Laddering: A “How to Do It”
Manual – with a Note of Caution
By
Abrafi Saaka
Graduate Student-CIRP Program
Chris Sidon
Graduate Student-CIRP Program
Brian F. Blake, Ph.D.,
Director-CIRP Program
Methodology Series
February 2004
Cleveland State University
Brian F. Blake, Ph.D.
Senior Editor
Jillian M. Hughes
Co-Editor
Entire Series available: http://academic.csuohio.edu:8080/cbrsch/home.html
2
RESEARCH REPORTS IN CONSUMER BEHAVIOR
These analyses address issues of concern to marketing and advertising professionals and to academic
researchers investigating consumer behavior. The reports present original research and cutting edge
analyses conducted by faculty and graduate students in the Consumer-Industrial Research Program at
Cleveland State University.
Subscribers to the series include those in advertising agencies, market research organizations,
product manufacturing firms, health care institutions, financial institutions and other professional
settings, as well as in university marketing and consumer psychology programs. To ensure quality
and focus of the reports, only a handful of studies will be published each year.
“Professional” Series - Brief, bottom line oriented reports for those in marketing and advertising
positions. Included are both B2B and B2C issues.
“How To” Series - For marketers who deal with research vendors, as well as for professionals in
research positions. Data collection and analysis procedures.
“Behavioral Science” Series - Testing concepts of consumer behavior. Academically oriented.
3
AVAILABLE PUBLICATIONS:
Professional Series
Lyttle, B. & Weizenecker, M. Focus groups: A basic introduction, February, 2005.
Arab, F., Blake, B.F., & Neuendorf, K.A. Attracting Internet shoppers in the Iranian market,
February, 2003.
Liu, C., Blake, B.F., & Neuendorf , K.A. Internet shopping in Taiwan and U.S., February,
2003.
Jurik, R., Blake, B.F., & Neuendorf, K.A. Attracting Internet shoppers in the Austrian
market, January, 2003.
Blake, B.F., & Smith, L. Marketers, Get More Actionable Results for Your Research
Dollar!, October, 2002.
How To Series
Blake, B.F., Valdiserri, J., Neuendorf, K.A., & Nemeth, J. Validity of the SDS-17 measure
of social desirability in the American context, November, 2005.
Blake, B.F., Dostal, J., & Neuendorf, K.A. Identifying constellations of website features:
Documentation of a proposed methodology, February, 2005.
Saaka, A., Sidon , C., & Blake, B.F. Laddering: A “How to do it” manual – with a note of
caution, February, 2004.
Blake, B.F., Schulze, S., & Hughes, J.M. Perceptual mapping by multidimensional scaling:
A step by step primer, July, 2003.
Behavioral Science Series
Shamatta, C., Blake, B.F., Neuendorf, K.A, Dostal, J., & Guo, F. Comparing website
attribute preferences across nationalities: The case of China, Poland, and the USA, October,
2005.
Blake, B.F., Dostal, J., & Neuendorf, K.A. Website feature preference constellations:
Conceptualization and measurement, February, 2005.
Blake, B.F., Dostal, J., Neuendorf, K.A., Salamon, C., & Cambria, N.A. Attribute preference
nets: An approach to specifying desired characteristics of an innovation, February, 2005.
Blake, B.F., Neuendorf, K.A., Valdiserri, C.M., & Valdiserri, J. The Online Shopping
Profile in the cross-national context: The roles of innovativeness and perceived newness,
February, 2005.
Blake, B.F., & Neuendorf , K.A. Cross-national differences in website appeal: A framework
for assessment, July, 2003.
Blake, B.F., Neuendorf , K.A., & Valdiserri , C.M. Appealing to those most likely to shop
new websites, June, 2003.
Blake, B.F., Neuendorf , K.A., & Valdiserri , C.M. Innovativeness and variety of
information shopping, April, 2003.
4
RESEARCH REPORTS IN CONSUMER BEHAVIOR
EDITORIAL DIRECTOR: DR. BRIAN BLAKE
Dr. Brian Blake has a wide variety of academic and professional experiences.
His early career... academically, rising from Assistant Professor to tenured Professor at Purdue
University, his extensive published research spanned the realms of psychology (especially consumer,
social, and cross-cultural), marketing, regional science, sociology, community development, applied
economics, and even forestry. Professionally, he was a consultant to the U.S. State Department and to
the USDA, as well as to private firms.
Later on...on the professional front, he co-founded a marketing research firm, Tactical Decisions Group,
and turned it into a million dollar organization. After merging it with another firm to form Triad
Research Group, it was one of the largest market research organizations based in Ohio. His clients ranged
from large national firms (e.g., Merck and Co., Dupont, Land o’ Lakes) to locally based organizations
(e.g., MetroHealth System, American Greetings, Progressive Insurance, Liggett Stachower Advertising).
On the academic side, he moved to Cleveland State University and co-founded the Consumer-Industrial
Research Program (CIRP). Some of Cleveland’s best and brightest young marketing research
professionals are CIRP graduates.
