Is Empathy Effective for Customer Service? Evidence From Call Center Interactions

Abstract

This study examines the nature and value of empathic communication in call center dyads. Our research site was a multinational financial services call center that we came to know through grounded study techniques, including analyses of 289 stressful calls. Examining calls as communication genre revealed that agents and customers have conflicting organizational, service, and efficiency needs that undermine communication. But three types of empathic expression can mitigate these conflicts in some interactions. Affective expressions, such as “I’m sorry,” were less effectual, but attentive and cognitive responses could engender highly positive responses although customers’ need for them varied tremendously. Thus, customer service agents must use both diagnostic and enactment skills to perform empathic communication effectively, a coupling that we call empathy work.

Full text

Article
Is Empathy
Effective for
Customer
Service? Evidence
From Call Center
Interactions
Colin Mackinnon Clark
1
, Ulrike Marianne Murfett
2
,
Priscilla S. Rogers
3
, and Soon Ang
2
Abstract
This study examines the nature and value of empathic communication in call
center dyads. Our research site was a multinational financial services call
center that we came to know through grounded study techniques, including
analyses of 289 stressful calls. Examining calls as communication genre
revealed that agents and customers have conflicting organizational, service,
and efficiency needs that undermine communication. But three types of
empathic expression can mitigate these conflicts in some interactions.
Affective expressions, such as ‘‘I’m sorry,’’ were less effectual, but attentive
and cognitive responses could engender highly positive responses although
customers’ need for them varied tremendously. Thus, customer service
1
University of New South Wales, Sydney, New South Wales, Australia
2
Nanyang Business School, Nanyang Technological University, Singapore, Singapore
3
Ross School of Business, University of Michigan, Ann Arbor, MI, USA
Corresponding Author:
Priscilla S. Rogers, University of Michigan, 701 Tappan St., Ann Arbor, MI 48109, USA.
E-mail: psr@umich.edu
Journal of Business and Technical
Communication
27(2) 123-153
ªThe Author(s) 2012
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/1050651912468887
jbtc.sagepub.com
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

agents must use both diagnostic and enactment skills to perform empathic
communication effectively, a coupling that we call empathy work.
Keywords
empathy, empathic communication, call center communication, customer
service, genre analysis, dyadic communication
Customer: I sent in the request, but you all didn’t process until I call. Then
the lady immediately do for me. . . . I faxed the request on 29 December,
so expecting that it should be done. Then I call on the 12th [of January],
and they just managed to do on that very day.
Agent: Mm, mm. I see. I get what . . .
Customer: [interrupts] So the delay is already almost 2, 3 weeks already.
Agent: Okay, so sorry to hear that ma’am. Don’t worry. You do a check by
the end of today. If don’t have it tomorrow, you give us a call again.
In the preceding excerpt of a customer service call, the call center agent
empathizes with the customer (e.g., ‘‘I get what [you mean],’’ ‘‘Okay, so
sorry to hear that,’’ ‘‘Don’t worry’’). What else could the agent do, given
that his coworkers had evidently failed to perform the service requested?
The agent might have offered to check the status of the request and report
back to the customer. As for the agent’s expressions of empathy, they seem
peripheral to the task, do they not? Interactions like this raise the question:
Is there a role for empathy in customer calls?
The ability to empathize with customers is regarded as an important
competency for customer service agents in call centers (Bordoloi, 2004;
Burgers, Ruyter, Keen, & Streukens, 2000; D’Cruz & Noronha, 2008; Dor-
man & Zijlstra, 2003; Korczynski & Ott, 2004; Pontes & Kelly, 2000). Yet,
the nature of agent–customer interactions has raised doubts about the neces-
sity and desirability of empathic communication in this context. Although
call centers exist to foster customer relations through service, their setup
distances agents and customers. Relational small talk occurs less often in
telephone conversations than in face-to-face ones (Halbe, 2012). Agents
and customers are strangers, often from different cultures, and will probably
never interact again. Their respective roles as service giver and service
receiver are asymmetrical. Agents are taught to ‘‘maintain an emotional dis-
tance’’ yet to build rapport (Thompson, Callaghan, & Van den Broek, 2004,
p. 140). They are instructed to disguise their own feelings, yet to identify
124 Journal of Business and Technical Communication 27(2)
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

with customers’ feelings (Hochschild, 2003; Thompson et al. 2004, as cited
in Deerie & Kinnie, 2002).
But this imperative to enact empathy pales against other call center
goals, such as service efficiency. Customers want their queries dealt with
pronto; they want first-call resolution (Callaghan & Thompson, 2002).
Agents are pressed for time as other callers wait in the queue and as man-
agers monitor call length, rewarding shorter calls. In this customer service
environment, then, is empathy needed?
This study examines the nature and value of empathic communication in
call center dyads (hereafter labeled customer calls). We wondered if
empathic communication contributed meaningfully to agents’ customer ser-
vice, and if so, how? Therefore, we investigated several basic questions:
What is the communicative genre of a customer call? What agent and cus-
tomer purposes may invite (or disinvite) empathy? How is empathic com-
munication performed in calls? And finally, what do customer responses
to empathic communication tell us about its effectiveness in this context?
Using a grounded methodology including analyses of calls, we identified
three types of empathic responses that contributed to the success of some
calls. But we also found that expressing empathy was not always a good
thing—some customers wanted no empathy whatsoever. The fact that
empathic needs differed from customer to customer led us to propose that
just learning ways to express empathy is not sufficient for success; rather,
agents should be taught to do empathy work. We define empathy work as
listening attentively to assess the need for empathy and providing the nec-
essary communicative responses to meet that need expeditiously.
To the best of our knowledge, this is the first study that focuses on empathic
communication using a discourse analysis of interactions between call center
agents and customers. Some research has explored empathyas an interpersonal
skill. Operational definitions and scales were developed, and students were
scored on their ability to express empathy (Rogers, 1951, 1980), for example.
But this research was criticized because it used students, was too tightly
focused tocapture a full range ofexpression, and emphasized decontextualized
skills (Clark, 2007). To this day there is little universal agreement on what
empathy is, let alone how it is performed, a state of affairs that we detail later.
Research on call centers mushroomed after they were introduced in the
early 1990s as an efficient and cost-effective means of delivering customer
service. But studies examining the calls as communicative texts involving
both agent and caller sides of the interaction remain surprisingly few. Fri-
ginal (2009) provides a comprehensive linguistic description of calls and
identifies features that may cause miscommunication; Zu, Wang, Forey,
Clark et al. 125
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

and Li (2010) compared English and Chinese usage and described the generic
structure of calls; Forey and Lockwood (2007) identified communication break-
downs in call dialogue; Adolphs, Brown, Carter, Crawford, and Sahota (2004)
examined politeness and involvement strategies in health advisory calls; and
Cowie (2007) studied attitudes toward various English accents and their impli-
cations for training. Several studies also identified various types of responses
used by call center agents, including responses related to empathy, and the
impact of these responses on call effectiveness. Rafaeli, Ziklik, and Doucet
(2008) found that customers rated the quality of service interactions higher when
agents used specific customer orientation behaviors such as providing emotional
support. Our own research found that agent expressions of empathy correlated
with teamwork, an achievement we characterized as solidarity (Clark, 2011;
Clark, Rogers, Murfett, & Ang, 2008). We build on these studies here.
Research Context and Analyses
Our research site, which we pseudonymously call ABC Company, is a call
center for aftermarket financial services located in Singapore. As a multicul-
tural ‘‘showpiece of Asian capitalism,’’ where English is fully indigenized as
the language of business, government, and education (Clark & Rogers, 2005,
p. 12), Singapore is a rich and widely relevant context for studyingbusiness
communication such as the customer calls we investigated.
To examine customer calls, we used a grounded study approach (Strauss &
Corbin, 1990). This involved observing call center operations, shadowing
agents, interviewing 26 agents about their calls, communicating regularly with
the supervising manager, and analyzing multiple sets of calls from our corpus of
587 calls in English (often calls were in Singlish, a distinctive Singaporean Eng-
lish influenced by Southern Chinese dialects). Early in our introduction to this
customer service environment and prior to interviewing, we analyzed a set of
75 calls selected by the supervising manager, 25 calls each from high, medium,
and low performers. From these we learned a great deal about communication
features related to performance. Then we extracted from our corpus a set of
289 calls in which the agent was under social stress as a result of caller aggression
or ambiguity (Dorman & Zapf, 2004). Independently analyzing these calls, we
identified types of responses that agents can meaningfully use for relationship
building and conversational control (Clark, 2011; Clark et al., 2008). So by now
we are well acquainted with this service environment and its interactions.
For this study of empathic communication, we reviewed our set of 289
stressful calls, earmarking those in which agents demonstrated either a high
or no use of relationship-building responses that we associated with
126 Journal of Business and Technical Communication 27(2)
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

