CONTENTS
JSMAM VOLUME 10 , NUMBER 2 - SPRING 2010
From the Editor
by Dan C. Weilbaker, Ph.D.
7
ACADEMIC ARTICLES
Revelations from Data Mining: A Case Study of a Sales Territory
By Alan J. Dubinsky, Guoying Zhang, and Michele Wood
Developing a Valid and Reliable Measurement of Attitudes Towards
Salespeople
8
27
By Gregory S. Black and Scott Sherwood
APPLICATION ARTICLES
Why Unified Communications Makes Sense Today
40
By Andrew Gorski
What Candor Can Do For You
43
By John Costigan
Mission Statement
The main objective of the journal is to provide a focus for collaboration
between practitioners and academics for the advancement of application,
education, and research in the areas of selling and major account management.
Our audience is comprised of both practitioners in industry and academics
researching in sales.
©2010 By Northern Illinois University. All Rights Reserved. ISSN: 1463-1431
Journal of Selling & Major Account Management
Strategic Partner
Northern Illinois University

Journal of Selling & Major Account Management
Subscription Form
Name
Company
Title
Address
City
State
Zip
Country
E-Mail
Phone
Fax
Subscription Type
 Domestic Individual— $50
 Domestic Corporate— $60
 Foreign Individual – $70
 Foreign Corporate— $80
Payment Method
 Check Enclosed
 Please Bill Me
Card Type:  Visa  Mastercard
 Credit Card
 Discover
 American Express
Name as it appears on card
Card Number
Exp. Date
Signature
Mail This Form to:
Dr. Dan C. Weilbaker
JSMAM
Northern Illinois University
DeKalb, IL 60115
Or Fax this Form to:
JSMAM
Attn: Dr. Dan C. Weilbaker
(815) 753-6014
We appreciate your help!
If you know of colleagues who might benefit and would be interested in
subscribing to The Journal of Selling & Major Account Management,
please forward one of the subscription forms.
Thank-you,
Dan C. Weilbaker, Editor
Place
Stamp
Here
Dr. Dan C. Weilbaker
Journal of Selling & Major Account Management
Department of Marketing
128 Barsema Hall
Northern Illinois University
DeKalb, IL 60115
FOLD HERE
Spring 2010
Manuscripts
1. Articles for consideration should be sent by email to Editor: Dan C. Weilbaker, Department of Marketing Northern
Illinois University, DeKalb, IL 60115 dweilbak@niu.edu.
2. Articles in excess of 6000 words will not normally be accepted. The Editor does welcome shorter articles and case
studies.
3. A manuscript should be submitted via email to the Editor in Microsoft Word format, with author's name(s) and
title of the article. Contributors are advised to check by telephone that submissions have been received. Neither the
editor nor Northern Illinois University, Department of Marketing accepts any responsibility for loss or damage of any
contributions submitted for publication in the Journal.
Biographical note - supply a short biographical note giving the author(s) full name, contact information,
appointment, institutions or organization / company and recent professional attainments.
Synopsis - an abstract not exceeding 100 words should be included.
Diagrams / text boxes / tables - should be submitted without shading although a copy of how the authors
wishes the diagram to appear shaded may be submitted by way of illustrative example. These should be
numbered consecutively and typed on separate pages at the end of the article with an indication in the text
where it should appear.
References - should be cited using the Harvard method. No footnotes should be used for references or
literature citations. Wherever possible, full bibliographic details (e.g., volume number issue number or date,
page numbers publisher year of publication) should be included.
Footnotes - for clarification or elaboration should be used very sparingly. Footnotes should be typed at the
bottom of the page and numbered consecutively throughout the text.
4. Any article or other contribution submitted must be the original unpublished work of the author(s) not submitted
for publication elsewhere.
5. Manuscripts should be formatted on 8 1/2” x 11” paper with all margins of 1" and double-spaced. Font style
should be Times New Roman in 12 pitch.
6. Cross references should not be to page numbers but to the text accompanying a particular footnote.
7. An address for correspondence (including Email address) should be supplied as well as a telephone and fax number
at which the author(s) may be contacted.
8. Authors undertake the responsibility to check that the manuscript should be free of grammatical, syntax or spelling
errors. The Editor reserves the right not to accept any manuscript in which excess alterations or corrections need to be
made.
PERMISSIONS
The copyright owner‘s consent does not extend to copying for general distribution, for promotion, for creating new works, or for
resale. Specific written permission must be obtained from the publisher for such copying.
Subscriptions
To subscribe to Journal of Selling & Major Account Management, please go to www.cob.niu.edu/jsmam/subscription.asp or mail
the subscription form to The Journal of Selling & Major Account Management,. 128 Barsema Hall, Northern Illinois University,
DeKalb, IL 60115. Subscription prices are: U.S. Individual-$50; U.S. Corporation-$60; Foreign Individual-$70; Foreign
Corporation-$80.
EDITORIAL AND ADMINISTRATIVE STAFF
EDITOR—Dan C. Weilbaker, Ph.D.
McKesson Pharmaceutical Group
Professor of Sales
Department of Marketing
Northern Illinois University
dweilbak@niu.edu
EUROPEAN EDITOR—Kevin Wilson
Sales Research Trust
Peyrenegre
47350 Labretonie
France
Kevin@sales-research-trust.org
ASSISTANT—Joey Lata
Administrative Assistant
Professional Sales Program
Department of Marketing
Northern Illinois University
jlata@niu.edu
Vol.. 10 No. 2
Journal of Selling & Major Account Management
EDITORIAL BOARD
Ramon A. Avila
Ball State University
Terri Barr
Miami University—Ohio
Jim W. Blythe
University of Glamorgan
Pascal Brassier
ESC Clermont - Graduate School of Management
Steven Castleberry
University of Minnesota—Duluth
William L. Cron
Texas Christian University
Laura Cuddihy
Dublin Institute of Technology
René Y. Darmon
ESSEC Business School
Dawn R. Deeter-Schmelz
Kansas State University
Sean Dwyer
Louisiana Tech University
Paolo Guenzi
SDA Bocconi
John Hansen
University of Alabama—Birmingham
Jon M. Hawes
Indiana State University
Earl D. Honeycutt
Elon University
Thomas N. Ingram
Colorado State University
Mark C. Johlke
Bradley University
Buddy LaForge
University of Louisville
Terry W. Loe
Kennesaw State University
Richard McFarland
Kansas State University
Northern Illinois University
Daniel H. McQuiston
Butler University
Peter Naude
Manchester Business School
Stephen Newell
Western Michigan University
Nikolaos Panagopoulos, Ph.D.
Athens University of Economics & Business
Robert Peterson
Northern Illinois University
Nigel F. Piercy
University of Warwick
Richard E. Plank
University of South Florida, Lakeland
Ellen Bolman Pullins, PhD
University of Toledo
David Reid
Bowling Green State University
Gregory A. Rich
Bowling Green State University
Rick Ridnour
Northern Illinois University
Elizabeth Rogers
Portsmouth Business School
Charles Schwepker, Jr.
Central Missouri State University
C. David Shepherd
Georgia Southern University
Mary Shoemaker
Weidner University
William A. Weeks
Baylor University
Michael R. Williams
Oklahoma City University
John Wilkinson
University of South Australia
Frederick Hong Kit Yim
Hong Kong Baptist University
Spring 2010
From the Editor
The Journal of Selling & Major Account Management is pleased to provide the
following academic and practitioner articles for your education and
consideration.
In this issue, the lead author in the first academic article is a veteran author with
JSMAM. Along with his co-authors, they look at how a salesperson can
improve their sales performance by data mining techniques to better understand
their customers purchasing tendencies. Armed with this process, salespeople are better able to
target the right customer at the right time.
The second academic article tackles an issue that can be valuable to practitioners. It develops and
validates a scale to measure the customer‘s attitude toward salespeople. It is commonly believed
that the relationship between salespeople and their customers is a primary driver of success. Thus,
being able to accurately and reliably measure the customer‘s attitude can play a major role in
increasing performance of the sales force.
Both of the applications articles in this issue are provided by long-standing members of the
Northern Illinois University Sales Advisory Board.
The first application article is from a new author and focuses on using Unified Communication
tools. These tools assist with improving output to assist in reaching the ever-increasing
performance goals, as well as benefitting the work-life balance.
The final application article is from a repeat author for the journal and it provides a thought
provoking topic: Candor. The author suggests that it is imperative for the salesperson to use
candor when selling to not only build trust, but to also differentiate themselves from other
salespeople.
Our continued thanks also go to the University Sales Center Alliance for their financial support to
help the journal while we build our subscriber base. Our thanks also go to the dedicated members
of the Editorial Review Board and our ad hoc reviewers.
Dan C. Weilbaker, Ph.D.
Editor, The Journal of Selling & Major Account Management,
McKesson Pharmaceutical Group Professor of Sales,
Northern Illinois University
Vol.. 10 No. 2
8
Journal of Selling & Major Account Management
REVELATIONS FROM DATA MINING: A CASE STUDY
OF A SALES TERRITORY
By Alan J. Dubinsky, Guoying Zhang and Michele Wood
The economic climate and extensive product proliferation have created intense competition for most
sales personnel. Salespeople must out-maneuver competitors and strategically target appropriate
products and services to fit their customers‘ needs. In such a difficult business milieu, data mining for
business intelligence can provide salespersons a treasure trove of information with which to attend to
their customers‘ unique requirements.
This paper presents a data mining process that salespeople can utilize to analyze customer records to
help them better understand customers‘ purchasing tendencies and associations. The case study can be
utilized to tailor a salesperson‘s efforts by targeting specific customers based on their characteristics.
Customer sales data from a territory of a large industrial distributor were employed to apply three
different data mining techniques. Data mining results yielded valuable information that could be used to
increase territory sales, gain customer loyalty, and grow market share.
INTRODUCTION
Customers are the lifeblood of any firm
(Dubinsky 1999): they nourish the company
financially. Indeed, Kale (2004) posits that
developing and maintaining customer
relationships is vital for competitive advantage.
Failure to pay heed to customers and their
various needs can leave a marketer unable to
leverage the company‘s competencies. In an
effort to provide customers with the ―perfect
customer experience,‖ Customer Relationship
Management (CRM) has become requisite for
many organizations as a means of enhancing
performance (Javalgi, Martin, and Young 2006;
Payne and Frow 2005, 2006).
Successful CRM implementation requires
attention to four crucial areas: strategy, people,
processes, and technology (Crosby and Johnson
2001; Yim 2002).
Operationalizing this
perspective, Yim, Anderson, and Swaminathan
(2004) propose that implementation entails
focusing on key customers, organizing around
Northern Illinois University
CRM, managing knowledge, and incorporating
CRM-based technology. CRM is an ―enabler,‖
as it assists companies to manage their customers
effectively (Landry, Arnold, and Arndt 2005;
Moutot and Bascoul 2008; Raman, Wittman, and
Rauseo 2006).
Boulding et al. (2005) state that the core of CRM
is ―dual creation of value.‖ Through CRM
activities, value is created for both the marketer
and the customer. That is, rather than solely
trying to maximize value for the firm,
management should also be concerned with
creating value for the customer. Value creation
entails discerning what value the organization
might offer customers, ascertaining what value
customers provide the firm, and maximizing the
lifetime value of the customer (Payne and Frow
2005).
Customer lifetime value (CLV) is the predicted
profitability of a customer during the entire
relationship with the company (Kale 2004). CLV
changes over the life of the relationship and
Academic Article
requires integration of the organization with the
entire CRM process (Boulding et al. 2005). CRM
efforts necessitate an ongoing balance between
the customer and the firm as the relationship
evolves through its stages (Reinartz, Krafft, and
Hoyer 2004). Moreover, distribution of value
across customers is uneven (Chen et al. 2006;
Reinart, Krafft, and Hoyer 2004). Therefore,
marketers must balance their customer portfolios
vis-à-vis their customers‘ needs, company efforts
to foster customer value, and the means by
which customers provide value for the firm
(Boulding et al. 2005). Part of this entails
correctly identifying responsive and profitable
target customers (e.g., Cao and Gruca 2005).
A key goal of CRM is to efficiently and
effectively increase the acquisition and retention
of customers by selectively building and
maintaining mutually satisfying relationships with
them. As Dickson et al. (2009, p. 113) aver:
―When CRM is done right, it seamlessly
integrates the marketing-sales-delivery-customerservice core functional added value
processes….‖ Through advances in information
technology, companies can realize improved
customer relationships. CRM technology can
foster gathering, analyzing, and interpreting
various kinds of customer data in order to
enhance the relationship with the customer. By
converting customer information into usable
data, CRM can increase the overall performance
of a company (Yim, Anderson, and Swaminathan
2004; Stein and Smith 2009).
Use of information technology to manage
customer relationships is a critical dimension of
relationship marketing (Anderson, Dubinsky,
and Mehta 2007). In fact, it has been found to
enhance salesperson performance (Hunter and
Perreault 2007; Mathieu, Ahearne, and Taylor
2007; Stoddard, Clopton, and Avila 2006) and
Spring 2010 9
other CRM outcomes (Moutot and Bascoul
2008).
Although diffusing technological
innovation into a sales force may face resistance
(Sharma and Sheth 2010), providing technology
empowerment assists in salespeople‘s
technological usage (Chou, Pullins, and Senecal
2007).
A particularly beneficial technological tool for
transforming customer information into usable
data—and thereby assisting sales personnel to
build sound relationships with customers—is
data mining. Data mining refers to ―all aspects
of an automated or semi-automated process for
extracting previously unknown or potentially
useful knowledge and patterns from large
databases‖ (Olafsson, Li, and Wu 2008, p. 1429).
Its main objective is to reveal interesting
patterns, associations, rules, changes,
irregularities, and general regularities from the
data to improve decision making (Sumathi and
Sivanandam 2006, p. 321).
Data mining can help sellers (including sales
personnel) enhance their understanding of
customers (Shaw et al. 2001, p. 128), as it entails
―…searching and analyzing data in order to find
implicit, but potentially useful, information…to
uncover previously unknown patterns, and
ultimately comprehensible information from
large databases.‖ Furthermore, data mining
provides specific knowledge to organizations
(e.g., salespeople) about each customer‘s needs
(Shaw et al. 2001). Moreover, it allows users to
identify relationships between independent and
dependent variables (Zhang et al. 2008). In
addition, data mining approaches can reduce the
effect of human bias on data analysis (Tu 2008).