In the last few years...academically, he is actively focusing upon establishing CIRP as a center for cutting
edge consumer research. Professionally, he is market research consultant for a variety of clients.
EDITOR (2003): JILLIAN HUGHES
Currently a CIRP graduate student, she graduated Magna Cum Laude from Mount Union College, where
she majored in Psychology, with a focus on Consumer Behavior, and minored in Sociology. Among her
many research interests; she focuses on the effects of Social Desirability Bias on Innovativeness Scales
and differences between paper and pencil and Internet survey responses. She had the honor of presenting
research concerning age differences in brand labeling at the Ohio Undergraduate Psychology Conference
in April of 2002 at Kenyon College. She also presented original research at the annual Interdisciplinary
Conference for the Behavioral Sciences hosted by Mount Union College in April, 2002 and 2003.
5
FOREWORD
For over 15 years, researchers have turned to laddering as one of their qualitative tools.
Although a watered down version of the technique is often employed, the systematic application
of the procedure as initially developed by Reynolds and Gutman can be highly informative. But
researchers and managers using the results of a laddering study must keep in mind the limitations
of the procedure. If they do not do so, the technique can “come back and bite them.”
This report presents a step by step approach to conducting a laddering study, and then
outlines potential problems that may be encountered.
The report is intended for professionals who do not have a long track record in the
technique. It is for researchers who are considering using the technique and for
managers/executives reviewing whether or not to commission a study based on this tool.
6
TABLE OF CONTENTS
Page
I.
Introduction
• An Example
• Basic Perspective
• Overview
7
7
8
9
II.
Laddering Methodology
• Eliciting Differences Among Brands
• The Interviewing Environment
• In-Depth Interviewing Techniques
11
11
12
13
III.
Analysis
• Step 1: Converting the Raw Interview Data into Ladders
• Step 2: Content Analysis
• Step 3: Generating the Implication Matrix
• Step 4: Constructing the Hierarchical Value Map
• Selecting Components of the HVM
• Determining Dominant Perceptual Pathways
16
16
17
19
22
22
24
IV.
Applications
27
V.
Problems
• Expense and Sample Size
• Subjectivity
• Lost Data
• Respondent Lack of Self-Insight
• Framework for Action
29
29
30
30
31
31
VI.
Conclusion
31
VII. Additional References
33
7
I. INTRODUCTION
An effective marketing strategy for a product or service requires an understanding of
consumers’ purchasing behaviors. If marketing researchers can identify the salient factors that
consumers consider in evaluating alternatives and the personally relevant reasons why those
factors are important to consumers, then successful marketing strategies can be developed to
appeal to those consumers.
Laddering is a technique that has been used, for example, to suggest communication
themes that “tug at a person’s gut.” It has been employed to identify emotional obstacles that
face a political candidate running for elected office. It has revealed product features that can
appeal to buyers at a very deep level. While some researchers have used the technique with great
success, other researchers have been known to find themselves overwhelmed by the subtleties of
the technique and to have felt it necessary to drop it.
An Example
Consider a consumer purchasing a car. That car has various attributes or features, e.g.
leather upholstery, a turbocharged V6 motor, two doors. Each attribute has particular
consequences, as that consumer sees it. The motor lets him out accelerate many other cars on the
highway (a positive consequence in his eyes), but also can cost more to operate (a negative
consequence). These consequences are meaningful in light of that person’s values. The
turbocharged V6 helps him achieve his value of feeling strong and powerful, a master of his own
fate. The higher cost of operation with this motor, though, interferes with his value of seeing
himself as a responsible husband and father, a selfless provider for his family.
8
Basic Perspective
Laddering is based on a means-end theory; it attempts to identify the product attributes
that elicit preference within a particular product class category. The attributes of products and
the consequences (both positive and negative) that are associated with usage are the “means”.
The “ends” are the desired outcomes expressed in terms of the consumer’s personal values.
These values are assumed to be reached through the consequences. (Reynolds & Gutman, 1988;
Myers, 1996). Consumers assumedly choose actions that produce desired and/or minimize
undesired consequences. Therefore, consumers (through their buying behavior) learn to
associate specific consequences with specific product attributes and this knowledge drives them
to choose products that have the relevant attributes to help them achieve their desired goals.
A major assumption of the means-end theory is that consumers’ product knowledge is
organized in a hierarchy with concrete thoughts linked to more abstract thoughts in a chain
progressing from a means to an end. Thus, the more concrete features or characteristics of a
product, the attributes (A), are connected to the more abstract ideas about psychological and
social consequences of the attributes (C). These psychosocial consequences or benefits (derived
from using the product) are in turn connected to the most abstract element of the three, the values
(V).
personal
values
consequences
or benefits
attributes
9
The end state is purported to be the real reason why a consumer uses the product i.e., how it
helps the individual achieve his/her desired goals.