empathic communication: being attentive, offering emotional support, and
anticipating needs. We flagged 68 calls for closer analysis, 56 in which the
agent made a high effort to use these responses with customers and 13 in
which the agent made no relational effort whatsoever.
Two of us (Murfett and Rogers) independently analyzed agent–customer
turns in the high-effort calls, completing flowcharts of agent responses that
we had characterized as empathic and customer reactions that indicated the
negative or positive impact of the agent’s effort. Using the totality of each
customer’s response as our measure, we holistically scored if the customer
left the call satisfied,moderately satisfied,orunsatisfied. There was no dis-
crepancy in our scoring. As a further check, one of us (Clark) examined
calls that were difficult to score; he agreed with the original scoring. There-
after, we treated customer-satisfied calls as exemplary. We scrutinized calls
with moderately satisfied or unsatisfied scores as potential sites for
empathic communication using the linguistic technique of substitution.
Three of us (Clark, Murfett, and Rogers) independently analyzed calls in
which the agent made no effort to build a relationship with the customer,
flagging calls that would benefit from empathic communication. We then
discussed how the empathic responses that we identified earlier might have
been used to improve customer satisfaction in these flagged calls. These
analyses, coupled with our fieldwork at the research site, showed us that
agents enacted empathic communication by listening closely to customers
(attentive empathy), offering emotional support (affective empathy), and
anticipating needs (cognitive empathy). We discuss these three types of
empathic communication in detail later, but to understand their relevance,
we first examine the customer call as a communicative genre, particularly
the purposes that bring agents and customers to the call.
What Is a Customer Call?
Genre theory provides a foundation for examining the viability of empathic
communication in the call center context. Viewed as a genre, the customer
call is a typified communicative action in response to a recurring situation
that is recognized by its form, content, and shared purposes (Bakhtin, 1986;
Miller, 1984; Swales, 1990; Xu et al., 2010; Yates & Orlikowski, 1992).
Call Form and Content
In form and content, aftermarket customer calls are dyadic phone conversa-
tions with considerable turn taking, beginning with the call center agent’s
Clark et al. 127
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

opening, then promptly proceeding to the customer’s concern followed by
interchanges to address it, and ending with a quick close. The calls in our
corpus averaged 5 minutes in length, with the stressful calls among them
averaging 30 seconds longer. Similar to the calls in Xu et al.’s (2010) study,
these calls tended to proceed in six phases:
1. Greeting. Agents clicked an icon on the computer screen to accept the
call and repeated a standard greeting: ‘‘Good morning, ABC Company,
[Agent’s name] speaking. How may I help you?’’
2. Identifying. Customers identified themselves, and agents requested an
account number or an identity card number (issued to all Singaporean
citizens). If the customer was a foreign national, agents requested a
passport number. Agents then typically asked callers to wait while they
retrieved the relevant customer account data. If the caller had multiple
accounts, the agent asked which account was the topic of the call (e.g.,
‘‘Is this about your endowment policy or your son’s critical illness
coverage?’’).
3. Defining. Agents then asked callers how they might help them with this
account. Caller responses ranged from requests for information,
descriptions of problems, or complaints about service.
4. Negotiating. Agents responded with supplementary questions if needed
(e.g., ‘‘When did this happen?’’ ‘‘Have you discussed this with your
bank?’’).
5. Resolving. Agents then resolved the call in a number of ways, depend-
ing on the nature of the call. The agent might provide information,
adjust the account, or refer the caller elsewhere (e.g., to another depart-
ment or another organization, such as the customer’s bank, if payment
processing was the issue). The agent would pass the call to a superior if
the caller’s request required special approval or would refer the cus-
tomer elsewhere if the request was outside the call center’s scope of
service.
6. Closing. Agents asked if the customer needed anything else, and if the
customer had no further concerns, they were expected to say, ‘‘Thank
you for calling ABC’’ before hanging up. If needed, agents then revised
customers’ on-screen information and wrote notes about the call before
proceeding to the next call.
Phases 1, 2, and 6 (greeting, identifying, and closing) each typically
happened very quickly, in seconds, whereas phases 3, 4, and 5 (defining,
negotiating, and resolving) constituted the heart of the call. At times, phases
128 Journal of Business and Technical Communication 27(2)
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

3, 4, and 5 were recurring and iterative. For example, it sometimes became
clear during the resolution phase that the issue was different or more com-
plex than the customer initially indicated, so the call reverted back to the
defining stage. If a call was long, it was usually stuck in these phases.
The degree to which scripts are used for this process varies with a call
center’s purpose and management. Scripts are used less often than people
might assume. As Forey and Lockwood (2007) observed, ‘‘The notion that
call center discourse is scripted and predictable is outdated’’ (p. 323). For
highly repetitious tasks that require uniformity and standardization, such
as sales calls, scripts may play a greater role. But there is less scripting
of aftermarket calls such as those we studied because they necessitate more
customization to complete the service (Deery, Iverson, & Walsh, 2004).
Form and content expectations for the customer-call genre suggest where
and how empathic communication might be used (e.g., to acknowledge a
customer’s explanation of a problem or to mitigate an adversarial dynamic
during resolution), and research shows that agents have considerable liberty
to express it. What analyses of these external features do not fully reveal,
however, are the needs that the customer-call genre is intended to meet.
As Swales (1990) explained, the purpose of genre is more difficult to get
at and ‘‘may require the analyst to undertake a fair amount of independent
and open-minded investigation, thus offering protection against a facile
classification based on stylistic features and inherited beliefs’’ (p. 46).
Determining if empathic communication is integral to the customer-call
genre, then, requires some understanding of this genre’s purposes.
Shared Purposes and Divergent Needs
Our analysis of the call center context reveals that customers and agents
share three purposes for calls: (a) organizational support, (b) service com-
pletion, and (c) efficiency. But these shared purposes serve divergent needs
for customers and agents (see Figure 1). We describe these shared purposes
and divergent needs as well as how empathic communication might be used
to mitigate the tensions that they evoke.
Organizational support. Both the agent and the customer are in some way
dependent on the organization for support. The customer obtained a product
or service from the organization and needs some sort of information or help
in conjunction with it. Customers’ degree of dependence varies. But in some
instances, it is high, such as in the case of an illness or loss for which the orga-
nization provides insurance.
Clark et al. 129
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

The agent depends on the organization for employment and the quality
of life and social status such employment provides. This dependence is
multifaceted. Because call centers are typically outsourced, agents fre-
quently serve two masters, the center that monitors their performance and
the entity that buys that center’s services. Agents must understand the
products and services of the entity purchasing their labor, meet the perfor-
mance criteria of the center for which they work, and address the needs of
the customers who call. These demands are complex and sometimes
competing.
Divergent
Needs
Shared
Purposes
Divergent
Needs
•Maintaining employment Organizaonal support •Obtaining products and
services
•Idenfying customer’s real
concern
•Demonstrang experse
•Depersonalizing
•Finishing the task fully to
avoid repeat call
Service compleon •Expressing concern
•Receiving informaon,
remedy, or aenon
•Saving face
•Keeping composed if
feeling mistreated, upset
about a personal loss, or
fearful
•Compleng work
•Moving to the next call
expediously
Efficiency •Finding someone to help
or listen now
• Geng full and
immediate aenon
remotsuCtnegA
Figure 1. Shared purposes and divergent needs for the customer-call genre.
130 Journal of Business and Technical Communication 27(2)
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

