Content uploaded by Luk Warlop
Author content
All content in this area was uploaded by Luk Warlop on Apr 16, 2022
Content may be subject to copyright.
Broad vs. Narrow brand positioning: Effects on competitive brand performance.
Lars E. Olsen, BI Norwegian Business School
Ioannis Pappas, BI Norwegian Business School
Bendik Samuelsen, BI Norwegian Business School
Luk Warlop, BI Norwegian Business School.
This is a post-print version of the paper, as it was accepted for publication by the European
Journal of Marketing in December 2021.
2
Broad vs. Narrow brand positioning: Effects on competitive brand performance
Abstract
Purpose –Brand managers can choose among two fundamentally different brand positioning
strategies. One is a broad brand strategy, focusing on many favorable brand associations. The
other is a narrow brand strategy, focusing on just a few and thus more mentally accessible
associations. Building on associative memory theory, the current article examines which of
these brand positioning strategies performs better under dynamic market conditions.
Design/methodology/approach – Three experiments test the effect of brand positioning
strategy on memory accessibility and competitive brand performance. Study 1 tests how
brand strategy (broad vs. narrow) affects defensive brand performance. Study 2 tests how
broad vs. narrow brands perform differently in a brand extension scenario (offensive brand
performance). Study 3 uses real brands and situation-based attributes as stimuli in a defensive
scenario.
Findings – The results show that a narrow brand positioning strategy leads to a competitive
advantage. Narrow brands with fewer and more accessible associations resist new
competitors more easily and have higher brand extension acceptance than do broad brands.
Research implications – The article shows how to use accessibility as evidence of
associative strength and test how accessibility influences competitive brand performance in a
controlled experimental context.
Practical implications – Brand managers would benefit from a narrow brand positioning
strategy in accordance with the USP school of thought used by many marketing practitioners.
Originality – The paper demonstrates that narrow brand positioning performs better than
broad brand positioning in dynamic markets, and to our knowledge is the first to do so.
Key words – brand positioning, associative memory, competitive performance
3
Broad vs. Narrow brand positioning: Effects on competitive brand performance
1. Introduction
Markets are dynamic. Established brands must defend themselves against new
challengers in their current categories. They grow by extending into new product categories,
and thus also challenge other brands in new markets. An important task of brand managers is
to position the brand in preparation for these challenges (Jewell and Saenger, 2014). They
must decide whether to opt for a narrow positioning with a core benefit or choose a broader
positioning of several benefits. In this research, we address this strategic choice considering a
dynamic environment, where the focal brand is facing new entrants in the category (defense)
or struggles for expanding in new categories (offense).
The classic approach to brand positioning has been the USP school of thought (unique
selling proposition) pioneered by Rosser Reeves (1961). According to USP, brand managers
should position their brands on one or a few unique benefits that enable the brand to stand out
from competitors’ brands (Frazer et al., 2002; Niu and Wang, 2016), and address the most
valuable consumers according to how these benefits match their needs (Haley, 1968). The
goal is to establish a few favorable brand associations in consumers’ memory (Roedder John
et al., 2006), and work to increase the strength of those associations. Instead of broadening
the set of favorable benefits, brand managers focus on repeating a limited set of benefits in
their positioning efforts. These few benefits must be relevant for consumers and preferably
not shared by other brands (Keller et al., 2002). The communication of a clear USP is
considered critical in advertising to differentiate a brand in the marketplace (Dens and De
Pelsmacker, 2010). Or as Haley (1968, p. 34) puts it: “New and old products alike should be
designed to fit exactly the need of some segment of the market […] Yet, many products
attempt to aim at two or more segments simultaneously. As a result, they are not able to
4
maximize their appeal to any segment of the market, and they run the risk of ending up with a
dangerously fuzzy brand image.”
However, previous literature has also shown that broad brand positioning,
simultaneously conveying several benefits, makes brands salient in consumers’ memories
across consumption situations, and therefore has a positive relationship with brand choice.
For example, Alba and Marmorstein (1987) found that in low-involvement situations,
consumers prefer a fictitious car brand with nine features to a car brand with three features.
More recently, research has found that the more benefits a brand is associated with, the
greater the likelihood that consumers would consider a brand for purchase, that having more
benefits associated with a brand influences brand loyalty, and that consumers more likely buy
brand extensions from brands with many benefits (Cobb-Walgren et al., 1995; Romaniuk,
2003; Romaniuk and Sharp, 2003; Romaniuk and Nenycz-Thiel, 2013). Heckler et al. (2014,
p. 176) argue in the same direction: “Proper product positioning often requires
communicating multiple brand benefits.” And last, Parker et al. (2018) argue that there is a
major trend in branding of having fewer brands that through a series of category extensions
have broadened their benefit propositions across disparate product categories.