Laddering is an in-depth interviewing and qualitative analysis methodology based on the
means-end theory. A laddering interview involves using a series of directed probes to uncover
the full range of attributes (A), consequences (C), and values (V) associated with a selected
product in a given product class.
Overview
The in-depth interviewing technique prompts the respondent to think critically about the
connections between the product’s attributes and his/her personal goals (the motive behind one’s
preference for that particular product) thereby revealing the A-C-Vs. In figure 1 below, the
diagram on the left is a ladder from a single respondent in a diet soft drink study; the diagram on
the right is a ladder from a luxury car study:
Figure 1: Ladder Examples
Ladder from respondent in diet soft drink study
Ladder from respondent in luxury car study
(V)
(V)
(C)
(C)
(A)
(A)
self esteem
↑
look good in clothes
↑
maintain my figure
↑
less calories
↑
not syrupy
(C)
(C)
(A)
(A)
self esteem
↑
prestige
↑
willing to pay a little more
↑
quality
↑
sleek look
The analysis of the laddering data across respondents begins with a summary of the major
elements by content-analysis procedures. A summary table is then constructed to reflect the
number of connections between the elements. From this table, the dominant connections are
10
then graphically represented in a tree diagram, termed the hierarchical value map (HVM). The
HVM is structural in nature and is a representation of the linkages across levels of abstraction
without reference to specific brands. The HVM is then used to recommend marketing and/or
communication strategies.
11
II
LADDERING METHODOLOGY
The question that should be answered before a laddering study is launched is: who are the
relevant people to be interviewed? For developing positioning strategies for products, the
relevant people may be customers (brand users) whose beliefs are critical to fully understanding
the competitive set of brands in the market. Since laddering involves detailed probing about
consumers’ brand beliefs, respondents must be knowledgeable about the specific brands in the
category. One way often found to be useful is to classify brand users by frequency of use and
relative loyalty.
As a general rule of thumb, it has been suggested that a minimum of 20 respondents
should be included in any single subgroup. Because each respondent typically provides about 3
ladders, and ladders usually have an average of 5 elements, ladders from 20 respondents can
produce a minimum of 225 data points (taking into account that one-fourth of respondents
generally do not go beyond one ladder). Hence, a relatively small sample size can provide
considerable detail about consumer choice and brand distinctions.
The laddering procedure involves three stages. 1) Elicitation of differences among
brands, 2) in-depth interviewing and 3) analysis of the data.
Eliciting Differences Among Brands
In the first stage of the laddering methodology, respondents are asked to make
comparisons between brands in a product class. There are three general methods of eliciting
such distinctions between products:
1) Triadic sorting involves presenting the respondent with three products and asking
them to explain how two of the products are the same and therefore different from the
third product. In the case of a diet drinks study, for example, some of the distinctions
could be:
12
•
cola versus uncola
•
coke versus pepsi
•
plastic bottle versus glass bottle
2) Preference differences is another useful device for eliciting distinctions. Here,
respondents are asked to rank their preferences and explain why one is more desirable
than the other. Queries about instances where less liked brands are used more
frequently than more liked brands enables the respondent to think about other
instances in which an attribute of a less preferred brand may appear attractive. This in
turn helps respondents to make meaningful distinctions among the products.
3) Differences by occasion presents the respondent with a personally meaningful context
within which to make the distinctions. Sometimes the distinctions elicited are such
obvious characteristics of the product that they do not permit advancement to more
personally meaningful areas from this starting point. Thus, helping the respondent
think of some frequent usage occasions provides another way for the respondent to
think about differences among the stimuli.
Once a satisfactory number of distinctions has been elicited for a given product (typically
10 to 15 attributes), the interviewer can select which ones will serve as the basis for building
ladders, or he/she could have the respondent rate the relative importance of each of the attributes
and select those with the highest ratings.
The Interviewing Environment
Naturally, the interviewer should try to build rapport with the respondent even before the
first stage (elicitation of distinctions) of the interview, and one should do one’s best to maintain
good rapport throughout the rest of the interview. Letting the respondents know in advance that
there are no right or wrong answers may go a long way in helping them become relaxed. Doing
13
so would also further reinforce the notion that the purpose of the interview is merely to
understand the ways in which the respondent sees this particular set of products. Also, because
of the personal nature of the later probing process, creating a slight sense of vulnerability on the
part of the interviewer may help the respondent realize that the interviewer is merely a trained
facilitator of this discovery process and not a judge or evaluator of the respondent’s ideas. This
can be achieved by telling the respondents that many of the questions may seem “somewhat
obvious and possibly even stupid,” thus, associating this predicament with the interviewing
process, and creating the impression that the interviewer is a mere facilitator following certain
guidelines.
By continually asking, “why is that important to you?” the interviewer maintains control
of the interview and creates the perception of being genuinely interested. Because it is critical
that the interviewer is perceived as an interested but neutral recorder of the information, the
interviewer’s reactions (both verbal and non verbal) should be as neutral as possible. It is
essential that an interviewer has a thorough understanding of the means-end theory so that they
are able to identify the A-C-Vs as they are brought forth by the respondents.