Thus, customers and agents use the customer call for the shared purposes
of organizational support. Although their needs for support differ, neither
the customer nor the agent is typically an organizational insider. So their use
of the call as a means of active recourse is restrained. There is only so much
that either the customer or the agent can do. To the degree that it is profes-
sionally appropriate (which is a tricky issue), the agent may empathize with
the customer from the outsider perspective that they share (e.g., ‘‘Yes, it’s
frustrating that we can’t get this done’’).
Service completion. Both the customer and the agent need the call to get
the work of service done. At first glance, the customer may seem to sit in
the catbird seat, the agent being a mere gofer. But they play important, inde-
pendent roles.
Customers call to get information, receive help, or simply to be heard.
They are not expected to have expert knowledge, to behave particularly
well, or to prepare for the call. It is okay for a customer to be inarticulate.
Discourtesy and unkind words go unchallenged if the agent is an effective
emotional laborer (Hochschild, 2003), paid to turn the other cheek. Custom-
ers can also end the call if it is not going well (e.g., ‘‘Okay then. I’ll call
another time or handle it myself’’). But the customer’s role in getting ser-
vice is not demand free. Customers need sufficient understanding to com-
municate their reason for calling and to interpret the agent’s response,
which may require some technical knowledge about the products or services
at issue. A customer who does not understand may experience a loss of face.
We heard many a customer laugh apologetically or admit their embarrass-
ment when they failed to express themselves well or to grasp the agent’s
meaning. Customers may also be challenged by their need to maintain
emotional control. A customer’s need for service may stem from a wide
range of personal insecurities and crises, such as financial loss, life-
threatening illness, or the death of a loved one. Callers may be grief-
stricken, fearful, frustrated, or simply confused about the product or service
they purchased. So, it is not inconceivable that they may have difficulty
keeping such strong emotions under control.
On the other side of the call, the agent must serve with a smile, providing
expert knowledge on demand. Agents are paid to serve, and they are
expected to do it well. Their bosses listen to their calls, and customers are
asked to evaluate them. Call centers keep records on first-call resolution;
follow-up calls expend agent time and are costly. Completing the service
fully in one go, then, is a paramount goal for agents. Some observers have
noted that surveillance is rarely applied to its fullest extent because agents
Clark et al. 131
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

resist, and trade unions may get involved (Deery et al., 2004; Taylor, Mul-
vey, Hyman, & Bain, 2002). Nevertheless, monitoring is important in eval-
uating agent performance.
Customer and agent roles differ considerably, but they are codependents
when it comes to service completion. Customers are not expected to be
experts or even particularly cordial, but their lack of knowledge or emotional
control may hamper an agent’s ability to provide what they need. As the gen-
re’s keeper, the agent is under pressure to set emotion aside, find out what the
customer needs, and meet those needs while Big Brother watches with reward
power. Making customers feel comfortable may encourage their cooperation
to get the service done. We have heard both parties expressing empathy for
their counterpart (e.g., the agent comments, ‘‘I’ve found this provision in the
policy difficult to understand too,’’ and the customer replies, ‘‘I really
appreciate your patience in explaining this to me’’).
Efficiency. Efficiency is in the best interest of both the customer and the
agent. These calls are business transactions, not social interactions, and
the system for handling them is set up accordingly. Calls are stacked in a
queue and allocated to agents via an automatic call distributor (ACD). Cus-
tomers may call a center often, but they do not necessarily speak to the same
agent. Moreover, customers seek to obtain service and move on as quickly
as possible. As one agent observed, ‘‘Most customers just want to come on
and get their query dealt with, they don’t really care whether [you’re] . . .
best friends with them at the end of the call’’ (Callaghan & Thompson,
2002, p. 245). But this shared need for efficiency is conflicted.
Sometimes customers want more agent time, such as when they are
registering a complaint. We found that during stressful calls, customers
often made remarks such as, ‘‘Do you understand me?’’ or ‘‘But I already
submitted a form that I don’t want this.’’ Customers may also want more
agent time because they feel that they deserve it after waiting in a long
queue, negotiating an automated entry system, or both (e.g., ‘‘I had to dial
this number three times before I could figure out which category would get
me to you!’’). Customers may decide to linger as well, taking advantage of
the opportunity to be heard or asking off-topic questions (e.g.,‘‘Oh, and
while we’re on the phone, let me ask you about my medical policy’’). Cus-
tomers pick when to call, whether to prolong a call, and when to hang up.
They can choose to hasten the call or take their time. But for agents, effi-
ciency is tied to job performance. Haste is not a choice; it is a standard.
Technology enables management to obtain data on agent call times, the
number of calls handled, unanswered calls, and customers’ abandonment
132 Journal of Business and Technical Communication 27(2)
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

rate (Armistead, Kiely, Hole, & Prescott, 2002; Deerie & Kinnie, 2002).
Agents’ conversational abilities are difficult to measure, but monitoring
their efficiency is easy, unobtrusive, and quantitative, a manager’s dream.
It is not uncommon for call center managers to post individual agents’ call
times and number of calls handled (Bain, Watson, Mulvey, Taylor, & Gall,
2002). Thus, customers and agents have different efficiency needs that may
conflict. Customers expect to have information and answers on demand
(e.g., ‘‘Why don’t you have that information?’’); they may choose to extend
the call out of, say, frustration, anger, curiosity, or loneliness. Agents, how-
ever, must labor under efficiency’s heavy hand.
Our data include calls in which customers express some awareness that
agents must bring the call to an end (e.g., ‘‘I know you have other callers wait-
ing, but before you go, please be kind enough to repeat what you said I should
ask my banker’’). When customers showed no understanding of time pres-
sures, agents sometimes reminded them (e.g., ‘‘Like you, I’m anxious to find
what you need, but you’ll have to give me a moment to look it up’’).
We have identified three purposes that customers and agents share and
explored divergent needs that are associated with these purposes in cus-
tomer calls. Customers and agents need organizational support. Both
depend on the organization in some way, an organization that restricts the
manner in which support is rendered by imposing the call genre as the meet-
ing place. Customers and agents also want service to be done efficiently.
But their divergent service and efficiency needs tug at each other in funda-
mental ways, raising questions about the viability of the customer-call
genre. As Swales (1990) observed, ‘‘when purposive elements come into
conflict with each other . . . the effectiveness of the genre as a sociorheto-
rical action becomes questionable’’ (p. 47). It is not surprising that in practi-
tioner (e.g., Dawson, 2005) and research publications (e.g., Dorman &
Zijlstra, 2003), call center service is reported as less satisfactory than what
organizations anticipated. Might empathic communication make the cus-
tomer call more effectual for customers and agents, or does empathy dis-
tract from the fulfillment of organizational support, service, and
efficiency needs? These questions led us to explore what empathy is and
how it is communicated in customer calls.
What Is Empathic Communication?
The literatures of customer service, marketing, organizational psychology,
and counseling discuss the importance of empathy, but no definitive defini-
tion of it has emerged. As Clark (2007) observed, ‘‘empathy has multiple
Clark et al. 133
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

meanings that have been conceptualized variously’’ (p. xii). Some descrip-
tions of empathy in these literatures were well suited to the interactions we
studied. Hogan (1975) captured the essence of these interactions in his
description of the empathic speaker and listener:
An empathic ‘‘actor’’ will typically tailor his performances to the needs and
requirements of his or her audience; the actor will also tend to be an effective
speaker as a result of an ability to anticipate the informational requirements of
his or her listener. ... On the other hand, the empathic ‘‘audience person’’
will tend to be a tactful and appreciative listener, skillfully encouraging oth-
ers in their performances, thereby providing an accepting and generally
rewarding context for interaction. (p. 15)
In Hogan’s (1975) description, our literature review, and analyses of
customer calls, we have found three types of empathic communication:
(a) attentive, (b) affective (sometimes called experiential), and (c) cognitive
(or observational). Using our call center data, we define these types of
empathy, describe responses that express them, and suggest possible inhibi-
tors and outcomes of each type (see Table 1).
Attentive Empathy
Being a ‘‘tactful and appreciative listener’’ (Hogan, 1975, p. 15) coincides
with definitions connecting empathy to attentiveness. For example, Ford
(1995) regarded empathy as ‘‘attentiveness to customers and employees
with their best interests at heart’’ (p. 75). Parasuraman, Berry, and Zeithaml
(1991) defined empathy as ‘‘caring individualized attention’’ (p. 6).
Winsted (2000) described it ‘‘primarily as showing an interest in the cus-
tomer, paying attention to the customer’’ (p. 402).
Our call analysis revealed that agents demonstrated attentiveness in cus-
tomer calls by behaviors associated with active listening, such as acknowl-
edging, repeating, paraphrasing, elaborating the customers’ ideas,
summarizing, and asking questions. For example, one call consisted of 23
turns, including a progression of attentive agent responses: 11 acknowledg-
ments (‘‘Yeah. Understand’’), 5 questions (‘‘Do you actually have any agent
in mind?’’), 7 clarifications or explanations of the actions (‘‘Yeah, usually
they will follow up’’), and a concluding summary to check mutual under-
standing (‘‘Okay, so what I’ll do is ...’’). The attentiveness expressed in
this call garnered a positive customer response.
134 Journal of Business and Technical Communication 27(2)
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