Data mining can reveal a treasure trove of
customer information and assist in a firm‘s
knowledge management process (Shaw et al.
2001). This applies whether data mining is
Vol. 10, No. 2
10
Journal of Selling & Major Account Management
conducted at an aggregate company level or on
an individual sales territory. Consequently, it can
serve as an especially effective tool to help a
salesperson attend to the unique needs of each
of his or her customers. Moreover, it affords
enhanced capability to deal with each customer.
As Shaw et al. (2001, p. 128) assert:
There is an increasing realization that
effective customer relationship
management can be done only based
on a true understanding of the nees and
preferences of customers…[d]ata
mining tools can help uncover the
hidden knowledge and understand
customers better.
Given the important role data mining can play in
enhancing relationship marketing efforts,
salespeople should be knowledgeable about how
to perform data mining. They must pssess the
technological tools and know-how to conduct
data mining in a way that will allow them to
leverage such efforts to advantage. Indeed,
Anderson, Dubinsky, and Mehta (2007, p. 121)
state that data mining is an especially useful tool
for sales personnel, as it can ―identify important
relationships among buyers and the products
they purchase, along with associated
complementary products‖ and help salespeople
assist their customers to sell effectively through
their own accounts.
The current paper presents a non-compex,
useful approach to data mining that has bactually
been utilized by a salesperson in an industrial
setting.
Discussed are three data mining
techniques, each with a unique focus that the
host salesperson conducted with data from her
accounts. The case study that follow is an
attempt to show salespeople the value that their
own use of data mining can offer. As such, this
paper partially answers a call by Raman,
Wittmann, and Rausea (2006, p. 51) for ―...future
research [that] could focus on fine-grained
aspects of CRM implementation and use.‖
DATA MINING ANALYSIS
In general, the data mining process includes the
Northern Illinois University
following stages:
data preparation and
exploration, data training and validation, model
evaluation and selection, and application of the
model on new data (Berson et al. 1999). In the
stage of data training and validation, a number of
alternative methods exist.
They can be
categorized into the following four groups:
clustering, classification, regression, and
association rule learning (Fayyad, PiatetskyShapiro, and Smyth 1996). Shown in Table 1 are
these four data mining techniques with their
corresponding purposes.
The objective of this paper is to illustrate a data
mining process that salespersons could use to
develop a grow-sales strategy. This is done
through examining relationships among various
customer variables and sales information that are
conceivably already in a firm‘s data base. Using
Table 1 as a guide, three data mining methods
were employed in the present paper: multiple
linear regression, association rules, and
clustering. Regression analysis was used to
develop models to predict annual sales revenue
based on customer characteristics. Association
rules were generated for the purpose of finding
products that customers often purchase together;
such information would be helpful for
developing cross-selling strategies. Clustering
was conducted to profile customers and develop
corresponding marketing strategies for different
clusters (i.e., related groups of customers)1.
The remainder of this paper will illustrate how an
industrial salesperson undertook the three
foregoing data mining techniques. Specific
details of how the host salesperson conducted
data mining will be described, thus providing
user-friendly directions for sales personnel
interested in utilizing data mining within their
own territories.
Spring 2010 11
Academic Article
Technique
Clustering
Classification
Regression
Association Rule
Table 1. Common Data Mining Methods
Purpose
Discovering groups and structures in the data that are in some way ―similar.‖
Making classifications of new data with known classes. Common algorithms
include decision tree, nearest neighbor, naive Bayesian, neural networks, and
support vector machines.
Estimating parameters in an assumed function form by fitting the data with
the least error. The established function can be used to predict new data.
Searching for association relationships between variables. A common
example is referred to as ―market basket analysis‖ to determine products or
services that will be purchased together.
Data Collection
Initially, salespeople must identify and collect the
data that they feel is pertinent to their specific
data mining situation. Sales personnel interested
in data mining ideally should be able to obtain
from company records the data utilized in this
paper. Indeed, the host salesperson in this study
requested the data from her management in
order to undertake data mining. She focused
solely on her own accounts and described the
importance of data mining to management to
justify her being given the data. When proposing
to management the advantages of using data
mining, salespeople can make the following
arguments:
Data mining can reveal an abundance of
customer information.
Data mining helps salespeople improve their
understanding of customers.
It requires users to search and analyze data
to find useful information that can uncover
previously unknown patterns of relationships
among customer variables (e.g., sales volume
and such factors as customer size, channel of
distribution, use of e-commerce).
Data mining can offer detailed information
about each customer‘s needs.
It allows firms to make enhanced use of the
information they have about their customers.
Data mining can serve as an especially
effective tool to help salespeople attend to
the unique needs of their customers.
As such, the salesperson must first identify the
relevant group of accounts on which he or she
wishes to undertake data mining.
The
salesperson in our case study gathered data for a
12-month period from an average-sized sales
territory of Company DM (the real name of the
company is not disclosed for confidentiality).
DM is a Fortune 500 company and one of North
America‘s largest maintenance, repair, and
operations (MRO) product distributors.
It
competes with other broad-line distributors,
regional suppliers, and local suppliers. The host
salesperson‘s entire sales territory consists of
twenty customers in the commercial
manufacturing and health care segments2. The
average annual sales volume for the host
salesperson‘s customer data set was $103,487.56.
The standard deviation was $23,497.83. The
smallest customer generated annual sales of
$4,477.86 (a new health care customer), and the
largest customer in the data set purchased
products totaling $330,513.00 over the 12-month
period.
Data Preparation
After the salesperson has identified the relevant
Vol. 10, No. 2
12
Journal of Selling & Major Account Management
Table 2: Product Category Spending by Selected Customers
Customer LAMPS BALLASTS FIXTURES
1
$86
2
$596
3
4
$17,156
PORTABLE FLASHELECTRICAL
LIGHTING LIGHTS/ SUPPLIES
BATTERY
$82
$6
$29
$344
$61
$12,281
5
$364
$2,700
$706
$1,298
$489
$1,076
$44,541
$25
$1,904
$3,416
$40,797
$141
6
$5,190
7
$5,252
$1,205
8
$2,181
$276
$13,611
accounts for data mining, he or she should
classify the data that will be used in the analyses.
The classification could be at two levels—
product grouping and customer profiles.
Product Groupings In a multi-product firm,
the salesperson would need to determine how
the product mix can be meaningfully classified
(e.g., product line, complementary items,
individual items). With respect to product
grouping in our case study, the host salesperson
examined her product sales for each customer,
ultimately categorizing the sales into the 102
distinct DM product categories. Shown in Table
2 is the annual spending on selected product
categories for a portion of the host salesperson‘s
twenty customers during the 12-month period.3
Customer Profiles In addition to grouping
products, the salesperson needs to identify
germane characteristics that can be used to
classify customers vis-à-vis the nature of their
interactions with the firm. Doing so will
produce a profile for each customer.
Prospective characteristics might include such
features as customer size, customer
Northern Illinois University
$3,298
$136
$2,275
demographics, purchase patterns, account call
frequency, and so on. For each characteristic,
the salesperson will assign a numerical value (e.g.,
0, 1, 2) to represent the corresponding category
of the characteristic.
With respect to customer profiles in our case
study, the host salesperson identified certain
customer dimensions of her twenty customers
that could meaningfully portray her accounts.
Based on her knowledge of the territory and
input from DM management, the following
profile characteristics of each customer, as
portrayed in Table 3, were selected:
1. Customer Segment Type: 0=Manufacturing,
1=Health care.
2. Corporate Account: 0=No, 1=Yes, (i.e.,
whether the customer has corporate account
status with DM‘s corporate office).
3. VMI: 0=No, 1=Yes. VMI refers to
―vendor managed inventory.‖ This means
that a DM employee goes to the customer
location weekly with a bar code scanner to
order products, place replenishment product
Spring 2010 13
Academic Article
onto customer shelves, and assist the
customer with product selection for spot
(unplanned/impulse) purchases.
4. PPD (Pre-paid) Freight: 0=No, 1=Yes. DM
pays the freight for customers whose
purchase level reaches a particular threshold;
some of DM‘s corporate contracts receive
this incentive.
5. E-commerce Customer:
0=No, 1=Yes.
This variable indicates whether the customer
utilizes some kind of electronic means (e.g., e
-mail, EDI) to conduct transactions with
DM.
6. Distance from DM Store: Total distance in
miles from the closest DM store.
7. Primary Order Method: Three different
primary order methods were used, coded as
follows: 00 = Phone, 01 = Fax, 10 = Ecommerce.
8. MRO Potential: Total annual dollars that
the customer spends with all suppliers for
maintenance, repair, and operating products.
9. Number of Contacts: Total number of
individuals within the customer‘s firm who
order from DM.
10. E-commerce Share: The percentage of the
customer‘s sales that are procured through
electronic channels (i.e., EDI, DM.com,
email, or some other type of electronic
platform).
11. Primary Discount: The minimum discount
that the customer receives for any item
purchased from the DM catalog.
12. Count of Categories Purchased: The total
number of distinct product categories from
which the customer purchased items.
DATA ANALYSES
At this juncture in the process, the salesperson
now selects alternative data analytic techniques
that he or she could use for data mining. The
techniques ultimately used should be based on
the purpose for undertaking the data mining. As
noted earlier, shown in Table 1 are the four
general data mining techniques and their
corresponding purposes. In our cases study,
multiple linear regression, cluster analysis, and
association rules were the techniques selected.
The host salesperson opted for these three
techniques because her purposes for data mining
were to (a) employ a set of customer
characteristics to predict sales revenue per
customer (thus, multiple linear regression), (b)
segment the territory into customer groups (thus,
cluster analysis), and (c) identify cross-selling
opportunities (thus, association rules). Each of
these techniques will now be described vis-à-vis
salesperson data mining.
Multiple Linear Regression.
If a salesperson wishes to use multiple linear
regression in data mining, most likely he or she
will utilize it to forecast sales revenue per
customer.
Multiple linear regression is a
prediction-type analysis where historical data are
used to build a model that helps forecast a future
event (i.e., sales in this study). Accordingly, the
salesperson must identify a set of germane
customer characteristics that are likely to be
related to sales revenue. The customer profile
that the salesperson has already created (as noted
above) should conceivably consist of those
relevant customer characteristics.
Once the requisite data have been obtained for
use in multiple linear regression, the salesperson
Vol. 10, No. 2
Northern Illinois University
$108,091.26
$16,464.12
$330,513.00
$37,967.80
$43749.00
$220,556.48
$34,781.97
$19,962.18
$289,831.27
$46,821.18
$70,000.00
$51,417.57
$65,934.76
$4,477.86
$213,149.00
$13.678.01
$192,790.40
$258,902.14
$25,663.24
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
0
0
0
0
0
1
0
0
0
1
0
0
1
0
0
0
0
1
0
0
Customer
Type
0
0
1
0
1
1
0
0
0
1
1
0
1
0
0
0
1
1
0
0
Corporate
Account
0
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
VM
I
0
1
1
0
1
0
0
0
0
1
1
0
1
1
0
0
1
0
1
0
PPD
Freight
1
1
1
1
0
1
0
0
0
1
1
0
1
1
0
0
0
0
1
1
ECommerce
5
5
1
50
80
75
5
20
12
7
45
75
75
15
12
100
65
150
50
20
Distance
0
1
0
1
0
0
0
0
0
1
1
0
1
1
0
0
0
0
1
1
(e-commerce)
Primary Order
Method
1
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
0
0
Primary
Order
Method
700000
1100000
700000
600000
1400000
40000
100000
700000
400000
80000
300000
2000000
300000
200000
1300000
900000
800000
200000
2000000
2000000
MRO
Potential
15
31
43
17
9
8
8
12
13
5
32
17
8
82
30
11
31
14
42
27
Number
of Contacts
30.09
47.6
0
43.21
0.89
0
1.11
0
0
94.35
10.23
1.95
89.7
35.86
17.12
0
0
1.15
57.62
45.57
Ecommerce
Share
5
5
15
10
5
13
10
10
10
10
10
10
7.5
10
10
10
5
15
10
10
Primary
Discount
Note: Phone orders were coded as ―00‖. Therefore, if both columns for ―Primary Order Method‖ (e-commerce and fax columns) contain a ―0‖ for a particular customer, this
means that the customer was using the phone as the primary order method.
$25,000.00
1
Customer
12-month
Sales
Table 3. Customer Characteristic Data
14
Journal of Selling & Major Account Management
Academic Article
must select computer software to conduct the
analysis. An array of software options is
available on the market. Prototypical options
include SAS (Statistical Analysis Software), SPSS
(Statistical Programming for Social Science),
STATA, and XLMiner for Excel. Each software
package provides essentially similar routines for
multiple linear regression.
A primary research question of our host
salesperson was, ―What factors (those noted in
Table 3) determine the sales volume of a
customer?‖ For example, does giving the
customer pre-paid freight increase sales? Does
deploying labor in the form of vendor managed
inventory augment sales? If the customer‘s
corporate office enforces the contract, might
sales rise? The host salesperson‘s goal was to
understand what DM programs and services, if
any, may be used to forecast a customer‘s annual
sales. Therefore, the data set presented in Table
3 was utilized for regression analysis. The
resulting model could then be applied to develop
and implement a sales plan directed at enhancing
sales with other customers in the territory.
Results from the regression analysis are
presented in Table 4. The host salesperson
utilized Excel to conduct the analysis owing to
its widespread availability in Microsoft Office.
The multiple R-square of 0.99 indicates that the
regression model explains 99 percent of the
variation in total sales dollars of the customers in
the study. This means that the independent
variables (customer profile characteristics—
Table 3) chosen for this regression analysis are
helpful in explaining the variance in the
dependent variable—annual sales.
The
significant independent variables—as evidenced
by the statistically low p-value (less than 0.05)—
are corporate account, pre-paid freight, primary
order method/e-commerce, primary order
Spring 2010 15
method/fax, and E-commerce share.4
The regression model can be interpreted as
follows: First, if a customer is a corporate
account, he or she might generate $65,681.04 less
sales revenue (owing to the negative sign of the
value: -65681.04) than a customer who is not a
corporate account. This implies that DM‘s sales
program which offers local target customers
deeper discounts than corporate customers
seemingly yields unexpected results.
The
program apparently enhances sales from local
customers but not from corporate accounts.