Apparently then, brand managers can choose between two distinct brand positioning
strategies, each with its own support in the academic marketing field, either a narrow brand
positioning, using the classic USP approach, or a broad brand positioning, using multiple
benefits, as advocated by more recent research. This choice of strategies creates a dilemma
for brand managers. The narrow/USP strategy represents the established practice for many
brand managers and advertising agencies (Frazer et al., 2002). However, the benefits of this
strategy have never been explained theoretically. Broad brand positioning, on the other hand,
has been shown to provide many benefits (e.g., Romaniuk, 2003; Romaniuk and Sharp, 2003)
in more static market conditions, in which the target brands did not face any specific
5
competitive challenges. However, this strategy contradicts common marketing and
advertising practice and has yet to be tested in more dynamic market conditions.
The purpose of the current article is to test which of these brand positioning strategies,
broad vs. narrow brand positioning, performs better under dynamic market conditions, in
which new brands often are introduced into the market, thereby making the market supply
fluctuate. Our purpose is not to discuss consumer brand choice in a static market with a stable
set of brands. We aim to investigate how brand positioning strategies affect competitive
brand performance when established brands are being attacked by a new brand in their
current category (defensive brand performance), or when they are the attacker and extend the
brand into a new category (offensive brand performance). This is a subtle, but important
strategic difference. The current literature and theories in brand management do not offer any
guidance for when and why to choose between narrow or broad brand positioning in these
scenarios (e.g., Keller, 2012).
In this article we make the following contributions. First, we show that a narrow
brand strategy is a better strategic alternative for both defensive and offensive brand
performance. We show that a narrowly positioned brand outperforms a broad positioned
brand both in defending itself against a new entrant, and in extending to a new product
category in which its core association is relevant. Our second contribution is to show that
these findings indicate the associative strength (French and Smith, 2013) advantage of a
narrow strategy, making a target benefit more accessible in long term memory and thus
influencing attitudes in both defensive and offensive strategic scenarios. This contrasts with
previous findings, obtained under static market conditions, where brands with multiple
benefits are preferred (Alba and Marmorstein, 1987; Romaniuk 2003).
We organize the remainder of this article as follows. First, we ground our proposition
in associative memory theory, and show how competitive brand strategy, both defensive and
6
offensive, benefits from the accessibility of brand associations. Next, we explain why narrow
and broad strategies differ in the resulting accessibility of brand associations. This is the basis
for our proposition that a narrow brand strategy results in more accessible brand associations
(H1) and allows the brand to perform better when it is being attacked (H2) or when it is
trying to expand (H3). Then, we report the results of three studies. Study 1 shows how brand
strategy (broad vs. narrow) affects both brand associate accessibility and defensive brand
performance. Study 2 shows how broad vs. narrow brands perform differently in a brand
extension scenario (offensive brand performance). Study 3 is a test in a real-world setting,
using real brands and situation-based attributes as stimuli in a defensive scenario. We discuss
limitations and future research in the final section of the article.
2. Theoretical background
2.1.Associative networks and the fan effect
The brand management literature frequently utilizes the associative network model of
human memory as a conceptual foundation (see Keller, 2012; Roedder John et al., 2006). The
model postulates that consumers use brand names as retrieval cues about product attributes
and benefits stored in memory, and they use attributes and benefits as cues to retrieve brands
(Van Osselaer and Janiszewski, 2001). Information about the brand is stored in semantic
memory as a network of concept nodes (Roedder John et al., 2006) connected by associative
links varying in accessibility – or the ease and speed with which an association comes to
mind while processing an input (Bohner and Wänke, 2002, Higgins, 1996).
A consumer may have a lot of information about the brand stored in the associative
network, but it is not necessarily always accessible. Furthermore, consumers access some
associations from memory faster than others (Fazio et al., 1982). The more a consumer thinks
about an association in relation to a brand, the more accessible the association will be when
7
one encounters the brand (Smith and Queller, 2001). Association accessibility is an important
determinant of evaluative responses. If the most accessible brand association seems relevant
for the judgment, the consumer may stop further memory processing (Lynch et al., 1988),
and base their evaluation on the accessed information. Consequently, accessible associations
disproportionally influence consumer judgments and subsequently brand performance (Keller
2012).