Table 1. Empathetic Communication in Call Center Customer Service.
Type of Empathy Definition Expressions Inhibitors Potential Outcomes
Attentive Listening actively and
appreciatively to
customers
Acknowledging
Repeating
Paraphrasing
Elaborating customers’
ideas
Summarizing
Asking questions
Surface listening
Impatience
Confirming understanding
Diagnosing customers’ need for
empathy
Uncovering customers’ real
reason for calling
Affective Identifying with
customers’ feelings
Stating understanding
Offering an apology
Referring to the
experience of others
Seriousness
discounted by
either interactant
Inappropriate use
Customer
embarrassment
Customer desire
for objectivity
Communicating a shared
condition—both agent and
customer are beholden to the
organizational entity
Cognitive Assuming customers’
perspective to
provide help
Providing language
customers need for
their explanation
Proposing options
Stating what other
customers have done
Misdiagnosis of
customers’ needs
Insufficient time
to explain in
detail
Demonstrating genuine
customer understanding and
care
Moving the call toward
resolution
135
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

Agents used attentive responses to comprehend customers’ needs, thereby
providing timely service. For example, the agent in the following call used
attentive responses (italicized) to move this call toward service resolution.
Agent: This is actually for Mr. [XXX]? Can I know how are you related to
him?
Customer: I’m his daughter.
Agent: Ah, okay Ma’am. So how can I help you?
Customer: Just ah, last month we took a loan of 9,500 from this policy with
the interest of 6%...
Agent: Mhm
Customer: I would like to check if let’s say he wants to do a payment ah now
of the full amount . . .
Agent: Is it going to be a full repayment?
Customer: Correct
Agent: Okay, I’m sorry, ma’am, because for loan value we don’t reveal to
third party, but if you want I can actually call him directly, or I can post
the information to him
Customer: Ah . . .
Agent: Which do you prefer?
Customer: Okay, you can ...ifyoucall him, mmm, ah . . . you can call him,
but I do not know if he’ll understand, that’s why he asked me to call.
Agent: Ah
Customer: Because . . .
Agent: I just need to check. Can I . . . you know, whether I can reveal [the
loan value information] to the daughter or not.
Customer: Yeah . . .
Agent: Would that help?
Customer: Yeah, you can call him.
By using attentive empathy, the agent was able to recognize that there was
an organizational barrier preventing her from helping the customer—agents
are not allowed to reveal some information to third parties without the pol-
icy owner’s permission. The agent used a series of six questions to find out
if her hunch was correct while taking steps to scale the organizational bar-
rier to service completion.
Our call analysis and agent interviews also revealed two inhibitors of
agent attentiveness: surface listening and impatience. We found that agents
can fake attentiveness. Deploying acknowledging, repeating, and para-
phrasing expressions passively (e.g., ‘‘Aha, yes I know’’; ‘‘You’re saying
that . . .’’); using a flippant tone and rushed delivery when acknowledging;
136 Journal of Business and Technical Communication 27(2)
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

and repeating customers’ words were all signs of inattentiveness. Some cus-
tomers’ responses suggested that they knew when agents were bluffing
attention (e.g., ‘‘Yes, but do you hear what I’m saying?’’).
Affective Empathy
Affective empathy has been defined as identifying with what another person
is feeling or responding with the same emotion as that of the other person
(Aggarwal, Castleberry, Ridnour, & Shepherd, 2005; Parasuraman,
Zeithaml, & Berry, 1988). It is sometimes associated with the German
Einf ¨
uhlung, ‘‘feeling one’s way into,’’ or Mitgef ¨
uhl, ‘‘feeling with some-
one’’ (Wispe, 1986). Clark (2007) considered it ‘‘a mode of experiencing’’
(p. 6). ‘‘Identifying with another person’s experiences and feeling concern
for them when things go wrong’’ could be characterized as affective empa-
thy (Axtell, Parker, Holman, & Totterdell, 2007, p. 143; see also Betan-
court, 1990; Egan, 1990).
In our data, agents expressed affective empathy by offering emotional
support (‘‘I understand’’; ‘‘That must be difficult’’) or an apology (‘‘I’m
sorry, can you hear me now’’; Clark, 2011; Clark et.al., 2008; Rafaeli et.
al., 2008). We also found instances in which agents referred to customers
facing similar difficulties (e.g., ‘‘Actually, other customers also called to
clarify what the letter means’’).
Affective empathic expression also has inhibitors. First, customers may
doubt the sincerity of such expressions. Agents are expected to use them,
much like expressions of courtesy. Second, affective expressions can be
inappropriately used (‘‘I’m sorry you received such bad service’’), exceed-
ing boundaries imposed by the organization to protect its reputation. An
agent should not reinforce customers’ negative views about the service,
products, or organization. Third, affective displays may embarrass the
customer (‘‘It’s not so bad really. I’ll handle it’’), or the customer may want
to keep the conversation objective (‘‘This is a generic call’’). Expressions of
affective empathy may be easy to drop into a conversation, but they should
be used somewhat carefully.
Cognitive Empathy
Cognitive empathy involves intellectually assuming the other person’s
perspective while retaining sufficient judgment to helpfully intervene.
Clark (2007) characterized it as ‘‘a mode of observation’’ (p. 10; see also
Axtell et al., 2007; Miller & Koesten, 2008). Kohut (1991) described it
Clark et al. 137
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

as a higher form of empathy that entails the two competing functions of
experiencing near and experiencing distant.
We heard agents expressing cognitive empathy by providing language
that the customer needed, proposing options for eventualities that the cus-
tomer might face, or stating what other customers have done. Customers
sometimes lacked the necessary terminology and product knowledge.
Empathetic agents filled in the blanks. The following excerpt is an apt
example of an agent’s use of cognitive empathy in helping a customer who
is struggling to articulate a request. In reassuring tones, the agent provided
the customer with the needed terminology (‘‘Golden Years Plan’’) and reas-
surance that her assumptions were understood (‘‘Yeah, correct’’), thereby
preventing the customer from losing face:
Customer: Okay, I give you the reference number, ha.
Agent: Okay, sure
Customer: [gives the number]
Agent: Okay
Customer: Yeah, because we receive the pre . . . the what, the medical pro-
gram, the gold, ah ...
Agent: The Golden Years Plan, aha
Customer: Aha. So what happens is we want to cancel this, ah . . . this, lah,
what would it, this ah . . . insurance. Is it insurance?
Agent: Yeah, correct.
Customer: So I’d like to cancel this insurance. How, how do I need ...doI
need to fill in any form?
Agent: Yeah, if you wish to we can send you the form, and then, ah, you can
send it back to us after completing it.
Customer: Could you please send to us ...
Agents also used expressions of cognitive empathy to anticipate a custom-
er’s future needs and offer solutions in advance. For example, an agent
anticipated that a customer requesting a change of address would wish to
use this new address for other policies as well (e.g., ‘‘So you need to update
your address for all your policies?’’) or that a customer would want to avoid
incurring greater costs by paying a bill (e.g., ‘‘You should go down today.
Every day a little bit of interest is incurred’’). In such cases, intellectually
assuming the customer’s perspective enabled the agent to helpfully
intervene.
Consider how cognitive empathic communication might have been used
in the following call:
138 Journal of Business and Technical Communication 27(2)
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