Second, if a customer is accorded prepaid
freight, annual sales from that customer could
well be $182,378.35 higher than a customer who
does not receive prepaid freight. Therefore,
prepaid freight may be an important variable for
augmenting account sales. Third, a customer
who utilizes an e-commerce channel for ordering
(such as DM.com, EDI, or some other type of
electronic interface) tends to have sales revenue
that is $121,670.15 greater than a customer who
orders primarily via the phone. Similarly, a
customer who faxes orders to DM seems to have
$51,957.55 more sales revenue than a customer
who primarily phones in the orders. Finally,
every unit increase in a customer‘s E-commerce
share could well lead to a sales decrease of
$3,020.12 (owing to the negative sign of the
value: -3,020.12). Given the popularity of using
the internet on which to make orders, this result
is not intuitive. Discussions with the host
salesperson, as well as in-depth analysis of each
account‘s sales data, however, revealed a
potential rationale for the unexpected finding.
By and large, the territory‘s accounts are not
using e-commerce extensively with which to
place an order. However, there exist customers
who generate a relatively small level of sales
while having a disproportionately high
percentage of e-commerce. Accordingly, the
Vol. 10, No. 2
16
Journal of Selling & Major Account Management
Table 4. Multiple Linear Regression
Regression Statistics
Multiple R
0.995
R Square
0.990
Adjusted R Square
0.967
F-statistic
Observations
43.717***
20
*** p<0.001
Coefficients
Standard Error
Intercept
15964.84
40009.62
Customer Segment Type
-2901.01
70321.93
-65681.04*
26186.06
14911.31
25466.23
182378.35**
45172.09
E-Commerce Customer
16325.02
22359.10
Distance from DM Store
70.58
Corporate Account
VMI
PPD Freight
308.19
Primary Order Method (e-commerce)
121670.15*
35535.82
Primary Order Method (fax)
51957.55*
17764.02
MRO POTENTIAL
-0.02
0.02
Number of Contacts
-72.45
390.30
E-commerce Share
-3020.12*
864.93
Primary Discount
4849.27
5059.43
Count of Product Categories Purchased
216.60
838.95
**p <0.01, *p <0.05
host salesperson confirmed that she needs to
expend more effort to encourage e-commerce in
her territory.
Cluster Analysis
A prominent use for data mining is to create
underlying customer segments within a
salesperson‘s territory. More specifically, it
entails reducing the number of customers in a
territory into a manageable set of clusters so that
corresponding sales plans can be directed at each
Northern Illinois University
cluster. This reduction occurs by using the
customer profiles (identified above).
If a
salesperson‘s customers are segmented based on
customer profile characteristics, then selling
efforts could be tailored to each segment.
A variety of clustering techniques, as well as
software providing cluster analyses, is available. 5
Regardless of the technique or software used,
however, cluster analysis is a data reduction
technique seeking similarity within a customer
segment but dissimilarity across customer
Spring 2010 17
Academic Article
segments. In other words, customers within a
segment should have similar profile
characteristics but have dissimilar profile
characteristics from customers in other
segments.
clustering further grouped these four accounts
together.
Another example is the cluster containing
customers 4, 18, and 16. All three of these
customers were manufacturers who generated
relatively high sales volume, participated in the
VMI program, received pre-paid freight, rarely
used e-commerce for order-placement, had large
sales potential, and purchased many different
categories of DM products. A higher level of
clustering further grouped these three accounts
together with customers 2, 7, 10, and 19, all of
whom possessed similarities in certain
characteristics (noted in Table 3).
With respect to our case study, data for cluster
analysis were from the same data set used for
multiple linear regression. Shown in Figure 1 are
the results from the host salesperson‘s cluster
analysis. The clustering appeared to be accurate
vis-à-vis identifying distinguishing characteristics
that grouped the host salesperson‘s territory
customers together. For example, customers 1
and 17 were manufacturers that had the
following similarities (noted in Table 3): noncorporate account, non-VMI program account,
no pre-paid freight, primary order method of ecommerce, e-commerce share of sales, and level
of primary discount. Also, customers 6 and 20
were manufacturers having these overlapping
characteristics: non-corporate account, nonVMI program account, no pre-paid freight, and
primary order method/fax. A higher level of
One further example entails customers 3, 8, 15,
and 11, who ultimately form a cluster at a higher
level. What did accounts 3 and 15 have in
common?
They both were health care
institutions generating a relatively small sales
volume, were corporate accounts not
participating in the VMI program, were not
accorded pre-paid freight, were a relatively far
distance from a DM store, rarely utilized e-
Figure 1. Dendrogram (Ward’s Method) of Cutomer Segmentation
Dendrogram (Ward's m ethod)
80
20000
70
60
Distance
50
40
30
20
10
0
0
1
17
6
20
5
9
12 13 14
3
15
8
11
2
7
10 19
4
18 16
Vol. 10, No. 2
18
Journal of Selling & Major Account Management
commerce to place orders, had a small number
of buyers of DM products within their
organizations, and received the same primary
discount. Commonalities between customers 8
and 11 were as follows: health care facility,
corporate account with no participation in the
VMI program, pre-paid freight recipient, ecommerce ordering, small number of buyers of
DM products within the institution, and identical
primary discount. Results from the cluster
analysis (Figure 1) also accurately depicted
similarities across other clusters.
Association Rules
Another frequently used data mining technique is
association rules (Shaw et al. 2001). This method
is also known as affinity grouping or market
basket analysis. When a salesperson uses affinity
grouping, it is to determine which products are
purchased together. A common example for this
method is to determine which retail items go
together in a shopping cart at the supermarket.
An association rule has a generic format of the
following: If product A is purchased, then
product B should also be purchased. Product A
is often referred as the ―antecedent product,‖
and product B is the ―consequent product.‖
In order to derive a legitimate association rule,
three critical parameters will be generated from
the association rules software—namely, support,
confidence, and lift. Support is the number of
transactions for a particular customer that
included both the antecedent and consequent
products in the rule. Confidence is defined as
the probability of purchasing a consequent
product given that the antecedent product was
purchased. Lift is defined as the ratio of
confidence to the probability of purchasing the
consequent product. According to Shmueli et al.
(2007), a lift ratio greater than 1.0 suggests that
an association is significant between antecedent
Northern Illinois University
and consequent products. In other words, the
chance of purchasing the consequent product
given that the antecedent product was purchased
is higher than the general chance of purchasing
the consequent product. The larger the lift ratio,
the greater is the strength of the association (of
the antecedent and consequent products).6
With respect to our case study, the host
salesperson‘s third purpose of data mining was
to identify the potential for the host salesperson
to engage in cross-selling—selling an additional
product or service to an existing customer.
Association rules generated from various DM
product categories could reveal which products
tended to be purchased together. Armed with
this knowledge, the host salesperson may be able
to enhance sales in her territory via cross-selling
(Kumar 2002).
Because not all 102 product categories were
purchased by the customers in the host
salesperson‘s territory during the 12-month
period, data for the association rule analysis falls
into a subset of the total 102 DM product
categories. The objective of the analysis was to
find common associations between product
categories that are both purchased together by a
significant number of customers.
Prior to conducting association rules, data
transformation had to be conducted for use in
the association rules analyses. DM is a broad
line distributor with a multiplicity of products.
Based on customer sales records, a binary value
(e.g., 0 or 1) was assigned for each customer in
each product. For brevity, illustrated in Table 5
is a subset of the transformed data. When a cell
value equals ―1,‖ it indicates that the customer in
that row purchased the product in that column
in the past 12 months. If the cell value equals
―0,‖ it means that no such transactions could be
found in the data set for that customer for the
Spring 2010 19
Academic Article
particular product.7
The host salesperson‘s transformed data were
analyzed using association rules. In order to
generate a reasonable size of legitimate rules, the
minimum support and lowest confidence level
had to be determined. After testing alternates, a
minimum support of 4 and the lowest
confidence level of 50% were chosen because
these rules were generated with adequate lift
ratios. Shown in Table 6 are the association
rules generated when using a minimum support
of 4—which means there needs to be at least 4
customers having the purchasing combination of
the particular antecedent and consequent
products. The analysis also set the lowest
confidence level at 50%, which means that the
probability has to be at least greater than 0.5.
The findings revealed interesting product
combinations that the host salesperson did not
necessarily anticipate. Such revelations indicate
the value association rules can offer data mining
users.
According to the rules generated (Table 6),
customers who bought lamps from DM also
typically purchased safety products.
(The
confidence level of this rule was 80%, and the lift
ratio was 1.33.) Rule 2 shows that customers
who purchased tools and storage products also
bought safety products (with a 75% confidence
level and 1.25 lift ratio). Rule 3 reinforced Rule 2
by also showing that customers who bought DM
safety products also purchased tools and storage
products (with a 50% confidence level and 1.25
lift ratio). Rule 4 demonstrated that customers
who purchased electrical supplies also bought
safety supplies (with a 62.5% confidence level
and 1.04 lift ratio). The results from the
foregoing association rules offer insight for the
host salesperson in terms of identifying
prospective cross-selling opportunities.
SALESPERSON IMPLICATIONS
CASE STUDY FINDINGS
OF
The primary objective of this paper was to
introduce three relatively useful data mining
tools that can assist sales personnel to improve
their efficiency and effectiveness in their sales
territory. Through data mining of one industrial
distributor salesperson‘s sales for a 12-month
period, valuable insight into the host
salesperson‘s target customer groups and target
Table 5. Transformed Purchasing Data for Selected Customers
Customer LAMPS BALLASTS FIXTURES
PORTABLE FLASHLIGHTS/ ELECTRICAL DISTRIBUTION/
LIGHTING BATTERY
SUPPLIES
CONTROL
1
1
0
0
0
0
0
0
2
0
0
0
0
0
0
0
3
0
0
0
0
0
1
0
4
1
0
0
0
0
1
0
5
0
0
0
0
0
0
0
6
0
0
1
0
0
0
1
7
0
0
1
0
0
1
1
1
0
0
0
0
1
0
8
Note: When a cell value equals ―1‖, it indicates that the customer in that row purchased the product in that column in
the past 12 months. If the cell value equals ―0‖, it means that no such transactions could be found in the data set for
that customer for the particular product.
Vol. 10, No. 2
20
Journal of Selling & Major Account Management
Table 6. Association Rules with Minimum Support of 4 and Minimum Consequence of 50%
Rule #
1
2
3
4
Conf. % Antecedent (a)
80
75
50
62.5
A1-LAMPS=>
M1-TOOLS & STORAGE=>
J1-SAFETY=>
B1-ELECTRICAL SUPPLIES=>
Consequent (c)
J1-SAFETY
J1-SAFETY
M1-TOOLS & STORAGE
J1-SAFETY
products was obtained. Specifically, the findings
of the three data mining techniques revealed
information that affords this salesperson
opportunity to design a sales plan tailored to her
territory‘s customers. Purely as an illustration,
actions that the host salesperson could take vis-à
-vis the three data mining approaches are now
offered.
Sales Plan from Results of Multiple Linear
Regression.
Findings from the regression analysis revealed
that the host salesperson should focus on four
key variables in her efforts to grow sales in the
territory. Specific actions that she might take are
noted below.
1. Customers who are DM corporate accounts
should be given enhanced discounts vis-à-vis
local targeted accounts.
The current
discount program for which corporate
accounts are eligible is having a negative
impact on sales. Indeed, their discount
program generates lower sales volume than
discounts offered to local targeted
customers, thus penalizing them for being
corporate accounts. Therefore, the host
salesperson should work with DM to
develop an effective discount structure
specifically for corporate accounts that will
reverse the current adverse influence the
present program has on their level of sales
2. Customers receiving pre-paid freight tend to
Northern Illinois University
Support(a) Support(c)
5
8
12
8
12
12
8
12
Support(a U
Lift Ratio
c)
4 1.333333
6
1.25
6
1.25
5 1.041667
have much larger sales levels than do those
who do not receive this concession. To
receive pre-paid freight requires an account
to have a sufficient level of sales to qualify
for free shipping. Therefore, the host
salesperson should work closely with those
accounts not receiving free shipping (noted
in Table 2) in efforts to enhance their sales
to the qualifying level for pre-paid freight.
Using the results of the market basket
analysis (selling associated products) could
also be utilized to help increase such
accounts‘ sales.
3. Customers e-mailing or faxing their orders to
DM have higher sales levels than accounts
not doing so.
Therefore, the host
salesperson should work with those accounts
not using either of the foregoing expedited
ordering methods. Such customers need to
be encouraged about the speed, convenience,
and safety of submitting their orders
electronically. In fact, DM might consider
offering customers an incentive for sending
their orders via e-mail or fax.
Sales Plan from Results of Cluster Analysis
Findings from the cluster analysis indicate that
the sample territory can be divided into distinct
clusters that should receive special selling
attention. For example, the higher level cluster
layer revealed an especially relevant segmentation
outcome (Figure 1): Segment #1 with customers
1, 17, 6, 20, 5, 9, 12, 13, 14; Segment #2 with
Academic Article
customers 3, 15, 8, 11; and Segment #3 with
customers 2, 7, 10, 19, 4, 18, and 16.
Segment #1 was comprised of manufacturing
customers who spent a small amount of money
on DM products relative to their MRO potential,
were not corporate accounts, did not participate
in the VMI program, and were not recipients of
pre-paid freight. Clearly, these accounts merit
special attention owing to their potential sales
volume and lack of participation in DM
programs. The host salesperson should make
concerted efforts to enhance the sales from these
accounts so that they can become eligible for
DM customer programs and improved primary
discounts.
Segment #2 consisted of health care institutions
that were accorded corporate account status,
whose 12-month sales essentially were low
relative to their MRO potential, were not eligible
for the VMI program, and had comparably few
distinctly different buyers of DM products
within their organization. Customers in this
segment should be informed of the value of the
VMI program, thus enticing them to increase
their sales volume to qualify for it. Furthermore,
the host salesperson should seek to cultivate
additional buyers within each of these health care
facilities; doing so could help augment sales from
these customers.
Segment #3 constituted manufacturing accounts
that generated a large sales volume for DM, may
or may not be a corporate account, received prepaid freight, may or may not have participated in
the VMI program, and qualified for pre-paid
freight. The host salesperson should be especially
vigilant to protect the accounts in this segment
from being enticed by the selling efforts of
competitors. A loss of any of these customers
would result in a substantial decline in sales
revenue from the segment. Owing to the size of
these accounts, special attention should be given
Spring 2010 21
to these clients by providing extra customer
service (e.g., ensuring accurate and timely
deliveries, offering ways to optimize the use of
DM products). Additionally, informing these
customers about how they could receive higher
discounts by augmenting their purchases from
DM should be emphasized.