In Depth Interviewing Techniques
Two basic problems have been known to arise during the laddering interviews: 1) the
inability of respondents to articulate why a lower level (ie, more concrete) issue is important to
them; 2) the tendency of respondents to try to avoid answering probes that are too personal or
sensitive. In the first instance, asking the respondent to imagine what would happen if the
consequence expected is not delivered may help uncover the nonconscious reason—this is
known as “negative laddering”. In the case of stalling or avoidance due to sensitive issues, the
interviewer could try to move the conversation into a third person format, thereby creating a role
playing exercise, or use self disclosure (a delicate procedure) to overcome this obstacle. Self-
14
disclosure should only be attempted by an experienced interviewer, though, since it involves the
interviewer revealing a relevant personal fact about himself/herself to help the respondent feel
less inhibited in comparison. A frequently more feasible option is for the interviewer to make
note of the problem area and return to it when other relevant information is uncovered later in the
interview.
The following is an excerpt of a hypothetical interview done by Reynolds and Gutman
(1988) in a wine cooler study. It illustrates the technique of unblocking respondents when they
cannot advance beyond a certain level.
Interviewer: You said you prefer a cooler when you get home after work because of the fullbodied taste. What’s so good about a full-bodied taste after work?
Respondent:
I just like it. I worked hard and it feels good to drink something satisfying.
Interviewer:
Why is a satisfying drink important to you after work?
Respondent:
Because it is. I just enjoy it.
Interviewer:
What would you drink if you didn’t have a cooler available to you
Respondent:
Probably a light beer.
Interviewer:
What’s better about a wine cooler as opposed to a light beer when you get home
after work?
Respondent:
Well, if I start drinking beer, I have a hard time stopping. I just continue on into
the night. But with coolers I get filled up and it’s easy to stop. Plus, I tend to not eat as much
dinner.
Interviewer:
So why is continuing to drink into the evening something you don’t want to do?
Respondent:
Well if I keep drinking I generally fall asleep pretty early and I don’t get a chance
to talk to my wife after the kids go to bed. She works hard with the house and kids all day—and
it’s really important that I talk to her so we can keep our good relationship, our family life, going.
15
Typically, two or three ladders can be obtained from approximately three-fourths of the
respondents interviewed. This means that one-fourth of the respondents cannot go beyond one
ladder. A typical laddering interview takes about 60 to 75 minutes to complete—from eliciting
distinctions to completing an in depth interview that has solicited enough elements to complete a
ladder.
16
III.
ANALYSIS
The basic analysis steps can be summarized as follows:
Step 1: Reducing the raw interview data into the A, C, or Vs ladders. This process
involves a thorough review of the verbatim notes of video/audio tapes of the
interview.
Step 2: Content analysis of the element selected in step 1
Step 3: Summation of relations in content codes, resulting in an implication matrix of
all paired relationships.
Step 4: Construction of a diagram to meaningfully represent the main implications of
the study, the hierarchical value map (HVM).
Now for more detail let us use the wine cooler study with its data and tables from
Reynolds and Gutman (1988).
Step 1: Converting the raw interview data into ladders.
Below is a summary ladder from the interview from the wine cooler study (see Table 1
for the full set of ladders).
(V)
good family life
(C)
able to talk to my wife
(C)
don’t fall asleep
(C)
consume less alcohol
(A)
filled up/easy to stop
(A)
full-bodied taste/less alcohol
17
Table 1: Set of Ladders for Hypothetical Wine Cooler Study
LADDER 1
V
C
C
A
Sense of belonging (part of the group)
Socialize
Avoid getting drunk
Less alcohol
LADDER 2
V
C
C
C
A
A
LADDER 4
V
C
C
A
A
Like my coworkers (belonging)
Sophisticated image
More feminine
Bottle shape
Fancy label
Good family life
Able to talk to my wife
Don’t fall asleep
Consume less alcohol
Filled up/easy to stop
Full bodied taste/less alcohol
LADDER 3
V
C
C
C
C
A
LADDER 5
V
C
C
C
A
Self-esteem
Status symbol
Impress others
Quality
expensive
Responsibility to family
Waste money
Throw it away (don’t drink all of it)
Gets warm
To much to drink
Larger size
LADDER 6
V
C
C
A
A
Completing a chore (accomplishment)
Reward
Thirst quenching
Crisp
Carbonation
18
Step 2: Content Analysis
The first step in the analysis is to record the entire set of ladders on a separate form and
appropriately label each item that is an A, C, or V. After inspecting them for completeness, a set
of summary codes is developed that reflects all the elements (A-C-Vs) elicited. Table 2 provides
the summary content codes for the Reynolds and Gutman (1988) hypothetical Wine Cooler
study.