Customer: For the claim, right, let’s say my husband got a tumor, then he
needs to stay in a hospital. So can we use this for [that] sort of claim?
Agent: The [xyz] policy does not cover hospitalization.
Customer: So that means it only covers accident and . . ..
Agent: It covers incapacitation as well as death.
In this call, the agent stated that the policy did not cover hospitalization, but
did not elaborate on what the policy did provide until asked. This suggests
little concern for the caller, who may in fact be discussing a real problem,
not just a hypothetical one. Alternatively, the agent could have empathized
with the customer’s need to understand the coverage of his policy and
helped him move forward:
Customer: For the claim, right, let’s say my husband got a tumor, then he
needs to stay in a hospital, so can we use this for [that] sort of claim?
Agent: I see here that your policy with us covers incapacity, where the
insured person cannot work, or death. It does not cover stays in hospital for
illness. You may want to consider other policies that cover hospitalization
in case someone is sick.
The possibility of misdiagnosing customers’ needs is an inhibitor of cogni-
tive empathy. In the counseling context, cognitive empathy is described as
evolving over time. Unlike affective empathy, which can be expressed
immediately, cognitive empathy is said to require ‘‘prolonged immersion
in the broader perspective of a client’s life’’ (Clark, 2007, p. 11; Ornstein,
1979). But customer calls are one-time events that are expected to begin and
end quickly. Thus, cognitive empathy is inhibited because the customer-call
genre disallows relationship building over time.
Despite the need for efficient call resolution, some agents performed cog-
nitive empathy with good customer effect. We hypothesize that agents’ famil-
iarity with the genre may facilitate their ability to effectively enact cognitive
empathy. Call topics orbit around the particular service sold, which agents are
prepared to discuss. Agents also become familiar with the range of concerns
that customers present in these calls (e.g., ‘‘Oh I see. You’re interested in tak-
ing a loan on your current insurance policy’’). Familiarity that comes with
experience is useful unless agents presume that customers have concerns that
they do not have and thus propose irrelevant solutions.
Having examined the customer call as a genre—its form, content, and pur-
poses—and having identified three types of empathic expression and what might
inhibit their effectiveness, we return to the basic question driving this study:
Clark et al. 139
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

Does empathic communication meaningfully contribute to customer service at
call centers? Once skeptical about empathy’s value in this service context, we
have found that empathic communication plays a selective but highly significant
role in aftermarket customer calls. But our data suggest that simply learning and
repeating formulaic expressions of empathy may undercut its contribution. Con-
sequently, we propose using an approach that we call empathy work.
What Is Empathy Work?
Our analyses suggest that realizing empathy’s full potential for contributing to
customer service at call centers requires not only knowing expressions of empa-
thy but also making decisions about their use based on customer needs. That is, it
requires empathy work—listen attentively to assess the need for empathy and pro-
viding the necessary communicative responses to meet that need expeditiously.
Empathy work harbors back to Hogan’s (1975) notions of being an appreciative
listener and tailoring performance to meet audience requirements. We will now
explain and illustrate how empathy work operated in the calls that we analyzed.
Listening Attentively to Assess Need
Empathy work is work in part because it requires paying close attention to
determine customers’ need for empathy. This conclusion coincides with
Clark’s (2007) observation that ‘‘empathy involves a commitment to grasp
the internal state of an individual as accurately as possible’’ (p. 8; see also
Cochran & Cochran, 2006; Egan, 2002; Lewig & Dollard, 2003). This
‘‘commitment to grasp’’ the customer’s need for empathy requires actively
listening for clues. Kent (1993) likened it to a tenuous guessing game:
When we communicate, we make guesses about the meaning of others’
utterances, and we, in turn, guess about the interpretations that others will
give our utterances. This guesswork is paralogical in nature because no logi-
cal framework, process, or system can predict in advance the efficacy of our
guesses. (p. 5)
In our analysis of customer calls, we found eight types of responses that indi-
cate the customer’s need for empathy. In these responses, the customer invites
empathy, disinvites empathy, expresses discontent, demonstrates misunder-
standing, repeats the concern, asks the agent to repeat, requests affirmation,
or criticizes service. Table 2 provides examples of each of these response
types.
140 Journal of Business and Technical Communication 27(2)
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

Table 2. Examples of Eight Response Types Indicating the Customer’s Need for
Empathy.
Response Type Examples
Invites empathy ‘‘Slow down. You’re going too fast’’
‘‘How come nobody inform me?’’
‘‘Hum. It’s very cumbersome. . . I’m surprised . . .
‘‘It’s all very complicated ...It’svery hard to weigh the
pros and cons’’
‘‘Actually, honestly, I need this amount ... Is there any
way you can help me?’’
‘‘That’s not very good, right?. . . If you were a
policyholder yourself and you wanted to see. . . You
know what I mean?’’
‘‘So you can see how’s my feeling now.’’
‘‘Let’s say I cannot pay back.’’
‘‘Please I do need a lot of help because I cannot speak
properly. ...’’
‘‘So that means I lose money. ...Oh that is dreadful, isn’t
it’’
‘‘Oh, dear.’’
Disinvites empathy ‘‘This is a generic inquiry.’’
But to keep it short, I really don’t want to waste a lot of
time.’’
‘‘For the claim, right, let’s say my husband got a tumor,
then he needs a stay in a hospital. . . .’’
‘‘Do I need to declare the health?’’
‘‘No, you listen to me first.’’
‘‘Say, yes. Can you check for me also the...?’’
Expresses discontent ‘‘I understand but ...’’
‘‘Yah but we still have to ...’’
‘‘You are saying that if I . . . I will only be getting 55,000?
Any other thing?. . . After 14 years death benefit is only . . .
I mean you give me a heart attack.’’
‘‘You better send me a form. I’m not going to write you
a letter. I don’t have too much time you know’’
‘‘Okay, so I’ll have to check’’
‘‘So that means I lost money. Oh that is dreadful, isn’t it’’
‘‘Cannot be’’
Demonstrates
misunderstanding
‘‘What does commencement date mean?’’
‘‘I would like to withdraw half of my, ah . . . what do you
call it? Half of my savings.’’
‘‘I don’t understand about this policy.’’
‘‘I started the payment but I don’t know which month.’’
‘‘I don’t know what’s really happening. Every month I
receive . . . ’’
(continued)
Clark et al. 141
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

Using such indicators to assess customers’ need for empathy is critical.
Empathic communication may be unnecessary or inappropriate in some
cases. Not all customers seek, need, or appreciate empathy. Some want to
avoid it completely, preferring an objective answer, even if their circum-
stances are grim. For example, in the following call, a customer just wants
an objective answer to her question about reimbursement for treatments for
her newly diagnosed cancer:
Customer: This is a generic inquiry ....Let’s say about the critical illness
policy. If I’m diagnosed with breast cancer, and I don’t need the money . . .
I don’t claim?
Table 2. (continued)
Response Type Examples
Repeats the concern But I already submitted a form that I don’t want this ....
I already submitted in October the form .... As soon as
I received the letter I sent it. . . .I mean what’s the
point?’’
‘‘You not understand English or what? I said you e-mail
to me’’
‘‘Oh, a letter at. . . . Okay if I mail in a letter. . . . So this
letter is. . . ’’
Asks agent to repeat ‘‘What? Say again’’
‘‘Sorry. Sorry. Come again?’’
‘‘Say that again. I can’t hear you’’
Requests affirmation ‘‘Do you know what I’m trying to say?’’
‘‘So he has enough in his account to cover it, am I right?’’
‘‘If sufficient funds are not in the account for coverage,
you will let me know?’’
‘‘You understand my concept?’’
Criticizes service ‘‘And then you sent a letter saying ...’’
‘‘So why did [your agent] ask me to write into the
bank?’’
‘‘But the trouble is . . .
‘‘The 24th letter [from you] was, hah, hah, a mistake.’’
‘‘Did you give us a policy . . . book? Is this in the
information at all? Is it stated? Usually when you buy
these you have all these. ...’’
‘‘So your offer is not very good.’’
‘‘No. I explained to her very clearly . . . [Your colleague]
passed the phone to you. No head; no tail.’’
142 Journal of Business and Technical Communication 27(2)
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