Sales Plan from Results of Association Rules
Transaction information utilized for the study
included every product category sold to the
salesperson‘s twenty customers during a 12month period. The association rules revealed
particular propensities for customers who buy
items in one product category to also buy items
in another category. The rules with the highest
number of support occurrences and lift ratios
greater than 1 are illustrated in Table 6 and offer
specific actions the host salesperson could take
with her customers.
Findings indicate that customers purchasing
tools and storage from DM also tend to buy
safety products (Table 6). Indeed, DM is one of
the largest distributors in North America of both
of these product categories. Although tools are
not closely related to safety products, safety
products are DM‘s largest product category.
Accordingly, a sales plan directed at selling these
two product categories to any customer who is
not purchasing from both of these categories
seems warranted. The sales effort could entail
use of targeted e-mail promotions, product
demonstrations at the customer site, discount
promotions, and cost savings demonstrations for
high quality, economical, private-labeled, globally
-sourced brands.
Shown in Table 6 is another set of association
rule results. Findings from this analysis suggest
that two other product categories which show
associations with safety products are lamps and
electrical supplies. This result suggests that a
Vol. 10, No. 2
22
Journal of Selling & Major Account Management
similar sales plan as that above should be
designed to present cross-selling opportunities to
buyers of safety products and lamps and/or
between safety products and electrical supplies.
Utilization of association rules analysis offers the
host salesperson insight into the particular
products that could be emphasized to grow sales.
Concomitantly, generating increased sales from
customers not only enhances the performance of
the salesperson, but it also assists customers to
generate a sales volume that affords qualification
for prepaid freight and a deeper primary discount
with DM (i.e., special concessions DM offers to
qualifying customers).
Association rule results also imply that the host
salesperson demonstrate to DM retail stores
where to stock items. More specifically, the DM
salesperson should convince DM stores in her
territory to arrange items in close proximity to
each other that are often purchased together
(e.g., position safety products near the tools,
lamps, and electrical supplies). Furthermore, an
online selling effort could be directed at DM
retail store customers by sending targeted e-mails
from DM‘s web site. For example, the e-mail
could state (à la Amazon.com) ―customers who
purchased this item also purchased the items
noted below.‖
MANAGERIAL CONCLUSIONS
Integrating data mining into a sales plan creates
direction and focus with increased sales results
and top-flight customer relationship
management as the end goals. The sheer
number of customer profiles, products,
programs, and market segments can be
overwhelming to a salesperson. Through the use
of data mining, however, specific information
can be uncovered that salespeople can utilize in
the management of their sales territory. Such
insights can assist sales personnel in maintaining
Northern Illinois University
superior customer relationships. By applying
data mining techniques to existing customer data,
the potential exists for a salesperson to leverage
the results to advantage. Findings from our case
study suggest that the use of multiple linear
regression, association rules, and cluster analysis
as data mining techniques offer sales personnel
customer-oriented tools that could enhance the
consequences of their territory efforts. Two
caveats about data mining, though, should be
kept in mind. First, the findings obtained in a
given salesperson‘s territory should not be
generalized to other salespeople‘s territories; the
results are likely to be territory specific.
Therefore, sales personnel interested in using
data mining should undertake their own efforts
rather than relying on data mining knowledge
gleaned from another salesperson. Second,
statistics and analytics cannot replace human
intuition. Salespeople must still carefully analyze
the findings to assure that the data mining results
are logical.
This paper has provided guidelines for utilizing
data mining in the industrial supply industry.
Using a sample sales territory, a framework was
presented about how to apply three data mining
techniques with actual sales data to generate
feasible sales plans. In order to fully incorporate
advantages of data mining into the sales force,
periodic investigation of customer data is
requisite. Once details of a sales plan are
complete, the program should be launched.
After a given period of time, data from the
results of the new sales plan should be analyzed
to test the effectiveness of the plan. The
response data from that selling effort could be
data mined, thus affording a new look at the data
and potential modification of the sales plan.
Admittedly, many sales personnel are unlikely to
be immediately skilled to undertake data mining
on their own. After all, they generally are trained
Spring 2010 23
Academic Article
to serve as conduits between their company and
their accounts, not as statistical or computer
technicians.
Therefore, most salespeople
interested in using data mining to enhance their
relationships with each customer will need to
become cognizant of (a) the importance of data
mining, (b) useful data mining methods vis-à-vis
their respective territories, (c) means of
conducting these analyses and interpreting their
results, and (d) requisite data to undertake data
mining. As Hair et al. (2009, p. 258) state, ―…
sales reps need to know how to use the software,
what information they are required to maintain
electronically about their customers…and how
to use equipment and applications…that
enhance the automated sales environment.‖ For
progressive sales managers who see the wisdom
of using data mining at a sales territory level, the
following recommendations are offered:
1. In initial and refresher sales training
programs, trainees should be made cognizant
of what is meant by data mining and how its
tools can be beneficial to sales personnel
wishing to make use of it. Lectures may be
ideal for dissemination of such information.
Firms that already use data mining, but at a
macro (company) level, should present to
trainees the value derived from utilizing such
efforts. Company employees who use data
mining should serve as trainers. After all,
they are conversant with data mining‘s
strengths, weaknesses, and requisite
knowledge, particularly vis-à-vis the firm and
its industry.
2. Sales personnel should be trained about what
data are necessary to use various data mining
tools. This training could be received at a
community college, a four-year university, an
online university, or the company itself. The
salespeople‘s firm should inform
salespersons what data are readily available in
its data base for undertaking data mining
efforts.
3. Company salespeople will need to know how
to actually conduct various data mining
techniques (such as those used by the host
company salesperson in our case study).
Again, this training could be received at a
community college, a four-year university, an
online university, or the company. Firms
that make extensive use of data mining at a
company-wide level could have its data
mining experts do the training, as they can
relate specifically to the environment in
which its company‘s sales personnel interact.
4. Salespeople whose territories are extremely
small (e.g., fewer than 10 customers) should
be apprised by company personnel that use
of data mining might not be particularly
beneficial. Such sales personnel may well
find that acquiring unique knowledge about
each of their customers through the use of in
-depth interviews may yield more reliable
results than relying on data mining tools.
5. To enhance understanding of data mining
techniques and interpretations of those
techniques, management may suggest to
salespersons that they enroll in a basic
statistics course at some college-level
institution to facilitate their use of data
mining tools.
END NOTES
1
Classification analysis was not conducted
because all customer types were known to
the host salesperson.
2
Because of the confidential nature of
customer records, data were only available
from the host salesperson‘s territory.
3
Table 2 is presented to provide readers with
a foretaste of products the host salesperson‘s
Vol. 10, No. 2
24
Journal of Selling & Major Account Management
customers purchase. As such, information
about all twenty customers is not shown.
The brevity is consistent with DM
management‘s request to maintain as much
confidentiality as possible.
4
5
6
7
For a detailed discussion about how to
interpret multiple linear regression results,
see Hair et al. (2010). Although multiple
linear regression is often employed to
suggest a cause-effect relationship between
two variables, caution should be exercised
regarding the multiple linear regression
findings in our case study. In the present
analysis, the significant independent
variables are correlated with customer sales
revenue; whether they definitively lead to a
higher (or lower) level of sales, however, is
uncertain.
For detailed discussion about how to
conduct cluster analysis, see Hair et al.
(2010).
Prototypical software packages
containing cluster analysis include SAS,
SPSS, STATA, and XLMiner for Excel.
Each software package provides essentially
similar options for cluster analysis.
For detailed discussion about how to
conduct association rules, see Schmueli et al.
(2007) and Shaw e al. (2001). Prototypical
software packages containing cluster analysis
include SPSS Clementine, IBM Intelligent
Miner for Data and XLMiner for Excel.
Each software package provides essentially
similar options for association rules.
In addition, irregular purchases had to be
eliminated from the dataset so that the
finding would reflect the customer‘s regular
purchasing patterns. According to Kumar
(2002), a cut-off rule has to be identified to
rule out the ―accidental‖ purchase which
might lead to an ―accidental‖ association.
Northern Illinois University
Therefore, different thresholds for
eliminating the irregular purchase (also called
a ―spot buy‖) were tested. After these tests,
the host salesperson discerned that if a
purchase value was less than 5% of the
customer‘s total annual sales, it would be
regarded as an irregular purchase, and thus,
that particular purchase would be eliminated
when conducting the binary (0,1) data
transformation.
Alan J. Dubinsky is the Dillard Distinguished
Professor of Marketing at Dillard College of
Business Admnistration, Midwestern State
University in Wichita Falls, TX. He is also
Professor Emeritus at Purdue University in Wet
Lafayette, IN. Email: Dubinsky@purdue.edu
Guoying Zhang is an Assistant Professor of
Management Information Systems in the Dillard
College of Business Administration, Midwestern
State University in Wichita Fall, TX . Email:
grace.zhang@mwsu.edu
Michele Wood, MBA, Dillard College of
Business Administration at Midwestern State
University in Wichita Falls, TX.
Email:
Michele_wood@sbcglobal.net
REFERENCES
Anderson, R. E., A.J. Dubinsky, and R. Mehta
(2007), Personal Selling: Building Customer
Relationships and Partnerships, Florence, KY:
South-Western/Cengage Learning.
Berry, M.J. and G. Linoff (1997), Data Mining
Techniques for Marketing, Sales, and Customer
Support, Hoboken, NJ: John Wiley & Sons,
Inc.
Berson, A., S. Smith, and K. Thearling (1999),
Building Data Mining Applications for CRM,
New York, NY: McGraw-Hill Companies.
Academic Article
Boulding, W., R. Staelin, M. Ehret, and W.J.
Johnston (2005), ―A CRM Roadmap: What
We Know, Potential Pitfalls, and Where to
Go,‖ Journal of Marketing, 69 (October), 155167.
Cao, Y. and T.S. Gruca (2005), ―Reducing
Adverse Selection through Customer
Relationship Management,‖ Journal of
Marketing, 69 (October), 219-229.
Chen, Y., G. Zhang, D. Hu, and S. Wang
(2006), ―Customer Segmentation in
Customer Relationship Management Based
on Data Mining,‖ Knowledge Enterprise:
Intelligent Strategies in Product Design,
Manufacturing, and Management, 207, 288-293.
Chou, A., E.B. Pullins, and S. Senecal (2007),
―Technology Empowerment As a
Determinant of Salesforce Technology
Usage,‖ Journal of Selling and Major Account
Management, 7 (2), 20-29.
Company DM (2008), Company DM 2008 Annual
Report.
Crosby, L.A. and S.L. Johnson (2001), ―High
Performance Marketing in the CRM Era,‖
Marketing Management, 10, (3), 10-11.
Dickson, P.R., W.M. Lassar, G. Hunter, and S.
Chakravorti (2009), ―The Pursuit of
Excellence in Process Thinking and
Customer Relationship Management,‖
Journal of Personal Selling and Sales Management,
24 (Spring), 111-124.
Dubinsky, A.J. (1999), ―When Salespeople Fail:
Assessing Blame,” Industrial Marketing
Management, 28 (January), 19-26.
Fayyad, U., G. Piatetsky-Shapiro, and P. Smyth
(1996), ―From Data Mining to Knowledge
Discovery in Databases,‖ AI Magazine (Fall),
37-54.
Hair, J.F., W.C. Black, B.J. Babin, and R.E.
Anderson (2010), Multivariate Data Analysis,
Upper Saddle River, NJ: Prentice Hall.
Spring 2010 25
Hair, J.F., R.E. Anderson, R. Mehta, and B.J.
Babin (2009), Sales Management, Boston:
Houghton Mifflin Company.
Hunter, G.K. and W.D. Perreault (2007),
―Making Sales Technology Effective,‖ Journal
of Marketing, 71 (1), 16-34.
Javalgi, R.G., C.L. Martin, and R.B. Young
(2006), ―Market Research, Market
Orientation, and Customer Relationship
Management:
A Framework and
Implications for Service Providers,‖ Journal of
Services Marketing, 20 (1), 12-23.
Kale, S.H. (2004), ―CRM Failure and the Seven
Deadly Sins,‖ Marketing Management, 13 (#5),
42-46.
Landry, T.D., T.J. Arnold, and A. Arndt (2005),
―A Compendium of Sales-Related Literature
in Customer Relationship Management:
Processes and Technologies with Managerial
Implications,‖ Journal of Personal Selling and
Sales Management, 25 (Summer), 231-251.
Mathieu, J., M. Ahearne, and S.R. Taylor (2007),
―A Longitudinal Cross-Level Model of
Leader and Salesperson Influences on Sales
Force Technology Use and Performance,‖
Journal of Applied Psychology, 92 (2), 528-537.
Moutot, J.M. and G. Bascoul (2008), ―Effects of
Sales Force Automation on Sales Force
Activities and Customer Relationship
Management Processes,‖ Journal of Personal
Selling and Sales Management, 28 (Spring), 167184.
Olaffson, S., X. Li, and S. Wu (2006),
―Operations Research and Data Mining,‖
European Journal of Operational Research, 187,
1429-1448.
Payne, A. and P. Frow (2006), ―Customer
Relationship Management: From Strategy to
Implementation,‖ Journal of Marketing
Management, 22, 135-168.
Vol. 10, No. 2
26
Journal of Selling & Major Account Management
Payne, A. and P. Frow (2005), ―A Strategic
Framework for Customer Relationship
Management,‖ Journal of Marketing, 69
(October), 167-176.
Sumathi, S. and S. N. Sivanandam (2006), ―Data
Mining in Customer Value and Customer
Relationship Management,‖ Studies in
Computational Intelligence, 29, 321-386.
Raman, P., C.M. Wittmann, and N.A. Rauseo
(2006), ―Leveraging CRM for Sales: The
Role of Organizational Capabilities in
Successful CRM Implementation,‖ Journal of
Personal Selling and Sales Management, 26
(Winter), 39-53.
Tan P., V. Kumar, and J. Srivastava (2002),
―Selecting the Right Interestingness Measure
for Associations Patterns,‖ Proceedings of the
8th ACM SIGKDD International Conference on
KDD, 32-41.
Reinartz, W., M. Krafft, W.D. Hoyer (2004),
―The Customer Relationship Management
Process: Its Measurement and Impact on
Performance,‖ Journal of Marketing Research,
41 (August), 293-305.
