also keeps customers from opting out of
receiving marketing messages. But how?
Evolving The Call Center
From the ashes of mounting restrictions against unsolicited telemarketing,
namely the do-not-call list, a phoenix may well rise in the form of the call
center. Industry analysts and experts are in agreement that marketers need
to change their methods of reaching customers. The days of casting a wide
net and hoping for a decent catch are over. Marketers must now transfer
their time and resources from getting the best call list to truly
understanding their customers and prospects and making them the right offer.
Much of this begins at the customer's main point of contact ' the call
center.
Shifting Gears From A Cost Center To A Profit Center
The call center business model has historically been based on reducing the
amount of time spent on each call to reduce costs and personnel. The
quantity and not the quality of each call was the driver. Operational
efficiency, in this model, was the overriding goal, not customer
satisfaction, customer retention, nor cross-selling or upselling for more
revenue. Rarely was a call center's operation tied in with an organization's
business goals or strategies.
With the advent of customer relationship management (CRM) initiatives in the
late 1990s, call centers took on a more strategic role. CRM systems promised
to bring all the threads together. When a customer called in, an operator in
a call center could quickly see the entire history of an organization's
relationship with that customer ' what they'd bought, when they'd bought it,
what issues had come up and how they had been resolved. Service
representatives could see what sales representatives had done and to which
direct marketing campaigns a customer had responded. Everyone could see who
was interacting with the customer.
Although CRM systems increased operational efficiencies, these initiatives
have failed to deliver significant value to customers and, as a result, have
not generated high returns. Operational CRM systems improve connections to
customers, but they don't do much to improve customer understanding, and
without good customer understanding, organizations can't be sure of the best
thing to say to each customer at a given point in time. Operational CRM
systems generate huge amounts of customer data, but have not been designed
to transform those data into information organizations can effectively act
on. Raw data has no real value until it is turned into information.
Put another way, CRM systems by themselves are great at facilitating
communication and at illuminating what customers have done in the past, but
by themselves, they do not tell an organization what to communicate. They
cannot predict what the customer phoning into your call center, at this
moment, would like to do but isn't telling you ' and that is the information
that every customer-facing channel, such as a call center, needs to sell
more efficiently and improve bottom line profitability.
The missing link is predictive analytics, which can turn the raw data
generated by CRM systems into usable information. The predictive analytic
process discovers the meaningful patterns and relationships in data '
separating signals from noise ' and provides decision-making information
about the future. Predictive analytics solutions, properly deployed and
seamlessly integrated with a CRM solution, do several things:
' They enable an organization to analyze all of its customer data and
identify patterns that predict customer behavior.
' They enable organizations to understand that certain kinds of customers
behave in ways that return a higher value to the organization ' and not
necessarily because they have purchased the most expensive product or
service (but because, for instance, they are likely to repay a loan over a
longer term, or are more likely to maintain a balance on a credit card).
' They also enable organizations to understand which types of customers pose
the greatest risk to the organization or stand to decrease their value to
the organization through default, high claims or fraud.
Because a solid predictive analytics solution can be integrated with a CRM
system for real-time use of in-session data, such a solution can enable call
center agents to act ' in real time ' in ways that hold the greatest
potential for increasing the value of the call center to the organization.
Not only does this predictive information enable customer-facing agents to
make upsell and cross-sell offers that are likely to be accepted by the
customer and that return the highest value to the organization, but it also
enables them to avoid pursuing offers with customers whose behavior suggests
significant risk.
Generally, predictive analytics incorporate techniques from four categories:
Statistical analysis refers to a collection of methods used to process large
amounts of data to uncover key facts, patterns and trends. Although there
are a number of statistical analysis procedures, the two most commonly used
are classification and segmentation. Classification uses predictor fields to
predict a categorical target field, such as which groups of people will
respond to an offer. Segmentation divides subjects, objects or variables
into a number of relatively homogeneous groups (e.g., segmenting consumers
into usage pattern groups). Use of statistical analysis to classify and
segment can help to increase the likelihood that the right offer is made to
the right person at the right time.
Online analytical processing (OLAP) enables data to be easily and
selectively extracted and then viewed from different perspectives. For
example, a user can request that data be analyzed to display a spreadsheet
showing all of a company's widgets sold in Wyoming in the month of August,
compare revenue figures with those for the same products in October and then
see a comparison of other product sales in Wyoming in the same time period.
To facilitate this kind of analysis, OLAP data are stored in a
multidimensional database, which considers each data attribute (such as
product, geographic sales region and time period) as a separate 'dimension.'
OLAP enables marketers to quickly review history and trends to take
advantage of emerging opportunities and take corrective action on developing
problems.