Agent: If she doesn’t want to claim first, but next time she decides to claim,
just make sure she has all her medical reports with her.
Later in the call, this customer coupled the personal Iwith the soft modal
may when she asked how a death would be treated under the policy, but the
agent doing empathy work continued to respect the customer’s desire for
emotional distance, selecting the person rather than using you.
Customer: I may die of cancer. I may die of road accident so ...
Agent: Okay . . . as long as it’s, it’s death, [the] bonus will still accumulate up
to then. . . .
Customer: You wouldn’t ask for the cause of death?
Agent: No, [it’s] a living policy. Living policy also covers death, so we don’t
look at whether it’s a major disease or not . . . the person [has died] already,
so we just admit the death claim.
With other customers, genuinely expressed affective empathy sufficed.
Examples include the agent who admitted he was speaking too fast (‘‘Oh
sorry. I said . . .’’) or the agent who affirmed understanding in authentic
tones (‘‘Yes. Don’t worry. We will let you know if there are not sufficient
funds in your account when the payment is due’’).
Other customers invited empathy (‘‘Please I do need a lot of help
because I cannot speak properly’’) or demonstrated considerable misunder-
standing (‘‘I don’t understand about this policy’’). Customers who were
greatly discontented or critical of the service seemed to benefit most from
agents’ use of cognitive empathy with proposals of ways to meet upcoming
needs. Sometimes agents effectively used a combination of attentive, affec-
tive, and cognitive empathy, as examples in the next section show. An agent
doing empathy work detects and honors customers’ differing needs for
empathy and tailors responses to meet those needs expeditiously.
Tailoring Responses to Meet Needs Expeditiously
Our data analyses suggest that empathic expression can potentially mitigatethe
underlying, and sometimes conflicting, needs inherent in the call genre (see
Figure 1 and the fifth column of Table 1). Agents used affective empathy—
stating understanding, offering apology, or referencing what others have expe-
rienced—less than we expected, and this kind of empathy was often not essen-
tial, even for customers in personal distress. But after agents expressed
affective empathy, some customers moved to the core issue motivating their
calls, which suggests that affective expressions may help move some calls
Clark et al. 143
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

toward resolution. Moreover, in listening to calls and talking with agents, we
found that expressing affective empathy may help some agents cope when
organizational constraints limit their efforts to assist customers (e.g., ‘‘So
sorry. Multiyear projections of possible gains after bonuses, are not shown’’).
In our subset of calls in which the agent made a high effort to use empathic
communication, we found that attentive and cognitive empathic responses had
a greater effect on customer satisfaction and call efficiency than did affective
empathic responses. Some agents used attentive repetition, inquiry, and ela-
borative comments as investigative tools (e.g., ‘‘So you’re asking if hospitali-
zation would be covered?’’) for uncovering a customer’s real reason for calling
and resolving the call. The most effective of these calls were those in which the
agent employed cognitive empathic responses, such as providing terminology
that the customer needed or suggesting actions that the customer might take.
For example, in the following call, the agent employs all three types of
empathic communication, but cognitive empathy expedites the call (cus-
tomer–agent turns are numbered as locators for the subsequent analysis):
1. Customer: I used to have an insurance officer; after she left I was assigned
three different officers for different policy. This is inconvenient for me,
and so far none has contact me. I’m not sure that this is the service they
are supposed to provide in the sense that they don’t follow up with me, they
don’t check the status of my policy and then in fact when I once tried to call
them, one of them never ever pick up her phone. . . . I’m really terribly
unhappy with the service I’m getting from ABC. . . . Furthermore I want
to be assigned just one officer for all my policies. . . .
2. Agent: Yes, I understand, ma’am. So you want one common officer for all.
3. Customer: Who is contactable and when I need any help I can get her eas-
ily. And maybe someone who bothers enough to like call up and check.
4. Agent: I understand what you mean. So what I’ll do is that I shall inform
your account manager who [is] servicing your needs that she’ll assign you
an officer for all your three cases, and we should also get the officer to give
you a call as soon as possible. It may take a few days because I’m sure she
needs to read through all your policies first.
The customer’s annoyance is expressed with negatively loaded adverbs,
adjectives, and phrases (‘‘inconvenient,’’‘‘none has contact me,’’ ‘‘terribly
unhappy’’) and some sarcasm (‘‘not sure that this is the service they are sup-
posed to provide, maybe someone who bothers enough to, like, call up’’).
The agent’s ‘‘Yes, I understand, ma’am’’ (turn 2) and ‘‘I understand what
you mean’’ (turn 4) display affective empathy by acknowledging the call-
er’s feelings. But after each of these affective expressions, the agent
144 Journal of Business and Technical Communication 27(2)
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

immediately enacted cognitive empathy to meet the customer’s informa-
tional needs: by summarizing the gist of the customer’s complaint (turn
2), thereby reassuring the customer that she did indeed understand correctly,
and by explaining to the customer what she, the agent, would do to solve the
problem (turn 4). The agent also volunteered additional information for the
customer (‘‘It may take a few days because I’m sure she needs to read
through all your policies first’’). This additional information was important
in this call because delays in establishing contact were part of the reason for
the customer’s unhappiness in the first place.
If the agent had not used much cognitive empathy but had instead
focused on elaborating the affective dimension and reinforcing the caller’s
obvious anger, the agent might have said something like this in turn 2:
‘‘Yes, I understand, ma’am. Really very sorry you had a bad experience.’’
Such a response may have created some alliance between the agent and
customer, but it probably would not have advanced the call toward a reso-
lution. Instead, the agent managed customer anger well by keeping affective
empathic expression (‘‘I understand’’) brief and interjecting cognitive
empathy quickly: (‘‘So you want one common officer for all’’).
Cuttings from another call further demonstrate the interplay between
attentive and cognitive empathic responses that serves the call’s generic
purpose while addressing the users’ underlying needs. The call began with
a rapid interchange, which remained friendly although neither the agent nor
the customer observed the customary turn taking:
Customer: ...I really like to feedback to you, ah, I’m getting really very
frustrated with your, ah, telephone, ah. . . .
Agent: You mean the IVR [interactive voice response] system?
Customer: Your IVR is very nice, sound very good, but is really making me
feel like a fool, you know.
Agent: Sir, can I know what happened when you actually ...
Customer: Aaah, you know . . .
Agent: You actually trying to connect? Do you have a e-connect? I mean,
sorry, a password to actually connect ...?
Customer: I don’t have a password. I just call your normal help desk,
customer service number.
Agent: Okay . . .
Customer: But everything go . . . you know, like long, long . . .
Agent: Maybe can actually be . . .
Customer: . . . but anyway to cut this short . . .
Agent: Okay . . .
Clark et al. 145
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

Customer: I don’t want to really waste a lot of time, you know . . .
Agent: . . . yes, does not help ....
Customer: I give you [my] policy number. One. Double seven.
Agent: One. Double seven.
The agent did not apologize for the routing system by expressing affec-
tive empathy but instead investigated the caller’s experience by asking
attentive questions, such as ‘‘You mean the IVR system?’’ and ‘‘Sir, can
I know what happened?’’ When the customer admitted that he lacked a
password to expedite the process, the agent’s simple ‘‘Okay’’ did not
challenge the customer’s face. Before the agent had time to assist with the
password issue, however, the customer provides his policy number and
gives his real reason for calling:
1. Agent: Okay, so Mr. [XYZ], how can I help with your policy?
2. Customer: I get this statement . . .
3. Agent: Okay . . .
4. Customer: I just could not understand. One entry I kind of find funny.
Special cash bonus allocated for ...
5. Agent: Okay, Mr. [XYZ], what happened is that right . . . ahm, last year . . .
we actually allocate a special bonus for certain policy only . . . . Because
it’s a one-time special bonus, so we ... actually post out a check ....
6. Customer: Oooh!I see.
7. Agent: Yeah, so there’s a check actually . . . 505 . . . sent out to all the
policyholders that may be affected.
8. Customer: Aah!When would I expect to receive this check?
9. Agent: Ah, we sent out on the first of July.
10. Customer: Ooh . . .
11. Agent: . . . did you receive this check?
12. Customer: . . . last year, first of July . . .
13. Agent: . . . but not exactly on first of July, but we sent out actually in July
period.
14. Customer: I really don’t, don’t recall whether I get this, you know ....
Too long ago . . .
15. Agent: . . . because maybe expiration date six months . . .
16. Customer: Is too long ago already, I could not remember whether I
received this thing. . . .
17. Agent: Yeah, Mr. XYZ . . . let’s say this check was not banked in, right . . .
18. Customer: . . . yeah?
19. Agent: Six months later exactly, the check would . . . expire.
20. Customer: Yeah.
146 Journal of Business and Technical Communication 27(2)
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