Sharma, A. and J.N. Sheth (2010), ―A
Framework of Technology Mediation in
Consumer Selling: Implications for Firms
and Sales Management, Journal of Personal
Selling and Sales Management, 30 (Spring), 121129.
Shaw, M.J., C. Subramaniam, G.W. Tan, and
M.E. Welge (2001), ―Knowledge
Management and Data Mining for
Marketing,‖ Decision Support Systems, 31, 127137.
Shmueli, G., N.R. Patel, and P.C. Bruce. (2007),
Data Mining for Business Intelligence, Hoboken,
NJ: John Wiley & Sons, Inc.
Stein, A. and M. Smith (2009), ―CRM Systems
and Organizational Learning:
An
Exploration of the Relationship between
CRM Effectiveness and the Customer
Information Orientation of the Firm in
Industrial Markets,‖ Industrial Marketing
Management, 38, 198-206.
Stoddard, J.E., S.W. Clopton, and R.A. Avila
(2006), ―An Analysis of the Effects of Sales
Force Automation on Salesperson
Perceptions of Performance,‖ Journal of
Selling and Major Account Management, 6 (1), 38
-56.
Northern Illinois University
Tu, Y. (2008), ―An Application of Web-Based
Data Mining: Selling Strategies for Online
Auctions,‖ Online Information Review, 32 (2),
147-162.
Yim, F.H.K. (2002), ―CRM Orientation:
Conceptualization and Scale Development,‖
Master‘s Thesis, Shatin: Chinese University
of Hong Kong.
Yim, F.H.K., R.E. Anderson, and S.
S wa m i na t ha n ( 2 0 0 4 ) , ― C u s t om e r
Relationship Management: Its Dimensions
and Effect on Customer Outcomes,‖ Journal
of Personal Selling & Sales Management, 24
(Fall), 263-278.
Zhang, G., F. Zhou, F. Wang, and J. Luo (2008),
―Knowledge Creation in Marketing Based on
Data Mining,‖ in 2008 International Conference
on Intelligent Computation Technology and
Automation, Hunan, China: ICICTA, 782786.
Spring 2010 27
Academic Article
DEVELOPING A VALID AND RELIABLE MEASUREMENT OF
ATTITUDES TOWARD SALESPEOPLE
By Gregory S. Black and Scott G. Sherwood
The authors explain the need for a valid and reliable scale to measure attitude toward salespeople. Based
on this need, a 24-item scale is developed in a step-by-step manner, following a process prescribed by
marketing scholars. Three studies are conducted and reported in developing the scale. The first two
studies use university students to develop the items, refine the measure, and begin the process of reliability and validity analysis in a consumer setting. The third study reports a further validation of the refined
measure in an organizational setting (the electronics industry). Resulting is a valid and reliable scale
made up of two dimensions: personal characteristics and social interaction characteristics of salespeople.
This scale offers a tool that both scholars and practitioners can use confidently in assessing customer
attitudes toward salespeople. Particularly valuable is the application of this measure to helping sales
managers assess customer satisfaction.
INTRODUCTION
Of the promotional tools a marketer may use in
this integrated marketing strategy, personal
selling may be the most important. This
importance is indicated by two factors. First,
organizations spend more on personal selling
than on any other of the promotional tools,
including advertising (e.g., Johnston and Marshall
2010; Traynor and Traynor 1989). Second, an
organization‘s most significant, and sometimes
only, contact with its customers is through the
personal selling function as salespeople interact
with the customers (e.g., Black and Peeples 2005;
Johnson and Black 1996; Puccinelli 2006).
In addition to producing sales for an
organization, a primary objective of personal
selling is to create a predisposition on the part of
customers to respond favorably to the
salesperson, which in turn facilitates a favorable
attitude toward the company represented by the
salesperson (Brown 1995). In pursuit of these
objectives, companies seek to influence customer
attitudes toward their sales representatives. If
this favorable attitude toward companies,
following a positive attitude toward its
salespeople, is achieved, a company‘s customers
are likely to act in ways that facilitate not only
single transactions, but also maintain long-term
relationships (Brown 1995). Companies are
increasingly demonstrating the importance of the
salesperson/customer interaction by including
customer satisfaction in sales force
compensation plans (Widmier 2002). Despite
this importance, little effort has gone into
developing any means to assess attitudes toward
the salesperson.
As customers continue to expect increased
customer orientation by salespeople (Widmier
2002), being able to assess customer attitudes
toward salespeople will become even more
important. Salespeople are boundary spanners
who bridge the gap between a business and its
customers. It is imperative that companies know
as much about their customers as possible and
their attitudes toward salespeople should be very
important.
Customer attitudes toward a
company‘s salespeople are often the most
important factor when judging the company,
regardless of the quality of its product or other
Vol. 10, No. 2
28
Journal of Selling & Major Account Management
factors (Jones et al. 1998). In consideration of
this ever-increasing importance of the personal
selling function and salespeople in creating
customer satisfaction and establishing a
favorable reputation for a company, the purpose
of this research is to develop a valid and reliable
scale to measure customer attitude toward
salespeople.
a consistently favorable or unfavorable manner.
Further, this ability to measure bipolarity is
essential in assessing attitudes. Also, some
attitudes are held much more strongly than are
other attitudes. Thus, to measure attitudes,
bipolar items should be offered with more than
two simple answer categories. Their resulting
suggestion is the semantic-differential scale.
BACKGROUND
Guidelines for generating an appropriate pool of
potential items were also provided by Osgood et
al. (1957), as well as by Churchill (1979). The
items should be relevant to the concepts being
judged and they should exhibit a certain factorial
composition. This suggests that more than one
factor may exist in an attitude and items
generated to measure that attitude should be
related and grouped into distinct, unidimensional
factors. Furthermore, Osgood et al. (1957)
recommended that frequency of usage, as
determined by free association, be used to
generate scale items. This guideline suggests
item-generation processes, such as free
association, be used and those items that are
frequently listed to describe the attitudes are the
ones most salient in expressing those attitudes.
Churchill (1979) also recommends similar
processes to generate items.
Young and Albaum (2003) developed a measure
of trust in salesperson-customer relationships,
successfully assessing a potentially important
element of a customer‘s attitude toward
salespeople. However, this measurement falls
short of being able to robustly assess overall
attitudes toward salespeople.
Brown (1995) came closest to developing a scale
to measure overall attitude towards salespeople.
In his research, a five-item scale using the
semantic-differential format (bad-good,
ineffective-effective, not useful-useful,
unpleasant-pleasant, and helpful-unhelpful) was
used to measure attitude toward salespeople in a
business-to-business setting. However, the scale
was not developed using prescribed methods
used in marketing and was simply one variable
among many in this larger study to determine if
attitudes toward salespeople impacted both the
attitudes of salespeople and product attitudes.
The scale proved to be reliable (Cronbach‘s
alpha = .91), but no validity was assessed. Thus,
a void still remains where a valid and reliable
measure of customers‘ overall attitudes toward
salespeople, using the steps suggested in the
literature (e.g., Churchill 1979), has not been
developed.
The theoretical foundations for developing the
scale were derived directly from the work of
Osgood et al. (1957). Accordingly, attitude is a
learned predisposition to respond to an object in
Northern Illinois University
METHOD
Item Generation and List Refinement
The first step was to generate as many items as
possible from a review of the relevant literature
(Brown 1995; Comer and Jolson 1985; Gibson et
al. 1980/1981; Hayes and Hartley 1989;
Muehling and Weeks 1988; Simpson and Kahler
1980/1981; Swan and Oliver 1991).
This
resulted in a list of 60 items.
Research suggests the best salespeople have both
feminine and masculine characteristics (e.g.,
Jones et al. 1998; Siguaw and Honeycutt 1995).
Spring 2010 29
Academic Article
Thus, in an attempt to capture these dimensions,
and to generate more items, we asked students in
an undergraduate Sales Management class on the
first day of class to generate two to five words or
phrases that came to mind when they heard the
word ―salesman.‖ Two weeks later, we did the
same for ―saleswoman,‖ and two weeks after
that, we asked them to accomplish the same task
for ―salesperson.‖
We then held a focus group including eleven
undergraduate students--six women and five
men. The issue we wanted to explore in depth
was the femininity/masculinity issue.
By
discussing specific products and situations, we
found evidence from this focus group that
gender stereotyping exists in people‘s
perceptions of salespeople. For example, before
we revealed to the group that we were probing
for discussion on femininity and masculinity, we
asked them to describe a hypothetical situation
where they were faced with a computer
salesperson. Without exception, all the members
of the focus group referred to the computer
salesperson as a male. Also, in later discussion,
we learned that without exception, people would
be more comfortable having a female
salesperson come to their home at night rather
than a male. Further discussion suggested that
focus group members were not so concerned
about the actual gender of the salespeople, but
rather were more interested in the masculine and
feminine characteristics they perceived needed to
be demonstrated by salespeople in the discussed
situations. We then asked the group to perform
word generation activities to produce more
items. Combined with the items from the
literature review and from the Sales Management
class, we now had a pool of 214 words and
phrases describing salespeople.
Many of these items could be paired with other
items generated in this process to form semantic-
differential pairings. For the items that could
not be paired from the generated list, we paired
them with words that intuitively seemed to be
opposites. To begin reducing the number of
items into a more usable subset, a panel of six
business Ph.D. students was used. Each panel
member was asked to independently evaluate the
list of item pairs and eliminate ambiguous pairs
and/or suggest another word or phrase that
would be a suitable pairing for items. After this
procedure, the list was given to two marketing
professors who have extensive experience in
developing scales. They were asked to do the
same as the Ph.D. students and to make any
other suggestions or additions to the list. In
addition, while examining the items, both the
Ph.D. students and the marketing professors
were asked to give their opinions about whether
these items would measure attitudes toward
salespeople. They were unanimous in their
affirmation, suggesting face validity.
Undergraduate Sales Management students were
called upon again to evaluate each one of the 214
words or phrases and to rate them as feminine,
masculine, or both. The resulting scale consisted
of 91 items: 23 feminine items, 22 masculine
salesperson items, and 46 items that could not be
classified as either.
To achieve the parsimony recommended to
measure an attitude, the number of items
obviously had to be reduced. This task was
accomplished by three data collections. This
described procedure is common in marketing
(e.g., Malhotra 1981).
Data Collections 1 and 2
In the first data collection, data were obtained
from student subjects in various undergraduate
business classes at a major western university.
This sample resulted in 176 usable
questionnaires. We did not disclose information
Vol. 10, No. 2
30
Journal of Selling & Major Account Management
about the femininity/ masculinity of scale items;
however, we did tell respondents we were
collecting data on attitudes toward salespeople.
To manipulate the frame of mind of the
respondents and introduce the femininity/
masculinity issue, three versions of the
instructions and introductory vignette were used:
one version described a salesperson named
Theresa, another version described a salesperson
named Travis, and the final version did not give
the described salesperson a name.
Being
otherwise identical, the vignettes described the
salesperson selling coupon booklets to students
in their dorms or apartments. In addition, six
versions of the questionnaire itself were
prepared; the only difference between these
versions was the order of the items in an attempt
to control for question order bias (Asher 1988;
Kalton and Schuman 1982). Finally, some items
were designed to be reverse-coded to attempt to
control for subjects who responded to the
questionnaire without reading the items.
The second data collection used similar
procedures and employed the same disclosures
and precautions. Data were collected from an
additional 161 students. However, we wanted to
increase the possibility of femininity/masculinity
being significant, so we used names that were
perceived by the respondents in the first data
collective to be more masculine and feminine
than the names we had previously used – Chuck
and Heather. Again, a third version described a
names salesperson. The vignettes were identical
in all other aspects.
Because of the objectives of these first two data
collections, student samples were deemed to be
appropriate (Calder et al. 1981). The particular
product selected, a coupon booklet, was
intuitively thought to be familiar, relevant, and
meaningful to students.
Northern Illinois University
Purifying the Instrument
One objective was to reduce the initial set of 91
items to a smaller set of items to constitute the
final scale with a minimal loss of ability to
capture attitudes toward salespeople and to
identify dimensions of these attitudes that might
emerge. Toward these ends, three analytic
procedures—exploratory factor analysis,
confirmatory factor analysis, and reliability
analysis—were employed so that multiple criteria
could be adopted to select final scale items, as
has been previously advocated (e.g., Churchill
1979; Gerbing and Anderson 1988; Nunnally
1967).
Exploratory Factor Analysis
The main purpose of exploratory factor analysis
is data reduction. However, it can also be used
for identifying dimensions or factors that may be
related to the construct of interest. A statistical
indication of the extent to which each item is
correlated with each factor is given by the factor
loading.
In analyzing the data from the first study, the
first step, analyzing the scree plot, did not
indicate a clear solution. Many of the factors had
Eigenvalues over 2.0, the preferred cutoff value
(Gerbing and Anderson 1988). The next step
was to look at the varimax rotated factor matrix.
Two things are accomplished in this step. First,
those scale items that do not show factorial
stability and those items with factor loadings
lower than .50 were identified for elimination. In
addition varimax rotation revealed that only the
first four factors had item loadings greater
than .50. A closer examination of these four
factors was revealing. Fifteen items loading on
the first factor can be described as personal
characteristics of the salesperson (Eigenvalue =
24.82; explained variance = 27.3%). Eleven
Spring 2010 31
Academic Article
items loading on the second factor can be
described as social interaction traits of the
salesperson (Eigenvalue = 11.96; explained
variance = 13.1%). Items related to femininity/
masculinity did not appear until the third and
fourth factors. Four items loading on the third
factor were all previously identified as masculine
items (Eigenvalue = 2.72; explained variance =
3.0%). Finally, four items loading on the fourth
factor were all previously identified as feminine
items (Eigenvalue = 2.43; explained variance =
2.7%). The resulting scale consisted of 34 items.
analyzed to see if indeed they do fit with the
appropriate items. The comparative fit index for
the model resulting from the first data collection
is .5745. This is not a good fit, according to the
established guideline of .90 (Gerbing and
Andersen 1988). In addition, the normalized
estimate should be at least .70, but this analysis
produced only a .287. The interpretation of
these results is that the four dimensions
identified are not unidimensional. As this
analysis is not normally a data reduction
measure, no items were eliminated in this step.