Table 2: Summary Content Codes for Wine Cooler Study
Values
(20)
(21)
(22)
(23)
Accomplishment
Family
Belonging
Self-esteem
Consequences
(8) Quality
(9) Filling
(10) Refreshing
(11) Consume less
(12) Thirst –quenching
(13) More feminine
(14) Avoid negatives
(15) Avoid waste
(16) Reward
(17) Sophisticated
(18) Impress others
(19) Socialize
Attributes
(1)
(2)
(3)
(4)
(5)
(6)
(7)
Carbonation
Crisp
Expensive
Label
Bottle shape
Less alcohol
Smaller
At this level of the analysis, the focus of interest is the relationship between the elements
and not the elements themselves. For example, “avoids negatives of alcohol” is a summary of
several more detailed elements (e.g. not too drunk, don’t say dumb things, etc.). Numbers are
then assigned to the codes, and these numbers are used to label each element in each ladder
producing the matrix in Table 3. The rows in the matrix represent an individual respondent’s
ladder (a respondent can have multiple ladders, thus multiple rows). Columns represent the
elements in each ladder.
19
Table 3: Raw Data from Hypothetical Wine Cooler Study
Respondents
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Content codes
1
1
1
3
4
2
1
3
1
1
3
2
1
1
1
3
1
2
1
1
2
3
1
1
3
3
3
3
3
3
3
3
4
4
5
10
10
10
6
17
10
12
8
12
10
8
10
12
12
10
16
10
10
10
10
10
20
10
10
6
6
8
18
16
8
8
17
13
13
13
12
16
12
20
20
12
16
20
16
16
20
12
16
16
12
20
12
12
12
16
12
0
12
16
16
16
18
23
23
18
17
18
17
17
17
16
0
16
0
0
16
20
0
18
0
0
16
20
18
16
0
16
16
16
0
16
0
16
0
20
18
20
0
0
22
18
23
18
18
23
20
0
16
0
0
18
0
0
23
0
0
18
0
23
20
0
20
18
18
0
18
0
20
0
0
23
0
0
0
0
23
0
23
22
0
0
0
23
0
0
22
0
0
0
0
0
22
0
0
0
0
0
22
23
0
22
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Step 3: Generate the Implication Matrix
The Implication Matrix (Table 4) is a square matrix that displays the number of times
each element leads to every other element in the same row (operationally defined as those
20
elements in a row which precedes other elements in the same row). Two types of relations are
represented, direct and indirect relations. Direct relations are those in which one element leads
to another without any intervening element. For example, in the wine cooler study, “less
alcohol” (in ladder 1) has a direct link with “avoid getting drunk”, and an indirect link with
“socialize”. The numbers in the matrix are expressed in fractional form with direct relations to
the left of the decimal and indirect relations to the right of the decimal. Thus, “carbonation”
leads to “thirst-quenching” 4 times directly and 6 times indirectly.
21
Table 4: Implication Matrix
8
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Carbonation
Crisp
Expensive
Label
Bottle shape
Less alcohol
Smaller
Quality
Filling
Refreshing
Consume Less
Thirst-quenching
More feminine
Avoid negative
Avoid waste
Reward
Sophisticated
Impress others
Socialize
Accomplishment
Family
Belonging
Self-esteem
1.00
3.00
12.00
2.00
1.00
9
10
11
10.00
4.00
12
13
14
4.06
.04
15
16
.01
.14
.04
2.04
2.02
2.02
1.00
1.00
1.00
5.00
.01
1.00
3.00
4.00
17
.03
1.01
2.04
1.03
.01
3.00
1.00
18
.04
.04
1.09
.02
.01
19
1.00
1.06
.01
1.01
.01
.01
4.03
4.04
5.10
.01
.06
7.00
.08
.02
1.00
5.00
11.00
4.00
1.00
.04
.02
3.02
1.03
.04
.04
14.00
22
23
.04
.01
4.00
5.00
21
.06
.04
10.00
20
.02
.06
4.01
2.00
8.00
1.00
1.00
3.00
.07
.05
.02
.02
.01
.01
.09
.03
.05
.03
.04
1.03
.04
.06
4.02
10.00
5.00
.05
.03
.03
.04
.02
.02
.04
.04
1.05
5.03
9.00
22
Step 4: Construct the Hierarchical Value Map
The Implication Matrix may be considered the “blueprint” for drawing up the hierarchical
value map (HVM). The HVM provides a meaningful way of representing subjective data and
acts as a tool to facilitate decision-making and problem solving. HVMs are created by
reconstructing “chains” from aggregate data. Chains refer to sequences of elements that emerge
from the aggregate Implication Matrix. Considerable ingenuity is needed for constructing the
HVM, because the only guideline is that one should try at all costs to avoid crossing lines. A
common approach in constructing an HVM is to set a “cut off,” i.e., a minimum number of links
that must be present before one considers that item. Multiple cutoffs (usually from 3 to 5) should
be used because they permit the researcher the freedom to choose the one that offers the most
information and the most stable set of relations.