21. Agent: Okay so, Mr. [XYZ], ahm . . . can I just trouble you that, can you
actually refer to your bank account. ...[Might] you actually have a quick
[look for] this 505? ...
22. Customer: I . . .
23. Agent: ...inJuly?
24. Customer: I . . . You see the problem is this, ah . . . first of all, ah ...I
don’t recall. Second is, ah . . . I’m having this passcard, so I don’t have a . . .
25. Agent: Oh, you don’t have an account book . . .
26. Customer: . . . yeah, you know . . .
27. Agent: Okay, Mr. [XYZ], I try to check whether ... this check [has]
been presented or not.
28. Customer: Why not, yeah . . .
29. Agent: Would you like to give me a contact number?
The customer gives his number, and the agent confirms it. The call ends
on a high note:
30. Agent: Okay. I can’t give you the information immediately ....
31. Customer: I’m sure . . .
32. Agent: It’ll probably be the end of the day or tomorrow morning.
33. Customer: Okay, no problem ....
34. Agent: . . . so I’ll try to call you back by end of today or tomorrow ....
35. Customer: Alright, marvelous ....
In the preceding call excerpt, attentive and cognitive empathy work
together to help resolve the customer’s concern. The agent signaled that she
is listening attentively by acknowledging what she hears (e.g., by saying
okay in turns 3, 5, 21, 27, and 30), asking a clarifying question in turn 11
(‘‘did you receive this check?’’), and repeating information for the customer
throughout the call. The cumulative effect of these attentive responses is to
let the customer know that he has the agent’s full attention, thereby mitigat-
ing his negative feelings from being made to ‘‘feel like a fool’’ by the IVR
system, which he had difficulty navigating to initiate the call.
The agent displayed cognitive empathy by proposing options for the cus-
tomer. In Turns 21 and 23, the agent politely posed an option as a question
rather than an imperative (‘‘Can I just trouble you that, can you actually
refer to your bank account. ... [Might] you actually have a quick [look for]
this 505 . . . in July?’’). In turn 25, the agent expedited the call by providing
the answer that the customer was trying to articulate (‘‘Oh, you don’t have
an account book’’) and, in turn 27, by offering another, very different option
to him (‘‘Okay, Mr. [XYZ], I try to check whether . . . this check [has] been
Clark et al. 147
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

presented or not’’). This last option is the solution that the customer was
seeking, and he leaves the call entirely satisfied (turn 35). The agent
efficiently diagnosed the customer’s needs—not just the customer’s need
for information but also his need for respect—and, by offering different
options (the latter involving more work for the agent), demonstrated genu-
ine care for the customer while moving the call toward closure.
As these examples suggest, empathy work challenges agents to move
away from scripted responses and one-size-fits-all thinking to adopt an
investigative approach that focuses on customer needs. Empathy work
involves discerning the appropriateness of empathy for each customer in
order to determine whether and how to use it. Customers come to calls with
a variety of underlying needs, such as those concerning face issues, finan-
cial difficulties, and personal health concerns that frequently emerged in the
calls we studied. But we found that customers’ need for empathic commu-
nication varied: Some customers did not want or need empathy; others
benefited greatly from its thoughtful enactment, particularly from the atten-
tive and cognitive empathic responses that demonstrated the agent’s effort
to understand their problems and find solutions.
Our data suggest that attentive, cognitive, and, to a lesser degree, affec-
tive empathic responses can keep calls moving forward. But we also
observed that cognitive empathic responses—particularly those helping
customers find future solutions—do take time. Whether training in empathy
work would help agents to use cognitive empathy to increase call efficiency
remains to be explored as does whether cognitive empathy contributes to
one-call resolution, thereby reducing the need for customer follow-up.
Conclusion
In this study, we sought to identify how empathy is expressed in customer ser-
vice and to explore whether empathic communication is beneficial in aftermar-
ket customer calls. Our findings present both theoretical and practical
implications. From the perspective of genre theory, this study suggests that
empathy work, as a construct, is a complex effort that can help mitigate the ten-
sions underlying the shared purposes that engender customer calls. Practically,
our data show that empathic communication is not a remedy to be universally
applied in stressful calls. But some customers responded positively to its enact-
ment and evensought it. By offeringgenuine emotional support, the enactment
of affective empathy did ameliorate tensions that some customers brought to
calls; other customers indicated a preference for objectivity, even in the face
of personal loss. Cognitive empathy was in greater demand and often highly
148 Journal of Business and Technical Communication 27(2)
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

appreciated although the degreeof its impact on customers remains to be inves-
tigated. In summary, this study shows that by diagnosing customers’ need for
empathy and providing appropriate empathic responses, agents can affect the
tone of a call and the efficiency of call resolution.
We believe these findings are important for training. Customer-service
agents will benefit from understanding the different types of empathy:
attentive, affective, and cognitive. Learning the concept of empathy work
may encourage their increased use of attentive listening strategies to deter-
mine when to employ or withhold expressions of affective empathy and
when to deploy cognitive empathy in order to address customer concerns
and expedite service.
In business and technical communication classes, we have used
examples from this research to stimulate discussion about dyadic commu-
nication and relationship building in general. Around the world, the
aftermarket customer call is a familiar genre for users of credit cards or new
technologies. Some students identify personally with such calls and share
their experiences eagerly. Those who have not used a call center benefit
from learning about the challenges this genre presents and applying what
they learn to other types of dyadic communication.
Future research should validate the multidimensional nature of empathy
work, examine the relationships between the three types of empathic com-
munication, and build and test theories about the differential predictors and
consequences of each. Practically, the demands of empathy work suggest
that it is insufficient to simply train agents to feel for the customer or to
memorize types of affective responses. Agents must rather learn the art
of puzzle solving—analyzing customers’ responses to determine if
empathic communication is needed and, if so, selecting attentive, affective,
and cognitive responses that best meet that need.
Acknowledgments
The Ross School of Business at the University of Michigan and the Nanyang Busi-
ness School at Nanyang Technological University supported this research. Special
thanks also are due to JBTC’s editor, David Russell, managing editor, Lori Peterson,
and two anonymous reviewers whose suggestions were very helpful.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research,
authorship, and/or publication of this article.
Clark et al. 149
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

Funding
The authors received no financial support for the research, authorship, and/or
publication of this article.
References
Adolphs, S., Brown, B., Carter, R., Crawford, P., & Sahota, O. (2004). Applying
corpus linguistics in a health care context. Journal of Applied Linguistics,1, 9–28.
Aggarwal, P., Castleberry, S., Ridnour, R., & Shepherd, C. D. (2005). Salesperson
empathy and listening: Impact on relationship outcomes. Journal of Marketing
Theory and Practice,13, 16–31.
Armistead, C., Kiely, J., Hole, L., & Prescott, J. (2002). An exploration of manage-
rial issues in call centres. Managing Service Quality,12, 246–256.
Axtell, C., Parker, S. K., Holman, D., & Totterdell, P. (2007). Enhancing customer
service: Perspective taking in a call centre. European Journal of Work and
Organizational Psychology,16, 141–168.
Bain, P., Watson, A., Mulvey, G., Taylor, P., & Gall, G. (2002). Taylorism, targets
and the pursuit of quality by call centre management. New Technology, Work and
Employment,17, 170–185.
Bakhtin, M. M. (1986). Speech genres and other late essays.(C.Emerson&M.
Holquist,Eds;V.W.McGee,Trans.).Austin: University of Texas Press.
Betancourt, H. (1990). An attribution-empathy model of helping behavior:
Behavioural intentions and judgements of help-giving. Personality and Social
Psychology Bulletin,16, 573–591.
Bordoloi, S. K. (2004). Agent recruitment planning in knowledge-intensive call
centers. Journal of Service Research,6, 309–323.
Burgers, A., Ruyter, K. D., Keen, C., & Streukens, S. (2000). Customer expectation
dimensions of voice-to-voice service encounters: A scale-development study.
International Journal of Service Industry Management,11, 142–161.
Callaghan, G., & Thompson, P. (2002). We recruit attitude: The selection and shap-
ing of routine call centre labour. Journal of Management Studies,39, 233–254.
Clark, A. J. (2007). Empathy in counseling and psychotherapy: Perspectives and
practices. London, England: Lawrence Erlbaum.
Clark, C. M. (2011). Communication strategies of call center agents: A
multi-method study of solidarity building and conversational control on agent
performance (Unpublished doctoral dissertation). Nanyang Business School,
Nanyang Technological University, Singapore.
Clark, C. M., & Rogers, P. S. (2005). Singaporean and US evaluation of business
writing: Consistency and national perspectives matter. Journal of Asian Busi-
ness,21, 1–18.
150 Journal of Business and Technical Communication 27(2)
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