Furthermore, we suggest that size of the associative network around a brand is an
important determinant of the accessibility of any of its associations, and hence of brand
performance. Sohn, Anderson et al. (2004) named the negative relationship between number
of associations and their accessibility the fan effect. The term fan refers to the number of
facts, or linked nodes, that “fan out” of a specific memory node. Research on the fan effect
has shown that as the number of linked nodes increases, the time to activate a particular
memory node also increases (see Sohn et al., 2004). Specifically, we expect that broader
brands have more fans (more associations) than do narrow brands. Thus, when the brand
pursues a broad brand strategy it requires more time to activate any association of a brand
than when it pursues a narrow brand strategy. Consequently, we predict that brands with
narrow and broad associative structures will have marked differences in brand association
strength, and that this difference in strength influences competitive brand performance.
Theoretically, associative strength refers to how closely brand associations are related
to the brand name in consumers’ memories. Empirically, strength is observed as accessibility
– the speed at which an association becomes activated from memory (Higgins, 1996;
Zdravkovic and Till, 2012). For example, Pullig et al. (2006) use the term “aspect
accessibility” to describe the likelihood that a given brand association comes to mind when
the brand name is activated. By increasing this likelihood, the speed at which people access
and recognize brand associations, relevant associations are strengthened. We formalize our
8
first hypothesis as:
H1: Brand associations formed as part of a narrow positioning strategy will be more
accessible from memory than brand associations formed as part of a broad strategy, as
evidenced by shorter reaction times to verify the truth of brand-benefit claims.
In the next section we outline how the differential effect of broad and narrow
positioning strategies on the brand associative network drives defensive and offensive brand
performance.
2.2. Competitive brand performance relies on brand association strength
In dynamic strategic scenarios, brand managers are generally concerned with either
defensive performance, which is the brand’s ability to protect profit margins, market shares,
and existing customer bases against competitive entry, or offensive performance, which is the
brand’s ability to grow the firm’s business by entering markets – e.g., brand extensions
(Hoeffler and Keller, 2003). We argue that because of the resulting greater accessibility of a
target association, the narrow positioning strategy will outperform a broad strategy on both
defensive and offensive performance.
2.2.1. Narrow positioning facilitates defensive brand performance
Defensive performance is associated with reducing customer exit and brand switching
(Keller, 2012). An important requirement for defensive performance is that consumers can
access relevant brand associations from memory (Lynch, 2006; Lynch et al., 1988). In other
words, defensive performance is the brand’s ability to defend its brand positioning in
consumers’ memories. When presented with a new competitor, brand B, consumers will
compare the new information with the established brand A’s corresponding brand
9
associations. If relevant associations are accessible, the memory search for more brand
associations is terminated and consumers will most likely evaluate brand A more favorably
than brand B.
Broad brand positioning focuses on increasing the size of consumers’ associative
networks by adding favorable and relevant brand benefits (Heckler et al., 2014). One
example is the clothing-retail brand H&M, which has positioned itself on multiple benefits:
fashion, reasonable price, modern/trends, models, etc. (Böger et al., 2017). Narrow brand
positioning focuses on brand concept consistency (Park et al., 1986). Brand managers work
to increase the strength of a few especially relevant brand associations, and instead of adding
new favorable benefits focus on repeating a limited set of benefits in their positioning efforts.
A typical example could be Volvo, which for many consumers is strongly associated with
“safety.” When a new brand with a strong safety message enters the market, the
corresponding association of Volvo is easily activated by members of the target group (e.g.,
families with babies/toddlers), helping the Volvo brand to fight off the attack. If Volvo would
be broadly positioned on multiple benefits (e.g. safety, comfort, and sleek design), the safety
association would be less easily activated, leaving the incumbent brand more vulnerable to
targeted attack.
The relationship between strong and accessible brand associations and defensive
performance has been shown in previous research. For example, brands with stronger
associations have been shown to better withstand competitive advertising (Kent and Allen,
1994), and consumers with stronger brand connections are more likely to reject negative
information about the brand (Lei et al., 2008). We predict that the narrow strategy, because it
results in a tighter and more easily activated brand-association link, will also activate the
relevant associations of the incumbent brand quickly when it is challenged by a new entrant.
10
We formalize our second hypothesis as:
H2: A narrow positioning strategy will lead to a higher ability to withstand a targeted attack
using the same benefit by a new entrant in the category than a broad strategy which
incorporates the benefit.