Selecting Components of the HVM
The most efficient way to construct the HVM is to start in the first row for which there is
a value at or above the chosen arbitrary cutoff level. Since by this point the reader is familiar
with it, we will use the Reynolds and Gutman hypothetical wine cooler data yet again. Starting
with a cutoff of 4, the first noteworthy value is “carbonation-refreshing” (Row 1, Column 10)
relationship with a value of 10 signifying 10 direct relations and no indirect relations between the
two elements. Since carbonation is related to refreshing, the next row we look at is 10
“refreshing”. “Thirst-quenching” in column 12 appears to be the first significant value. Moving
to row 12 we find the next value in our growing chain is “reward” in column 16. From row 16
we find our next value to be “impress other” in column 18, and following that we find our final
value of “belonging” in column 22, producing a chain of 1-10-12-16-18-22.
After a chain is completed, it is advisable to go back to the beginning and check to see if
there are other important links in the rows of the matrix which were not picked up in the
completed chain. For example, inspecting the rows reveals that “accomplishment” (20) and
23
“self-esteem” (23) are linked to “carbonation”. A similar pattern is observed when links with
“Thirst-quenching”, “reward” and “impress others” are inspected, producing the chain like the
one below:
self-esteem (23)
Accomplishment (20)
Impress others (18)
Reward (16)
Thirst-quenching (12)
Refreshing (10)
Carbonation (1)
Moving to row 2, we find that the connections are almost identical to “carbonation” with one
exception; “Crisp” has a link with “Quality” (8). We therefore plot it next to “carbonation” and
then start a new chain with “quality”. Looking up quality reveals 12 direct relations with
“expensive” (3), significant links with “reward” (16) and with “sophisticated image” (17).
Scanning row 17 we find significant links with “Impress others”. This produces a 3-8-16-17-18
chain. “Fancy label” (4) and “Bottle shape” (5) both have 4 (2 direct and 2 indirect) relations
with “More Feminine” (13) and “More feminine” has seven direct links with “Sophisticated
Image” (17), yielding 4-5-13-17 chain. Row 6, “Less Alcohol” has links with “Avoid Negatives
of Alcohol”, (14), which links up with “Socialize” (19) which in turns links up with “Family
Life” (21), making a 6-14-19-21 chain. The next row, “Filling” (7) links up with “Consume
Less” which connects with the rest of the “Less Alcohol” link; producing a 7-11-14-19-21 chain.
The last attribute item “Smaller Size” (7) links up with “Avoid Waste” which has a weak link to
“Family Life” (21). Going back to inspect to see if any important links were left out, we find
24
that “Family Life” links up several times with “Belonging” (22), which also links up several
times with “Self-esteem” (23). Thus, it appears that it is only at the value level, “Belonging,”
that the right side of the map is connected to the elements of the left side.
The emergent HVM is shown in figure 2.
Determining Dominant Perceptual Pathways:
Having plotted all the elements, it is important to look at the elements in terms of the
number of direct and indirect relations they have with other elements. Table 5 presents the sums
of the direct and indirect relations for each element. “Belonging” (22), at the value level,
appears to have the most elements leading from it. It may be the core value in terms of
importance to the product class.
Table 5: Summary of Direct (XX) and Indirect
(YY) Relations for Each Element (XX.YY)
Code
To
From
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
15.35
07.23
17.30
06.14
05.10
06.60
04.05
19.23
05.12
16.26
05.09
14.22
06.09
10.05
02.00
20.11
15.05
20.00
08.00
00.00
00.00
00.00
00.00
00.00
00.00
00.00
00.00
00.00
00.00
00.00
19.00
00.00
16.00
05.00
15.00
06.04
10.05
04.01
25.33
15.15
21.40
08.11
14.25
09.12
20.56
15.37
25
The other three noteworthy elements with a high frequency of elements leading from
them are “Impress Others” (18), “Reward” (16), and “Quality” (8). In fact, the quality—
reward—impress others—belonging chain appears to have a high number of relations among its
respective elements. Other dominant pathways with considerable direct and indirect links are the
“carbonation-self-esteem” chain and “carbonation-accomplishment” chains. While these
dominant pathways provide genuine insight into what consumers consider as important
characteristics of a brand, all the pathways warrant our attention because the weaker pathways
might represent an opportunity for a campaign to strengthen this tie.
Figure 2: Hierarchical Value Map of Wine Cooler Study.
(Reynolds and Gutman)
Self-esteem
(23)
• Self worth
• Self image
Accomplishment
(20)
• Get the most
from life
Socialize (19)
• Easier to talk
• Open up
Impress Others (18)
• Successful image
Sophisticated Image
(17)
• Personal status
• How others view
Avoid negatives of
alcohol (14)
• Not too drunk
• Not too tired
More Feminine (13)
• Socially
Quality (8)
• Superior
product
• Product quality
Refreshing (10)
• Feel alert
• Alive
Carbonati
on (10)
Better Life (22)
• Better family
Belonging (22)
• Security
• Camaraderie
• Friendship
Reward (16)
• Satisfying
• Compensation
Thirst-quenching
(12)
• Relieves thirst
• Not too sour
Crisp (2)
26
Expensive
(3)
Avoid waste (15)
• Doesn’t get
Consumer less (11)
Label (4)
(fancy)
Bottle (5)
(shape)
Less
Alcohol (6)
Filling (9)
Smaller
size (7)
27
IV. APPLICATIONS
1. Segmentation
One form of market segmentation is to attempt to put people into groups that have
common needs and will respond in similar ways to marketing initiatives. Information from the
Hierarchical Value Map can be used to help segment populations. Groups of consumers who
have similar personal values in can be identified and segmented. The resulting segments may be
different from the ones that would result from using demographic or socio-economic variables or
even from the ones developed in a benefit segmentation. The value based segmentation, though,
may be particularly useful for specific applications. Advertising based on personal values may
be particularly effective in engendering higher buyer involvement in the brand and long term
loyalty.