Clark, C. M., Rogers, P. S., Murfett, U. M., & Ang, S. (2008, April 29). Is courtesy
enough? ‘‘Solidarity’’ in call center interactions. Ross School of Business Paper
(No. 1103). Social Science Research Network Electronic Paper Collection.
Retrieved from http://papers.ssrn.com/sol3/papers.cfm?abstract_id¼1128246
Cochran, J. L., & Cochran, N. H. (2006). The heart of counseling: A guide to
developing therapeutic relationships. Belmont, CA: Thompson Brooks/Cole.
Cowie, C. (2007). The accents of outsourcing: The meanings of ‘‘neutral’’ in the
Indian call center industry. World Englishes,26, 316–330.
Dawson, K. (2005). Ignore dissatisfaction at your own risk. Call Center Magazine,18,6.
D’Cruz, P., & Noronha, E. (2008). Doing emotional labour: The experiences of
Indian call centre agents. Global Business Review,9, 131–147.
Deerie, S., & Kinnie, N. (2002). Call centres and beyond: A thematic evaluation.
Human Resource Management Journal,12, 3–13.
Deery, S., Iverson, R., & Walsh, J. (2004). The effect of customer service encounters
on job satisfaction and emotional exhaustion. In S. Derry & N. Kinnie (Eds.),
Call centres and human resource management: A cross-national perspective
(pp. 201–222). New York, NY: Palgrave.
Dorman, C., & Zapf, D. (2004). Customer-related social stressors and burnout.
Journal of Occupational Health Psychology,9, 61–82.
Dorman, C., & Zijlstra, F. R. H. (2003). Call centres: High on technology—High on
emotions. European Journal of Work and Organizational Psychology,12,
305–392.
Egan, G. (1990). The skilled helper: A systematic approach to effective helping (4th
ed.). Pacific Grove, CA: Brooks Cole.
Egan, G. (2002). The skilled helper: A problem-management and opportunity-
development approach to helping (7th ed.). Pacific Grove, CA: Brooks/Cole.
Friginal, E. (2009). The language of outsourced call centers: A corpus-based study
of cross-cultural interaction. Amsterdam, The Netherlands: John Benjamins
Publishing Company.
Ford, W. S. Z. (1995). Evaluation of the indirect influence of courteous service on
customer discretionary behavior. Human Communication Research,22, 65–89.
Forey, G., & Lockwood, J. (2007). ‘‘I’d love to put someone in jail for this’’: An
initial investigation of English in the business processing outsourcing (BPO)
industry. English for Specific Purposes,26, 308–326.
Halbe, D. (2012). ‘‘Who’s There?’’ Differences in the features of telephone and
face-to-face conferences. Journal of Business Communication,49, 48–73.
Hochschild, A. R. (2003). The managed heart: Commercialization of human feeling.
Berkeley: University of California Press.
Hogan, R. (1975). Empathy: A conceptual psychometric analysis. Counseling
Psychologist,5, 14–18.
Clark et al. 151
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

Kent, T. (1993). Paralogic rhetoric: A theory of communicative interaction.
London, England: Bucknell University Press.
Kohut, H. (1991). On empathy. In P. H. Ornstein (Ed.), The search for the self: Vol.
4. Selected writings of Heinz Kohut: 1978–1981 (pp. 525–535). Madison, CT:
International Universities Press.
Korczynski, M., & Ott, U. (2004). When production and consumption meet:
Cultural contradictions and the enchanting myth of customer sovereignty.
Journal of Management Studies,41, 576–599.
Lewig, K. A., & Dollard, M. F. (2003). Emotional dissonance, emotional exhaustion
and job satisfaction in call centre workers. European Journal of Work and
Organizational Psychology,12, 366–392.
Miller, C. R. (1984). Genre as social action. Quarterly Journal of Speech,70, 151–167.
Miller, K. I., & Koesten, J. (2008). Financial feeling: An investigation of emotion
and communication in the workplace. Journal of Applied Communication
Research,36, 8–32.
Ornstein, P. H. (1979). Remarks on the control position of empathy in psycho-anal-
ysis. Bulletin The Association for Psychoanalytic Medicine,18, 95–108.
Parasuraman, A., Berry, L. L., & Zeithaml, V. A. (1991). Perceived service quality
as a customer-based performance measure: An empirical examination of organi-
zational barriers using an extended service quality model. Human Resource
Management,30, 335–364.
Parasuraman, A., Zeithaml, V., & Berry, L. (1988). SERVQUAl: A multiple-item
scale for measuring customer perceptions of service quality. Journal of Retail-
ing,64, 12–37.
Pontes, M. C. F., & Kelly, C. O. (2000). The identification of inbound call center
agents’ competencies that are related to callers’ repurchase intentions. Journal
of Interactive Marketing,14, 41–49.
Rafaeli, A., Ziklik, L., & Doucet, L. (2008). The impact of call center employees’ customer
orientation behaviors on service quality. Journal of Service Research,10, 239–255.
Rogers, C. R. (1951). Client-centered therapy: Its current practice, implications,
and theory. Boston, MA: Houghton Mifflin.
Rogers, C. R. (1980). A way of being. Boston, MA: Houghton Mifflin.
Strauss, A., & Corbin, J. (1990). Basics of qualitative research: Grounded theory
procedures and techniques. Newbury Park, CA: Sage.
Swales, J. (1990). Genre analysis: English in academic and research settings.
Cambridge, England: Cambridge University Press.
Taylor, P., Mulvey, G., Hyman, J., & Bain, P. (2002). Work organization, control and
the experience of work in call centres. Work,Employment and Society,16, 133–150.
Thompson, P., Callaghan, G., & Van den Broek, D. (2004). Keeping up
appearances: Recruitment, skills and normative control in call centres. In S.
152 Journal of Business and Technical Communication 27(2)
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from

Deery & N. Kinnie (Eds.), Call centres and human resource management: A
cross-national perspective (pp. 129–152). New York, NY: Palgrave Macmillan.
Winsted, K. F. (2000). Service behaviors that lead to satisfied customers. European
Journal of Marketing,34, 399–417.
Wispe, L. (1986). The distinction between sympathy and empathy: To call forth a con-
cept, a word is needed. Journal of Personality and Social Psychology,50, 314–321.
Xu, X., Wang, Y., Forey, G., & Li, L. (2010). Analyzing the genre structure of
Chinese call-center communication. Journal of Business and Technical Commu-
nication,24, 445–475.
Yates, J., & Orlikowski, W. J. (1992). Genres of organizational communication: A
structurational approach to studying communication and media. Academy of
Management Review,17, 299–326.
Author Biographies
Colin Mackinnon Clark is a senior project officer with the University of
New South Wales in Sydney, Australia. As a lecturer in business communication
at Nanyang Business School in Singapore, he completed his PhD thesis on call
centers and won the Association for Business Communication’s Outstanding
Dissertation Award.
Ulrike Marianne Murfett is a senior lecturer in communication management at the
Nanyang Business School, Nanyang Technological University in Singapore. She
teaches communication management modules in undergraduate, MBA, and military
programs.
Priscilla S. Rogers is an associate professor of business communication at the Ross
School of Business, University of Michigan, teaching in the Global MBA and Exec-
utive Programs. She has research awards from the Association for Business Commu-
nication and the National Council of Teachers of English.
Soon Ang is the Goh Tjoei Kok Chair and professor of management at the Nanyang
Business School, Nanyang Technological University. She is the foremost expert in
cultural intelligence and author of two pioneering books on the topic, both published
by Stanford University Press.
Clark et al. 153
at NANYANG TECH UNIV LIBRARY on October 21, 2015jbt.sagepub.comDownloaded from