Using the same procedures with the second set
of data, only the first two factors had more than
one item loading greater than .50. Fourteen
items loading on the first factor again were
related to the personal characteristics of the
salesperson (Eigenvalue = 11.54; explained
variance = 33.9%). Ten items loading on the
second factor were also related and can again be
described as social interaction traits of the
salesperson (Eigenvalue = 6.15; explained
variance = 18.1%). The two dimensions based
on femininity and masculinity did not appear in
any factor at all, despite our overt attempt to
strengthen these factors.
The comparative fit index for the model
resulting from the second data collection
is .8594. This is a better fit, despite the
recommended level of .90 (Gerbing and
Anderson 1988). In addition, the normalized
estimate should be at least .70, but this analysis
produced only a .404. This is an improvement
from the previous version of the scale. The
interpretation of these results is that the two
dimensions or factors identified are reasonably
unidimensional and seem to provide a fair fit,
suggesting the scale is measuring a person‘s
attitude toward salespeople.
It is important to assess the correlations between
factors (p = 0.610), showing the two variables
are significantly correlated, but also
demonstrating they are different enough that
both factors are not measuring the same
dimension of attitude toward salespeople The
resulting scale consists of 24 items. See Tables 1
and 2 for a summary of these items.
Confirmatory Factor Analysis
In confirmatory factor analysis, the fit of the
model is assessed in order to determine
unidimensionality of the scale or of the various
dimensions of the scale.
The factors or
dimensions that have been identified are
Reliability Analysis
After the number of items has been reduced and
the unidimensionality of the scale is determined,
its overall reliability and the reliability of its two
dimensions should be assessed (Gerbing and
Anderson 1988). After reducing the scale from
the second data collection is reliable (Cronbach‘s
alpha = .92). The recommended minimum score
for Cronbach‘s alpha should be .70 (Nunnally
1967). It is also necessary to look at each of the
dimensions separately. The dimension related to
sonal characteristics of a salesperson is also
reliable (Cronbach‘s alpha = .94), as is the
dimension related to social interaction traits of a
salesperson (Cronbach‘s alpha = .89).
Vol. 10, No. 2
32
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Journal of Selling & Major Account Management
TABLE 1. FACTOR PATTERN MATRIX: SECOND DATA COLLECTION
Scale Items
Factor 1
Factor 2
Communality
Sympathetic-Indifferent
0.158
0.427
0.441
Willing to take risks– Afraid to take risks
0.529
-0.146
0.742
Self confident-Not confident
0.616
0.092
0.817
Assertive-Inhibited
0.561
-0.179
0.710
Enthusiastic-Not enthusiastic
0.577
-0.151
0.695
Outspoken-Quiet
0.744
-0.004
0.687
Trustworthy-Untrustworthy
-0.090
0.810
0.759
Talkative-Quiet
0.865
-0.020
0.783
Cheerful-Dreary
0.543
0.028
0.789
Understanding-Intolerant
-0.135
0.514
0.540
Loyal-Disloyal
0.015
0.768
0.613
Reliable-Unreliable
0.016
0.867
0.771
Good-Bad
Competent-Incompetent
Outgoing-Shy
Sincere-Phony
Ambitious-Lazy
Energetic-Lethargic
Courteous-Discourteous
Caring-Uncaring
Good talker-Bad talker
Has leadership abilities– Is unable to lead
Makes decisions easily– Finds decision-making difficult
Has a strong personality– has a weak personality
Factor 1/Factor 2 Correlation
Explained Variance
Third Data Collection: Establishing External
Validity
To further validate this measurement, the 24item scale resulting from the first two data
collections was included in a larger study to
assess various related phenomena in the
electronics components industry (SIC 3679).
Each of the firms in the chosen sample
(electronic components manufacturers) was
initially screened by telephone. During these
Northern Illinois University
0.157
0.436
0.838
-0.152
0.670
0.812
0.373
-0.009
0.772
0.535
0.569
0.578
0.610
33.90%
0.662
0.465
0.062
0.802
0.257
0.062
0.534
0.823
0.008
0.396
0.236
-0.155
0.653
0.634
0.757
0.741
0.727
0.727
0.754
0.752
0.760
0.777
0.770
0.700
18.10%
telephone conversations, these manufacturers
were asked to identify at least one major buyer
firm and to give contact information for that key
informant who should be most knowledgeable
about the relationships and the issues being
investigated (Anderson et al. 1994; Kumar et al.
1995).
Of the 1,062 original manufacturers, 768
provided the requested information. These 768
key informants were they contacted by telephone
Spring 2010 33
Academic Article
Table 2. SCALE ITEMS BY FACTORS/DIMENSIONS: SECOND DATA SET
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Scale Anchors
Factor
Willing to take risks- Afraid to take risks
Self confident-Not confident
Assertive-Inhibited
Enthusiastic-Not enthusiastic
Outspoken-Quiet
Talkative-Quiet
Cheerful-Dreary
Outgoing-Shy
Ambitious-Lazy
Energetic-Lethargic
Good talker-Bad talker
Has leadership abilities– Is Unable to lead
Makes decisions easily– Finds decision-making difficult
Has a strong personality- Has a weak personality
Sympathetic-Indifferent
Trustworthy-Untrustworthy
Understanding-Intolerant
Loyal-Disloyal
Reliable-Unreliable
Good-Bad
Competent-Incompetent
Sincere-Phony
Courteous-Discourteous
Caring-Uncaring
Personal Characteristic
Personal Characteristic
Personal Characteristic
Personal Characteristic
Personal Characteristic
Personal Characteristic
Personal Characteristic
Personal Characteristic
Personal Characteristic
Personal Characteristic
Personal Characteristic
Personal Characteristic
Personal Characteristic
Personal Characteristic
Social Interaction Trait
Social Interaction Trait
Social Interaction Trait
Social Interaction Trait
Social Interaction Trait
Social Interaction Trait
Social Interaction Trait
Social Interaction Trait
Social Interaction Trait
Social Interaction Trait
and asked to participate in our study.
Subsequently, each company was sent a prenotification letter, which explained the study,
emphasized its importance, reminded the key
informant that he or she agreed to participate,
and further urged him or her to respond in a
timely fashion.
Several days later, the
questionnaire package was mailed to the key
informants in each organization, including the
questionnaire, a cover letter explaining the study
and assuring confidentiality, and a postage-paid
return envelope.
The steps in the data collection procedure were
expected to result in a respectable response rate.
The response rate was expected to be equivalent
to similar research. Acceptable rates run from
12% to 30% (e.g., Boyle et al. 1992). Of the 768
companies, 107 returned usable questionnaires,
resulting in a response rate of 13.9%. The
response rate was lower than expected, but was
in the acceptable range for studies of this nature.
Exploratory Factor Analysis
Unlike the previous data collections, which were
designed to purify the instrument, this
collection‘s purpose was to confirm, or validate,
the two factors, using the same procedures as
outlined previously. Table 3 shows the loadings
Vol. 10, No. 2
34
Journal of Selling & Major Account Management
TABLE 3. FACTOR PATTERN MATRIX: THIRD DATA COLLECTION
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Scale Items
Sympathetic-Indifferent
Willing to take risks-Afraid to take risks
Self confident-Not confident
Assertive-Inhibited
Enthusiastic-Not enthusiastic
Outspoken-Quiet
Trustworthy-Untrustworthy
Talkative-Quiet
Cheerful-Dreary
Understanding-Intolerant
Loyal-Disloyal
Reliable-Unreliable
Good-Bad
Competent-Incompetent
Outgoing-Shy
Sincere-Phony
Ambitious-Lazy
Energetic-Lethargic
Courteous-Discourteous
Caring-Uncaring
Good talker-Bad talker
Has leadership abilities- Is unable to lead
Makes decisions easily-Finds decision making difficult
Has a strong personality-Has a weak personality
Explained Variance
of the 24 items on the two factors. A close
assessment of these two factors indicates
generally higher loadings on the factors than did
the first two data collections. The same fourteen
items loaded on the first factor (Eigenvalue =
16.12; explained variance = 39.8%). Also, the
same ten items loaded on the second factor
(Eigenvalue = 7.57; explained variance = 24.3%).
Reliability Analysis
Again, after the unidimensionality of the two
dimensions of the scale is determined, the next
step is to assess the scales overall reliability and
the reliability of each dimension (Gerbing and
Northern Illinois University
Factor 1
0.27935
0.65023
0.62058
0.58029
0.75580
0.75580
-0.18716
0.88980
0.54416
-0.11801
0.05753
0.07510
0.35184
0.38008
0.79591
-0.24662
0.73217
0.82819
0.22321
-0.00923
0.77624
0.58237
0.68078
0.69699
39.8%
Factor 2
0.59172
-0.30040
0.12597
-0.17242
-0.21264
-0.01360
0.84161
-0.06948
0.06378
0.55007
0.78007
0.88841
0.72394
0.46722
0.05666
0.78971
0.26041
0.05727
0.57987
0.86112
0. 007
0.23336
0.17423
-0.19545
24.3%
Communal0.43473
0.74671
0.87516
0.69841
0.70288
0.63588
0.76442
0.79973
0.80186
0.54966
0.62484
0.79494
0.60403
0.64917
0.77038
0.78348
0.69984
0.73795
0.78310
0.78727
0.77017
0.78409
0.77050
0.70236
Anderson 1988). The overall reduced scale (24
items) appears to be quite reliable (Cronbach‘s
alpha = .94). The dimension related to personal
characteristics of a salesperson is also reliable
(Cronbach‘s alpha = .97), as is the dimension
related to social interaction traits of a salesperson
(Cronbach‘s alpha = .91).
Nomological and Discriminant Validity
Nomological validity is known as the degree to
which a construct behaves as it should within a
system of related constructs. Discriminant
validity is described as the distinct difference
between a construct and other constructs in a
study. A common method to assess these types
Academic Article
Spring 2010 35
TABLE 4. SUMMARY OF CORRELATION ANALYSIS TO ASSESS DISCRIMINANT
AND NOMONLOGIAICAL VALIDITY
DTRU
CTRU BTRUS
ATTDMUN
CALST
ST
T
SALES
C
COM
ATTCOM
DTRUST
1.00
CTRUST
.05
1.00
BTRUST
.11
.84
1.00
ATTSALES
.05
.62
.72
1.00
DMUNC
.16
.27
.31
.23
1.00
CALCOM
.01
.13
.05
.24
.03
1.00
1.00
ATTCOM
.08
.67
.68
.55
.29
.45
of validity is to include the measure in a theorygrounded empirical study that includes
hypotheses relating this construct to others, and
then examine the correlations between the
construct of interest and other constructs
included in the study (Churchill 1979; Peter
1981). Table 4 presents these correlations.
As can be seen, attitude toward salespeople
(ATTSALES) is highly correlated (nomological
validity), but not perfectly so (discriminant
validity) to several of the other variables used in
this study, including credibility trust (CTRUST)
(p = .62), benevolent trust (BTRUST) (p = .72),
and attitudinal commitment (ATTCOM) (p
= .55), suggesting nomological validity. On the
other hand, it fails to correlate highly with the
other three variables in the study, including
dispositional trust (DTRUST) (p = .05),
calculative commitment (CALCOM) (p = .24),
and decision-making uncertainty (DMUNC) (p
= .23), suggesting discriminant validity.
DISCUSSION AND CONCLUSION
Regardless of the discovery that femininity- and
masculinity-related factors are not important in
the formation of attitudes toward salespeople,
the primary purpose of this research was to
develop a scale to measure customer attitudes
toward the salesperson. Through the three data
collections described, a valid and reliable scale to
measure these attitudes has been successfully
developed.
The development process
discovered two important dimensions or factors
that make up attitudes toward salespeople –
personal characteristic of the salespeople and
their social interaction traits. In today‘s world
where customers expect salespeople to provide
benefits to them personally (consumers) or to
their organizations, an ability to interact
favorably with customers is increasingly
becoming a valued and necessary characteristic
of the modern sales force.
A study by Anselmi and Zemanek (1997)
supports these findings. The findings of a
rigorous study of 450 industrial salespeople
indicate the more a salesperson exhibits good
interpersonal skills, the greater the level of buyer
satisfaction. Other studies have confirmed the
importance of both interpersonal skills
(Chakrabarty et al. 2010; Reinhard et al. 2006)
and personal characteristics of salespeople in
customer satisfaction (e.g., Birt and Vigar 2002;
Chimonas et al. 2007; Lan 2003; Reinhard et al.
2006; Zipkin and Steinman 2004), especially the
trustworthiness of the salesperson and the
Vol. 10, No. 2
36
Journal of Selling & Major Account Management
salesperson‘s knowledge of the products being
sold (e.g., Campbell et al. 2006; DeCarlo 2005;
Tsai et al. 2010; Young and Albaum 2003).
The implications of this study are valuable and
clear. There now exists a valid and reliable
instrument to capture customer attitudes toward
salespeople. For academicians, this scale can be
widely applied to examine various phenomena in
marketing.
Implications for practitioners are even more
pronounced and numerous. Companies invest
much time and money trying to establish and
then assess customer satisfaction (Reichheld
2003). For the past 15-20 years, the trend
toward including factors other than sales revenue
in salespeople‘s compensation plans makes the
ability to assess customer attitudes toward
salespeople crucial. Other than sales revenue,
customer satisfaction is the most commonly
used additional ingredient in a salesperson‘s
compensation (Journal of Accountancy 1995).
Compensation for customer satisfaction is
offered both as part of the salary of salespeople
(Hauser et al. 1997) and non-salary incentives
and bonuses (O‘Connell and Marchese 1995;
Widmier 2002). Customer satisfaction with
salespeople is likely made up of several factors.
A favorable attitude toward salespeople, as
indicated by a measure based on a salesperson‘s
personal characteristic and his or her social
interaction traits is likely to be important in
customer satisfaction.
Thus, the measure
developed in this study is important in assessing
customer satisfaction, and determining customer
satisfaction is important since as much of 58%
of organizations partially base their salespeople‘s
compensation (Journal of Accountancy 1995).
Favorable attitudes toward salespeople also have
the potential to play a crucial role in customer
retention, thus enabling the development of long
-term relationships with customers. Every time
Northern Illinois University
a company needs to find new customers to
replace lost ones, it costs the company resources
in time, effort, and revenue (Johnston and
Marshall 2010). Customers will not be likely to
become loyal and offer opportunities for
companies to establish these valuable long-term
relationships if they are forced to interact with
salespeople who are unpleasant either because of
their social interaction skills or because of their
personal characteristics. Customers will naturally
seek out interactions and relationships that are
more pleasant for them, especially when there
are no major differences in the products being
sold. Thus, salespeople who are able to generate
these favorable attitudes toward themselves from
their customers will provide a valuable
competitive advantage for their employers.