Data mining discovers the meaningful patterns and relationships in data and
provides decision-making information about the future. Data mining
procedures include: association, looking for patterns where one event is
connected to another event; sequence or path analysis, looking for patterns
where one event leads to a later event; classification, looking for new
patterns; clustering, finding and visually documenting groups of facts not
previously known; and forecasting, discovering patterns in data that can
lead to reasonable predictions about the future. Data mining provides a
clear picture of what is going to happen in time to change it. For example,
who the best customers might be, which customers are likely to defect or if
the right data are gathered, which carries the risk of adverse reaction to
marketing offers.
Continued...
'Creating Profit In The Call Center'
continued from previous page
Text mining analyzes unstructured textual data by finding and discovering
the patterns and relationships within thousands of documents, such as
e-mails, call reports, Web sites and other information sources. Text mining
extracts terms and phrases, and then automatically classifies the terms into
related groups, such as products, organizations or people, using the meaning
and context of the text. Text mining can be used to analyze call agent notes
and to provide real-time feedback, such as scripts that can be used to pitch
cross-sell and upsell offers. With the combination of text mining and data
mining, call center scripts can be changed instantly to reflect how the
caller matches the pattern of previous calls. As the customer speaks or
writes, the agent is immediately able to analyze that customer's current and
future needs.
All of these are powerful techniques, but analysis alone ' no matter how
sophisticated ' does not deliver value until the results can be used to make
decisions wherever and whenever they need to be made. For call centers, this
means predictive models have to integrate with the operational systems
supporting interactions and deliver recommendations based on content and
context of the call as well as on historical customer information. This has
to be done in real-time; no matter how smart the recommendation, the
customer conversation can't be stalled waiting for a response from the
model.
Call Centers As Profit Centers ' Real-World Examples
Consider the case of a major financial services provider. This firm has
moved from a product-focused to a customer-focused sales model and relies
heavily on its knowledge of customer behavior to facilitate upsell and
cross-sell efforts. Once they integrated a predictive analytical solution
with their CRM system, they were able to mine their data, develop an
understanding of different kinds of customer behaviors and use those models
in real-time to present in-calling customers with highly targeted offers. As
a result, the firm increased its conversion rate by 50 percent, even as it
decreased its marketing costs by 40 percent.
Another financial services provider wanted to find a way to turn a service
center that supported more than 2 million customer calls each year into a
profit center ' one that could facilitate the conversion of incoming
customer calls to sales opportunities. Through the use of predictive
analytics, service center operators were able to increase conversion rates
by a factor of 100. This contributed as much as $40 million to the firm's
revenues in the first year, even as it enabled the firm to decrease the
amount it was spending on other marketing endeavors. The net result was an
increase in bottom-line profits on the order of several million dollars each
year.
Other companies have had similar, some even more dramatic, successes.
Organiza-tions are increasing conversion rates for direct mail, Web, direct
sales and other channels ' and they are doing it by enabling their
customer-facing agents to work, in real-time, with customer behavioral
models discovered through the integration of predictive analytics software
and their CRM systems. Not only are these efforts improving bottom-line
profits for these companies, but they are also enabling these companies to
show real return on CRM systems that had not shown strong return in the
past.
The Customer Is King Again And There's No Turning Back
The backlash against unsolicited telemarketing and e-mails coupled with
increasing competition has forced more precise methods of selling and
marketing to customers. Customers are ' as they should have always been ' in
the driver's seat and are forcing organizations to work harder to win and
keep their business. Companies that have relied strongly on the
carpet-bombing approach to marketing are now using their call center to
better target customers at the right time with the right offers.
Achieving the right balance between providing good customer service and
discovering and acting upon every existing sales opportunity in the call
center will require more than a paradigm shift; it will require equipping
the call center agents with the tools and intelligence necessary to
understand customers. By turning the voluminous customer data available into
useful customer intelligence, the value of each customer interaction is
increased in real-time. This results in a higher degree of customer
satisfaction and increased call center profitability and a rapid return on
investment.
Colin Shearer is vice president of customer analytics at SPSS Inc. A
pioneer of data mining in the early 1990s, he was a founder of Integral
Solutions Ltd. (ISL) and the architect of ISL's Clementine system, the first
data mining tool aimed at non-technologist end-users. In December 1998, ISL
was acquired by SPSS Inc., a provider of predictive analytics. Shearer
joined SPSS to establish its data mining business and now oversees the
customer analytics business center. He is responsible for the product
management and marketing of products and solutions related to the analysis
of customer data. Colin can be reached at
[email protected].
If you are interested in purchasing reprints of this article (in either
print or HTML format), please visit Reprint Management Services online at
www.reprintbuyer.com or contact a representative via e-mail at [email protected]
or by phone at 800-290-5460.
For information and subscriptions, visit
www.TMCnet.com or call 203-852-6800.
[
Return To March 2004
Table Of Contents ] |