2. Market Insight
Laddering can provide a deeper basis for understanding how people make choices
between competing products or brands, according to laddering enthusiast. It can identify
positive and negative associations as well as perceived strengths and weaknesses of the product
or brand in terms of consumer preferences. Unlike large scale surveys based on self-reported
measures, the interviewer is there to help the person uncover the “real” or “underlying” motives
behind one’s product or brand preference. Enthusiasts of laddering often feel that people are not
always aware enough of their own mental processes to do this by themselves. Suppose a person
were taking a pencil and paper survey, and that person’s answer to a question is, “I don’t know
why I like it.” There is no way to get past this answer. In an interview situation however, the
interviewer is trained to help people explore why they like or don’t like a product.
3. Advertising
Advertising can also be evaluated at the different levels of abstraction uncovered during
the laddering interview (that is attributes, consequences and values). After the initial laddering
28
interview is completed, the consumer is more aware of what attributes are more important to
them, what benefits they receive from buying the product and what value they are trying to
reinforce. At this point they are shown several ads and asked to rate them in terms of how well
each ad communicates the attributes, benefits and values of interest to them. Consumers are also
asked to comment on exactly what it is about the ad that does or does not facilitate the
communication process. An ad might communicate just the benefits but not the main attributes
or values or just the attributes but nothing else. A successful ad will communicate at all three
levels according to Reynolds and his colleagues. The laddering approach can help marketers
determine how well their ads do this.
4. Sales Support
It has also been suggested that laddering can be helpful in sales contacts. A variation of
the laddering technique could be used by salespeople to better understand how their customers
make their decisions. The salesperson can then adapt his or her sales approach accordingly.
5. Devise Promotional Strategies
There is interest in the laddering technique as a way to help develop marketing strategies.
The ideas that are uncovered can be used to help reach the consumers in the target areas. For
example, laddering can help position a product or brand by helping to determine it’s points of
difference with competitors. Points of difference are extremely important because, obviously,
they help determine whether or not a new product will be successful. Laddering taps into the
meanings and associations that can help to determine these differences.
Wansink (2003) gives an example of how laddering insights can be used to develop
specific marketing actions; the illustration relates to Nike shoes and clothing. Some constructs
that were found are:
•
Gives me a sense of belonging to a certain group
•
Makes me more a part of the soccer community
29
•
I wish I could live the lifestyle of a professional soccer player
Some marketing actions that were developed in response to these are:
•
Put premier soccer players’ numbers on shoes
•
Buy time on the score box shown on the T.V. during games
•
Offer Nike Soccer newsletters and promotions at the point of purchase
6. Advertising Strategy
Laddering can also be used to develop advertising strategies. The MECCAS Model
(Means-End Conceptualization of the Components of Advertising Model), can help do this. In
this approach the HVM is used to identify elements that would be used in advertising strategy:
•
Message elements: the specific features of the product to be pictured.
•
Consumer benefits: the positive consequences of using the product.
•
Executional framework: the overall tone and style of the ad.
•
Leverage point: the presentation of the message to that the linkage between one’s
values and the specific product attributes will be readily apparent to buyers.
•
Driving force: the dominant end value on which the advertising will focus
The MECCAS model is beyond the scope of the present report. For an extended review of this
issue consult the sources in the “Additional References” section of this paper.
V. PROBLEMS
1. Expense and Sample Size
One of the major problems in conducting laddering research (including gathering,
analyzing, and summarizing the data) is that the method is extremely time consuming and
expensive to conduct (Baker, 2002). In order to obtain the insight necessary for useable results
the data must be collected manually through in-person interviews, while the analysis and
summary of the data may typically require days of work by highly trained experts; all of which
can add considerable expense. For this reason it may be of limited value when the research
30
involves collecting data from large sample of respondents, e.g. when segmenting large
populations. On the other hand, it may be valuable when a fairly small sample is adequate.
Laddering has been effective when used to analyze a specific population with a small
representative sample. In this case the benefits can outweigh the costs. For example, Federal
Express used the laddering technique to interview secretaries of companies they would like to
have as clients. By finding out exactly what the secretaries wanted from a delivery service and
why they wanted it, Federal Express was able to greatly increase its market share.