Finally, this measurement identifies 14 personal
characteristics and ten social interaction traits
important in establishing positive attitudes
toward salespeople.
Since the scale was
developed and validated in both the consumer
and the organizational environments, companies
should feel confident in using these
characteristics as part of a screening process
when recruiting salespeople.
In addition,
specific sales force training could be developed
to help sales teams develop these important
traits. Identifying the specific traits that make
salespeople more successful and then
implementing training to instill these traits into a
sales force is becoming more crucial in the everincreasing competitive environment faced in
most industries (Reday et al. 2009).
The development of this scale has helped to
identify several areas where additional research is
needed. Factors, such as customer experience
with salespeople (Bristow et al. 2006; Castleberry
1990; Honea et al. 2006), incidental similarity
(shared birthdays, etc.) (Jiang et al. 2010), and
customer moods and emotions (Dahl et al. 2005;
Academic Article
Puccinelli 2006) may also play critical roles in a
person‘s attitude toward salespeople. These
areas should be researched further to deepen our
understanding of the customer-salesperson
interaction.
Gregory S. Black, Ph.D., is an Assistant
Professor of Marketing, Metropolitan State
College of Denver, Department of Marketing.
Campus Box 79, P.O. Box 173362, Denver, CO
80217-3362. (303)352-7146. Email:
gblack4@mscd.edu.
Scott G. Sherwood is an Visiting Assistant
Professor of Marketing, Metropolitan State
college of Denver, Department of Marketing,
Campus Box 79, P.O. Box 173362, Denver, CO
80217-3362.
(303)352-4499/
Email:
sherwoos@mscd.edu
REFERENCES
Anderson, James C., Hakan Hakansson, and Jan
Johanson (1994), "Dyadic Business
Relationships within a Business Network
Context," Journal of Marketing, 58 (October),
1-15.
Anonymous (1995), ―A Shift in Sales
Compensation,‖ Journal of Accountancy, 180
(5), 17.
Anselmi, Kenneth and James E. Zemanek, Jr.
(1997), ―Relationship Selling: How Personal
Characteristics of Salespeople Affect Buyer
Satisfaction,‖ Journal of Social Behavior and
Personality, 12 (2), 539-550.
Asher, Herbert (1988), Polling and the Public: What
Every Citizen Should Know. Washington, DC:
Congressional Quarterly.
Birt, M. and D. Vigar (2002), ―A Study in Retail
Sales Person Effectiveness,‖ South African
Journal of Business Management, 33 (1), 31-39.
Spring 2010 37
Black, Gregory S. and Donna K. Peeples (2005),
―The Impact of a Propensity for
Relationalism and Market Growth on
Distribution Channel Outcomes.‖ Journal of
Business Strategies, 22 (2), 119-133.
Boyle, Brett, F. Robert Dwyer, A. Robicheaux,
and James T. Simpson (1992), "Influence
Strategies in Marketing Channels: Measures
and Use in Different Relationship
Structures," Journal of Marketing Research, 29
(November), 462-73.
Bristow, Dennis M., Rajesh Gulati, Douglas
Amyx, and Jennifer Slack (2006), ―An
Empirical Look at Professional Selling From
a Student Perspective,‖ Journal of Education for
Business, 82 (May/June), 242-249.
Brown, Steven P. (1995). ―The Moderating
Effects of Insupplier/Outsupplier Status on
Organizational Buyer Attitudes,‖ Journal of the
Academy of Marketing Science, 23 (3), 170-181.
Calder, Bobby J., Lynn W. Phillips, and Alice M.
Tybout (1989), ―Designing Research for
Application,‖ Journal of Consumer Research, 8
(September), 197-207.
Campbell, Kim Sydow, Lenita Davis, and Lauren
Skinner (2006), ―Rapport Management
During the Exploration Phase of the
Salesperson-Customer Relationship,‖ Journal
of Personal Selling & Sales Management, 26 (4),
359-370.
Castleberry, Stephen B. (1990), ―Selling & Sales
Management in Action: The Importance of
Various Motivational Factors to College
Students Interested in Sales Positions,‖
Journal of Personal Selling & Sales Management,
10 (Spring), 67-72.
Chakrabarty, Subhra, Gene Brown, and Robert
E. Widing, II (2010), ―The Effects of
Perceived Customer Dependence on
Salesperson Influence Strategies,‖ Journal of
Personal Selling & Sales Management, 30 (4), 327
-341.
Vol. 10, No. 2
38
Journal of Selling & Major Account Management
Chimonas, Susan, Troyen A. Brennan, and
David J. Rothman (2007), ―Physicians and
Drug Representatives: Exploring the
Dynamics of the Relationship,‖ Journal of
General Internal Medicine, 22, 184-190.
Honea, Heather, Andrea C. Morales, and Gavan
J. Fitzsimons (2006), ―1=2: When a Singular
Experience Leads to Dissociated
Evaluations,‖ Journal of Consumer Psychology, 16
(2), 124-134.
Churchill, Gilbert A., Jr. (1979), ―A Paradigm for
Developing Better Measures of Marketing
Constructs,‖ Journal of Marketing Research, 16
(February), 64-73.
Jiang, Lan, Joandrea Hoegg, Darren W. Dahl,
and Amitava Chattopadhyay (2009), ―The
Persuasive Role of Incidental Similarity on
Attitudes and Purchase Intentions in a Sales
Context,‖ Journal of Consumer Research, 36 (5),
778-791.
Comer, Lucette B. and Marvin A. Jolson (1985),
―Sex-Labeling of Selling Jobs and Their
Applicants,‖ Journal of Personal Selling and Sales
Management, 5 (May), 15-22.
Dahl, Darren W., Heather Honea, and Rajesh V.
Manchanda (2005), ―Three Rs of
Interpersonal Consumer Guilt: Relationship,
Reciprocity, Reparation,‖ Journal of Consumer
Psychology, 15 (4), 307-315.
DeCarlo, Thomas E. (2005), ―The Effects of
Sales Message and Suspicion of Ulterior
Motives on Salesperson Evaluation,‖ Journal
of Consumer Psychology, 15 (3), 238-249.
Gerbing, David W. and James C. Anderson
(1988), ―An Updated Paradigm for Scale
Development
Incorporating
Unidimensionality and Its Assessment,‖
Journal of Marketing Research, 25 (May), 186-92.
Gibson, C. Kendrick, Daniel S. Cochran, and
Donald E. Epley (1980/81), ―Impact of Sex
Role Differences upon the Frame of
Reference,‖ Journal of Personal Selling and Sales
Management, 1 (Fall/Winter), 41-7.
Hauser, John R., Duncan I. Simester, and Birger
Wernerfelt (1997), ―Side Payments in
Marketing,‖ Marketing Science, 16 (3), 246-255.
Hayes, H. Michael and Steven W. Hartley (1989),
―How Buyers View Industrial Salespeople,‖
Industrial Marketing Management, 18, 73-80.
Northern Illinois University
Johnson, Jean L. and Gregory S. Black (1996),
"The Effects of Relationalism and Supplier
Replaceability on Industrial Distribution
Channel Outcomes," Journal of Marketing
Channels, 25-44.
Johnston, Mark W. and Greg W. Marshall
(2010), Sales Force Management, 10th ed., New
York: Irwin McGraw-Hill, Inc.
Jones, Eli, Jesse N. Moore, Andrea J. S.
Stanaland, and Rosalind A. J. Wyatt (1998),
―Salesperson Race and Gender and the
Access and Legitimacy Paradigm: Does
Difference Make a Difference?‖ Journal of
Personal Selling & Sales Management, 18 (4), 7188.
Kalton, Graham and Howard Schuman (1982),
―The Effect of the Question on Survey
Responses: A Review,‖ Journal of the Royal
Statistical Society, 145, 42-73.
Kumar, Nirmalya, Lisa K. Scheer, and JanBenedict E. M. Steenkamp (1995), "The
Effects of Supplier Fairness on Vulnerable
Resellers," Journal of Marketing Research, 32
(February), 54-65.
Lan, Pei-Chia (2003), ―Working in a Neon Cage:
Bodily Labor of Cosmetics Saleswomen in
Taiwan,‖ Feminist Studies, 29 (1), 21-45.
Malhotra, Naresh K. (1981), ―A Scale to Measure
Self-Concepts, Person Concepts, and
Product Concepts,‖ Journal of Marketing
Research, 18 (November), 456-64.
Academic Article
Spring 2010 39
Muehling, Darrel D. and William A. Weeks
(1988), ―Women‘s Perceptions of Personal
Selling: Some Positive Results,‖ Journal of
Personal Selling and Sales Management, 8 (May),
11-20.
Simpson, Edwin K. and Ruel C. Kahler
(1980/81), ―A Scale for Source Credibility,
Validated in the Selling Context,‖ Journal of
Personal Selling and Sales Management, 1 (Fall/
Winter), 17-25.
Nunnally, Jum C. (1967), Psychometric Theory. New
York: McGraw-Hill, Inc.
O‘Connell, Bill and Lisa Marchese (2005),
―Paying for the Relationship, Not Just the
Sale,‖ Journal of Compensation & Benefits, 11
(1), 32-39.
Swan, John E. and Richard L. Oliver (1991), ―An
Applied Analysis of Buyer Equity
Perceptions and Satisfaction with
Automobile Salespeople,‖ Journal of Personal
Selling and Sales Management, 11 (Spring), 1526.
Osgood, Charles E., George J. Suci, and Percy
H. Tannenbaum (1957), The Measurement of
Meaning. Urbana, IL: University of Illinois
Press.
Traynor, Kenneth and Susan C. Traynor (1989),
―Marketing Approaches Used by High Tech
Firms,‖ Industrial Marketing Management, 23
(November), 281-87.
Peter, J. Paul (1981), ―Construct Validity: A
Review of Basic Issues and Marketing
Practices,‖ Journal of Marketing Research, 18
(May), 133-145.
Tsai, Ming-Tien, Chao-Wei Chin, and ChengChung Chen (2010), ―The Effect of Trust
Belief and Salesperson‘s Expertise on
Consumer‘s Intention to Purchase
Nutraceuticals: Applying the Theory of
Reasoned Action,‖ Social Behavior and
Personality, 38 (2), 273-288.
Puccinelli, Nancy M. (2006), ―Putting Your Best
Face Forward: The Impact of Customer
Mood on Salesperson Evaluation,‖ Journal of
Consumer Psychology, 16 (2), 156-162.
Reday, Peter A., Roger Marshall, and A.
Parasuraman (2009), ―An Interdisciplinary
Approach to Assessing the Characteristics
and Sales Potential of Modern Salespeople,‖
Industrial Marketing Management, 38 (7), 838844.
Reichheld, F. F. (2003), ―The One Number You
Need to Grow,‖ Harvard Business Review, 81
(12), 46-54.
Reinhard, Marc-Andre, Matthias Messner, and
Siegfried Ludwig Sporer (2006), ―Explicit
Persuasive Intent and Its Impact on Success
at Persuasion – The Determining Roles of
Attractiveness and Likeableness,‖ Journal of
Consumer Psychology, 16 (3), 249-259.
Widmier, Scott (2002), ―The Effects of
Incentives and Personality on Salesperson's
Customer Orientation,‖ Industrial Marketing
Management, 31 (7), 609-615.
Young, Louise and Gerald Albaum (2003),
―Measurement of Trust in SalespersonCustomer Relationships in Direct Selling,‖
Journal of Personal Selling and Sales Management,
23 (3), 253-269.
Zipkin, Daniella and Michael A. Steinman
(2005), ―Interactions Between
Pharmaceutical Representatives and Doctors
in Training: A Thematic Review,‖ Journal of
General Internal Medicine, 20, 777-786.
Siguaw, Judy A. and Earl D. Honeycutt, Jr.
(1995), ―An Examination of Gender
Differences in Selling Behaviors and Job
Attitudes,‖ Industrial Marketing Management, 24
(1), 45-52.
Vol. 10, No. 2
40
Journal of Selling & Major Account Management
Why Unified Communications Makes Sense Today
By Andrew Gorski
WHERE DID THE TIME GO?
I recently had the opportunity to spend some
one on one time with my father. Since I have a
wife and three kids, a job and other social
responsibilities, it seems that I have less and less
alone time with him. It seems that most times
that we‘re together are during the holidays
(where other family members are present) or at
one of my kid‘s activities (which do not allow for
much intimate interaction).
This year in the spring and we took some time to
get our family cottage open for the season. We
put the boat in the water, cut the grass and then
took time for dinner and a couple of beers at a
local tavern. Over a beer (or two), my father
commented to me that if he had one regret in
life, it was that he didn‘t spend enough time with
my brother, sister and me growing up. He was a
heating/air conditioning guy and spent a lot of
weekends doing side jobs for extra income.
As I thought about it, I told him that I don‘t
know many working parents today who don‘t
have the very same sentiment. My own job
requires me to travel several days per week and
although I make many of my kid‘s activities, I
miss more than I would like. However, when I
compare the time I am able to spend with my
family during the work week, I think back to the
way it was.
I spent over ten years working in a cubicle at my
corporation both as a salesman and a sales
manager. During that span of time, I was always
up early to the train, worked the day downtown
Northern Illinois University
and typically made it home at 6:30 PM. During
those years, I rarely had the opportunity to have
dinner with my family since my kid‘s day began
at 6am and they were often starving when they
got home from school.
Nowadays, when I don‘t travel, I wake up at 6
AM and take the sixty second ―commute‖ to my
home office, I make coffee, boot up the
computer and I‘m working. The tools my
company provides me with allows me to
communicate with and lead my team, who are
scattered throughout sixteen cities, as if I was
back in my cubicle in downtown Chicago.
What has allowed this miraculous
transformation? Collaboration tools! The tool
that integrate web and document sharing, video
and audio conferencing has made it possible for
the road warrior to work efficiently and
effectively and reduce costs. Even though these
tools are becoming more prevalent in the world
today, there is much more opportunity to
implement and use this technology than is being
realized. Putting together a personal strategy for
utilizing these tools can provide higher quality
meetings, higher rates of job satisfaction,
increased employee retention as well as
providing more time to spend with the family or
other desired activities.
HOW TO: CONDUCT HIGH QUALITY
REMOTE MEETINGS
Many of us host audio and web meetings where
attendees spend time ―multitasking‖ and paying
limited attention. In many cases, the larger the
Academic Article
11.4 if I‘m busy, on the phone or available for a
quick chat which save them time and
tremendously increases the ability to contact me
and get a quick answer when necessary. Instant
messaging like this has tremendously increased
one‘s ability to contact someone and be
contacted and in many cases has replaced the old
―email someone and wait back for a response‖
practice of days gone by.
I‘m also lucky enough to have a tool which will
take any phone call I get at my desk phone, to
my extension or any other phone I designate and
have it ring all phones including my smart phone
at the same time. This tool ensures that I never
miss the opportunity to pick up a caller which
keeps me productive.
Outside of being in flight (and we all know that
is changing too,) unified communications tools
like the ones I have at my disposal lead me to
always be ―in the office‖ when I‘d like to be.‖
During times where I‘m in the office and don‘t
want to be disturbed, these tools also offer users
the chance to update their ―status‖ to busy or do
not disturb and others allow users to hide. In
the end, these tools allow the user to design
when and how it‘s best to contact them as well
as when they don‘t wish to be contacted.
HOW CAN WE AFFORD IT?
Unified communications (UC) means different
things to different people, but much of what I
have discussed above is exactly that. While some
people choose to think of components of UC
such as VoiP, IM or email as unified
communications, I believe that the tools that
allow one to communicate anywhere, anytime,
with the device of their choosing and through
the richest medium possible is truly unified
communications.
Spring 2010 41
When taking a look at your own company, think
about how you communicate today and where
you think you could do better. Many companies
already have VoiP (Voice over IP) implemented
and many of the tools I discussed above are add
on products to what is already owned. When
measuring the impact and ROI, I recommend
looking hard at both soft and hard cost savings
as well as productivity, job satisfaction boosters
and time savings.
How much could be saved just by holding
meetings in high quality video?
Some of the commonly measured hard cost
savings would be relatively easy:
Business travel costs (Estimate or Calculate)
Cost of Flights
Transportation to and from airport +
parking
Cost of Rental cars + Insurance + fuel
Cost of Lodging
Cost of Meals
After examining how much each one of them
cost, then multiplying that by the frequency you
can get an estimate of the total hard costs of
making the face-to-face business trips. This then
becomes the amount of hard dollars that can be
saved (if no additional expenses for technology
are needed).
To illustrate, I‘ll share the story of one of my
customers, a public utility company in the
Midwest. They have over 5,000 employees
employed throughout the Midwest with three
main offices, all within about 300 miles of each
other. Their challenge was to cut costs while
maintaining a stable business climate. They had
many employees traveling between offices which
Vol. 10, No. 2
42
Journal of Selling & Major Account Management
led to expenses for mileage, lodging, food, etc.
This company figured out the cost of each trip,
the number of trips that could be avoided by
using high definition video and figured out that
their $1.2M investment would pay for itself in
less than one year. They ended up outfitting
about four to five conference rooms with Cisco
TelePresence in each location and cut down on
an incredible amount of travel. Other benefits
other than cutting costs were that employees
didn‘t have to travel overnight to as many
meetings, they didn‘t lose productivity while
making the flight/drive to other locations and
they were able to ―meet‖ with coworkers in
other cities and still make it home for dinner
with their families.
Some of the commonly measured soft cost
savings would be:
Time lost to travel (reduced productivity)
Fatigue of traveler
Stress on traveler (work-family balance out
of line)
Unhappy spouse/children
Cash flow issues is using personal credit
cards for travel
Then calculate your ROI to show the benefits of
using Unified Communications.
Besides Cost Savings here are a few Common
Productivity Boosters that can be obtained:
Reduction in attendee Multi Tasking – How
much more productive would your meetings
be if all attendees were actively engaged and
always paying attention?
Enhanced Communication – It has been
noted that over 60% of communication is
non verbal. Video allows people to utilize
both verbal and nonverbal communications.
Northern Illinois University
Always Available Coworkers
How much more work can a company
get done if they didn‘t have to wait for
emails and voicemails to be returned?
What is the value of reaching the right
personright away?
Job Satisfaction Boosters:
Reduced Commute Time
Without a commute, employee spends
more time working on the job while if
working from home.
Without a commute, employee is able to
take short breaks from the work day for
family activities like dinner, school
functions, etc. Also the employee is able
to easily get back to work after taking a
break.
The result of over three years of using unified
communication tools; I‘ve had more dinners
with my family and missed fewer activities than I
did in the ten previous years. Although I believe
that I‘ll always feel regret that I don‘t spend
enough time with my children, I‘m grateful that
technology is helping me be better than ever.
I‘m also more available to my coworkers, my
business partners and customers.
That‘s
something hard to put a price tag on.
Andy Gorski is a Sales Manager in the
Information Technology Industry for CDW,
120 S. Riverside Plaza, Chicago, IL.
Email: andygor@cdw.com
Spring 2011 43
Application Article
What Candor Can Do for You
By John Costigan
Webster‘s definition of Candor is defined as the
state or quality of being frank, open, and sincere
in speech or expression. In Jack Welch‘s book
―Winning‖ he explains the significance of how it
helps companies cut costs and speed up
decisions. In this article, we will explain how
candor will:
1. Help you win business
2. Set you apart from the competition
3. Allow you to make significantly more money.
THE KAREN STORY
Dateline:
Q/4, 2010
Location:
Redwood Shores, California
Customer:
Oracle
Department:
Oracle Direct / Inside Sales
In the fourth quarter of 2010, we conducted a
sales training class for approximately 50 inside
sales people where the focus was on prospecting
and cold calling. Upon completion of that class,
we split off with the sales people individually to
observe their performance in real life settings at
their desks. This reinforcement program is called
―Candor in the Cube‖ where honest feedback,
good or bad, is conveyed to the sales
representative and their managers to help change
behaviors for the long term. One such sales
person named Karen was in a difficult
predicament. Reason being, there was an
opening in the schedule and her boss wanted us
to take advantage of the gap in the schedule and
review her phone skills. There was just one
problem: She hadn‘t attended our training class!
To say she would have been flying blind would
have been an understatement.
Here is a
salesperson who has been with the company for
roughly 90 days, hardly knows where the rest
rooms are located let alone any breath to the
company‘s products and she is being asked to
call into the CIO of a Fortune 1000 company.
Karen‘s attitude was great and even though she
was nervous, she forged ahead to give it a shot.
Our goal on this particular call was to expand
the business of an existing customer. As
instructed, she started with the main 800 number
and went right to the President‘s office and read
the script provided to her perfectly, which
started with, ―I need some help…‖ In a
moments notice, we were being transferred to
the CIO‘s office where she left a perfect voice
mail message there as well.
What‘s amazing is Karen was shaking when she
hung up. She was telling us, ―Oh My God, there
is no way he is going to call me back. He
probably gets these messages every day, and if he
does call back, I don‘t know what I am going to
say. What if he ...‖ Ring, Ring. Just 2 minutes
after the voice mail was left, the CIO was calling
her back. She looked at us and said, ―Now
what? I didn‘t expect him to call back.‖ Our
answer was simple. Apply candor and simply tell
the truth. That‘s it, be courageous enough to tell
the truth and watch what happens.
―This is Bob Smith. Is this Karen? (Obviously
not a warm and friendly start.) Karen said, ―Hi
Bob. I‘m so glad you called back. The reason for
my slightly cryptic voice mail was most messages
ramble on and on and result in little or no
response. Secondly, I‘m new and need some
help. We wanted to say thank you for the
Vol. 10 No. 2
44
Journal of Selling & Major Account Management
current business you have with us and we
wanted ask you from a 1-10, with 10 being
perfect, how are we doing?‖ His next statement
was classic, ―Let me get this straight, you are
calling me to say thank you for the business?
Now that‘s a phone call I like taking.‖ And with
that, the conversation went on for 15 minutes
until Karen hung up the phone and everyone
around her desk applauded. (FYI – The account
enjoyed her so much over the next couple of
months, they flew her to NY to discuss
additional opportunities to help the client.)
The lessons learned:
1. ―Can you help me?” It‘s the number one
sales line of all time. Use it early and often.
She used it with every one she spoke with.
2. Tell the truth. Be completely honest and
sincere. She was honest on the voice mail,
and honest when he returned the call back.
3. Never say ―Thank you‖ when they call back.
You‘re a peer! You‘re not beneath these
people. They need you just as bad as you
need them. Say, “Glad you called back”
as opposed to Thank you.
4. Quit sounding like everyone else. When
you are genuine, honest and sincere, it‘s
different from most sales people.
5. Have Courage: She virtually had no
product knowledge (they have 9000 products
and she‘d been there for only 90 days, and
hadn‘t even had our training class yet!) and
she had the courage to pick up the phone
and ask for help.
6. Positive attitude. She was scared, yes. But
her attitude wouldn‘t let that deter her. She
pressed forward and truly saw the fruits of
her effort..
7. She was smiling the whole time. (We put a
Northern Illinois University
mirror in front of her and told her ―When
you‘re smiling, they hear it on the other end.)
There are a variety of lessons in that one story,
but candor is the leading factor in this and so
many other scenarios. For example, when we
perform our Costigan LIVE Prospecting calls to
demonstrate the power of candor, we may say
something like, , ―Bill, we could go into a bunch
of features and benefits about how we‘ve helped
others, and to be honest, it sounds like everyone
else. I figured I‘d ask you a simple question.
How the heck does someone break into your
account?‖ Rarely do you hear a representative
use this type of technique. Why? It takes
courage! Jack Welch said in his book, Winning,
―Candor in business -- or in any kind of
organization -- is a rare and wondrous thing. It is
rare because, as we have discovered over the past
seven years, so few companies have it.
Wondrous because when they do, everything just
operates faster and better.‖
Google produces 2,600,000 results in 0.18
seconds when ‗Best traits of a professional sales
person‘ is searched There were lists that covered
endless qualities; Prompt, Professional,
Knowledgeable, Articulate, Creative,
Hardworking, Sincere, Organized, Resourceful,
Focused, Positive, Self Confident, ………all
good qualities, correct? Not one article said,
―Trustworthy.‖ NOT ONE!
Question: If the customer doesn‘t trust you, but
you perform those other qualities well, do you
think you have dramatically limited your ability
to get a sale? You‘re darn right. Trust is
everything!
We ask thousands of people a year, ―Have you
ever had anyone like you, like you a lot, and still
not buy from you?‖ Every hand goes up. Lesson
learned: They can like you and still not buy from
45
Journal of Selling & Major Account Management
you. Liking me is an add-on. Trusting me leads
to $$.
important than you giving your answer validating
your price.)
Think about going into surgery. Sure liking the
Doctor is important, but trusting him? That‘s
everything! No one wants to hear ―whoops‖ in
surgery. Nor do your customers when you are
delivering the solution.
Here‘s another one, ―We‘re already happy. Just
send me some information.‖ The truthful
answer: ―Mr. Customer, I completely believe
you are happy. I really do. But my biggest fear is
we may have an opportunity to make you
happier but you‘re telling me you are happy
because you want to get me off the phone. And
if I were you, I‘d probably say the same thing.
Can you help me?‖ See, truth again and it‘s
different. Different is good as long as it‘s
genuine and sincere.
The obvious next question, , ―How can you gain
trust quicker in a day and age where it‘s more
difficult to attain?‖ Stephen R. Covey‘s The Speed
of Trust, confirms just that. ―Americans only trust
34% of the people they know and in the last 4
decades, In Great Britain, it dropped from 60%
to 29%.‖
The answer? Tell the truth. Let me say that
again. TELL THE TRUTH! Look at the two
words. TRUST / TRUTH. There is a reason
why they are so close in spelling; You need to
have one to get the other.
For example, if a customer asks, ―Why are you
so expensive?‖ Of course we have usually been
programmed to tell the customer ―Because we‘re
the best in service and support and …….yadayada -yada.‖ And usually that answer won‘t get
you too far. How about telling the truth. ―Mr.
Customer, there are so many reasons why we
have priced our products accordingly, and when
it gets right down to it, in my heart, I don‘t
know. It‘s what the company has felt is a fair
price for this service. But can I ask you a
question? You must have asked me that for a
reason. Why do you ask? See, THAT‘S the truth.
Did your company consult with you on the
pricing of the products? Probably not. Did they
run it by you on why they need to price things
they way they do? Probably not. So, in actuality,
you really DON‘T know the REAL reason. (Tip:
When a customer asks you that question, find
out why they asked you. That‘s way more
Northern Illinois University
What about the prospect who wants to run you
around like crazy grabbing resources, pricing
quotes, webinars, etc.. And without any
commitment what-so-ever. (We call that being
stuck in Hope Alley) The truthful answer: ―Mr.
Customer, I have no problem gathering
resources to get the answers for you. My biggest
fear is we do all that and nothing happens. What
are your thoughts?‖ Again, the truth, and in this
case, if you get to the truth quicker, a ―no‖ early
on is way better than a ―no‖ 6 months from
now. We have heard answers from prospects
that would astonish you. From ―I don‘t know
what happens?‖ to ―We‘re not going to buy from
you, but we just wanted to get pricing so we can
keep our current vendor honest.‖ (Think about
that. Would you actually run around getting
prices to a prospect who already said they
wouldn‘t use you? A lot of reps do because they
get confused with effort and results. They think
just working hard will get them there. I‘d take
smart over working hard any day of the week!)
So in summary, will the truth help you win
business? Absolutely! Why?
Truth gets you trust.
Truth saves you time.
46
Journal of Selling & Major Account Management
Truth builds long lasting relationships
Truth makes you money.
Finally, the truth separates you from the
competition? Why? As Jack Welch says, ―it‘s
rare, very rare.‖ And being rare and different is a
good thing!
I just wish someone would have taught it to me
sooner.
John Costigan is the President and Founder of
John Costigan Companies. Prior to founding his
own company, John was a top sales
representative and account executive for leading
software and Internet companies. He created a
unique sales training program in which he
demonstrates the effectiveness of his proven
techniques by making impromptu calls to
prospects that are provided to him by the client.
John holds a Bachelor of Science Degree in
Education from Northern Illinois University.
For more information, please visit
www.JohnCostiganCompanies.com
Northern Illinois University