2. Subjectivity
Another possible problem is the subjectivity involved in analyzing the data. The
decisions about what to include and how to categorize and summarize the data are highly
subjective. For example, it is not always easy to tell the difference between a consequence and a
value. The categorization procedure which uses common meanings to form the categories is also
subjective. Two people looking at the same group of attributes or consequences may not
generate the exact same groupings or names for the groupings. Different people are also likely
to consider different data to be relevant and/or important.
3. Lost Data
Loss of data is also a concern with the laddering technique. When the data are
transformed into a Hierarchical Value Map, it is inevitable that some data will be “lost.” “Lost”
in this context means that the responses are not included in the final analysis; they are simply
discarded or considered irrelevant. For example, Baker (2002) reports the loss of 55% of one
data set and 74% of another. She also reports that researchers who lose more than 77% of their
data frequently do not report the percentage in their research findings. This raises a question as
to how well the Hierarchical Value Map truly represents that data. A major reason why a
researcher may attempt to conduct a laddering study rather than to turn to other qualitative
31
techniques is the richness of the original data generated. Ironically, the richness of the data is
sacrificed through the loss of data.
4. Respondents Lack of Self-Insight
A potential problem with the interview process is that some respondents seem to
genuinely have very little idea as to the reasons they purchase an item. The job of the
interviewer then becomes one of trying to subtly evoke the desired responses. This effort to coax
a person into responding can make the interview seem more like a test to some people, which in
turn may engender an aversive reaction to the interview process.
5. Framework for Action
Finally, even though professionals are familiar with the laddering technique, they still
may experience problems going from a Hierarchical Value Map to execution and implementation
of a marketing strategy. There is no commonly agreed upon framework for translating the
strategies that are developed from the Hierarchical Value Map into creative, useable concepts for
advertisements or other marketing tools. This lack of an action framework makes it more
difficult for the creative staff to use the information from the study.
VI. CONCLUSIONS
In a global economy, consumers are inundated with numerous options from which to
choose. In this consumer-oriented market, consumers can decide which products to buy, when to
buy, in which packaging and of which quality. Thus, not paying attention to consumer
motivations could increase the risk of a mismatch between the characteristics of a product and
the aspects that are sought after by consumers. It is axiomatic that a marketer’s paying attention
to consumer motivations increases the likelihood for designing products that consumers would
desire improve the effectiveness of marketing strategies. Despite the problems in translating the
laddering data into a coherent HVM, the means-end theory, as implemented in the laddering
32
technique, offers a useful tool to explore the consumer’s values in terms of product choice
criteria.
On the other hand, the researcher and the end user of a laddering study must be alert to
the problems often encountered by this technique. These problems can turn out to be a fatal flaw
for a laddering project. CAVEAT EMPTOR!
33
VII. Additional References
Audenaert, A. & Steenkamp, J. E. M. (1997). Means-end chain theory and laddering in
agricultural marketing research. In B. Wierenga, A. van Tilburg, K. Grunnert, M.
Wedel and J.E. M. Steenkamp (Eds.), Agriculture marketing and consumer
behavior in a changing world (pp. 217-232). Amsterdam: Kluwer.
Bagozzi, R. P., Gurhan-Canli, Z., & Priester, J. R. (2002). The social psychology of
consumer behavior. Buckingham, UK: Open University Press.
Baker, S. (2002). Laddering: Making sense of meaning. In D. Partington (Eds.),
Essential skills for management research (pp. 226-253). London, UK: Thousand
Oaks Sage.
Gengler, C. E. & Reynolds, T. J. (2001). Consumer understanding and advertising
strategy: analysis and strategic translation of laddering data. In T.J. Reynolds
and J.C. Olson (Eds.), Understanding consumer decision making (pp. 119-144).
Mahwah, NJ: Lawrence Erlbaum Associates, Inc.
Myers, J.M. (1996). Segmentation and positioning for strategic marketing decisions,
(pp. 263-282). Chicago, IL: American Marketing Association.
Olson, J. C. & Reynolds, T. J. (2001). The means-end approach to understanding
consumer decision making. In T.J. Reynolds and J.C. Olson (Eds.),
Understanding consumer decision making (pp. 3-20). Mahwah, NJ: Lawrence
Erlbaum Associates, Inc
Reynolds, T.J., Dethloff, C., & Westberg, S.J. (2001). Advancements in laddering. In
T.J. Reynolds and J.C. Olson (Eds.), Understanding consumer decision making
(pp. 91-118). Mahwah, NJ: Lawrence Erlbaum Associates, Inc.
34
Reynolds, T.J. and Gutman, J. (1988). Laddering theory, method, analysis and
interpretation, Journal of Advertising Research, 28, 11-31.
Solomon, M. R.. (2002). Consumer behavior: buying having, and being. Upper Saddle
River, NJ: Prentice Hall.
Wansink, B. (2003). Using laddering to understand and leverage a brand’s equity.
Quantitative Market Research: An International Journal, 6 (2), 111-118.