2.2.2. Narrow brand positioning facilitates offensive brand performance
Brands are generally viewed as vehicles of growth (Dimitriu et al., 2017; Roberts, 2005;
Samuelsen and Olsen, 2012). Growth includes acquiring new customers or increasing the
revenues obtained from established customers. In general, offensive performance is the
brand’s ability to increase firms’ revenues. One much researched offensive capability of
brands is brand extension into new product categories (Parker et al., 2018; Palmeira et al.,
2019). Research on brand extensions has shown that associative fit between the brand and the
new product category is an important determinant for offensive performance (i.e., acceptance
of brand extensions; Aaker and Keller, 1990; Michel and Donthu, 2014; Hem et al, 2014;
Völckner and Sattler, 2006). Hence, when consumers are exposed to a brand extension, an
important requirement for evaluating fit between the extension and the brand is access to
relevant brand associations in memory (Miniard et al., 2020). More accessible fit associations
will increase consumers’ attention levels on the information contained in the brand
associations and will benefit the extensions evaluations when compared to an incumbent
brand in the new category. Consequently, consumers who can more quickly activate a fit
association will tend to evaluate the brand extension more favorably, and the brand will thus
achieve higher levels of offensive performance. We formalize our third hypothesis as:
H3: A narrow positioning strategy will lead to a higher ability to extend the brand into a
11
different category where the same benefit is relevant than a broad strategy incorporating this
benefit.
Figure 1 shows our research model.
- Figure 1 about here
3. Empirical studies
We test our hypotheses in three studies. We test H1 in all three studies. H2, stating that a
narrow brand should perform better than a broad brand in the face of attack from a new
entrant (defensive performance), is tested in Study 1 and Study 3. Study 2 tests H3 that a
narrow brand would perform better than a broad brand does in an extension context
(offensive performance). To maximize internal validity, Study 1 and 2 are conducted with
fictitious brands, and carefully pretested brand associations. Study 3 extends the ecological
validity of our findings by using real brands and their situation-based associations (Romaniuk
2003), in a different product category.
3.1. Study 1
If our assumptions hold, we expect that narrow brand positioning will be associated
with stronger and more accessible brand associations, and consequently, higher levels of
defensive performance than broad brand positioning is. Specifically, Study 1 tests the
defensive performance of a brand that is pursuing either a narrow or a broad brand
positioning strategy when under attack from a new entrant.
3.1.1. Pretest
We chose to use the shampoo category in studies 1 and 2. The shampoo market is a
dynamic market characterized by many products and brands coming and going each year. In
addition, we chose the sun lotion category as the extension category in study 2.
12
In a pretest we tested the relevance of benefit association to use in the manipulations.
We recruited 32 undergraduate business students (65.6% females; median age 21) from the
same population that we would use in study 1. They rated a range of benefits used in the
shampoo category (presented as statements) one by one on a 7-point semantic differential
scale with scale anchors “unimportant” (1) and “very important” (7). We selected the three
middle benefits (the rated mean benefit, plus one above and one below the mean: M = 3.94,
Std. Dev. = 1.69) as stimuli. These benefits
1
were: Has good PH values (M = 4.03, Std. Dev.
= 1.93), More washes with less shampoo (M = 3.94, Std. Dev. = 1.69), and Protects against
dangerous UV rays (M = 3.90, Std. Dev. = 1.66).
3.1.2. Participants, procedures, and measurements
In study 1, sixty-three undergraduate business students served as participants (55.6%
males; median age 23). We recruited them in the school library, and they participated in
groups of up to ten persons in a computer lab. The students participated voluntarily and
received a gift certificate upon completion of the test session. First, after they read a short
introductory text, we exposed participants to the manipulation: an informative text about the
new shampoo brand ZELL in one of two versions. In the narrow condition, we listed only one
benefit: Has good PH values. In the broad condition, we listed three benefits: 1. Has good
PH values; 2. Protects against dangerous UV rays; and 3. More washes with less shampoo.
We instructed participants to read the information carefully, to make them
cognitively process and learn the information provided about ZELL. Second, we told
participants that a series of statements would appear, one by one, on the screen (e.g.,
Copenhagen is the capital of Denmark), and that they should press one of two keys to
indicate whether the statement was true or false. This filler task had two purposes. First, a
1
These benefits have been translated into English from the native language used in the original data
collection.
13
temporary delay to clear out working memory was important to reduce hypothesis guessing
and to control for differences in mere retrieval of ZELL’s benefits (see Nayakankuppam and
Mishra, 2005). Second, the response times measured in this filler task served to measure the
individual participant’s natural latencies. We followed general principles for collection
latency data (see Fazio, 1990), and instructed participants to work as quickly as possible
without sacrificing accuracy.
In total, eighteen statements appeared on the screen (nine true and nine false
statements) in randomized order. We repeated this block once, so that each participant
provided answers on thirty-six true-false statements. Third, we informed participants that a
series of statements regarding ZELL would appear on the screen, and that they should
indicate as quickly and accurately as possible, by pressing an assigned key, whether the
statement was true or false (Jewell and Unnava, 2003). In essence, this procedure was similar
to the filler task procedure. In total, eighteen statements appeared in randomized order, and
we repeated this procedure once, so that each participant provided answers on thirty-six
statements. Fourth, participants evaluated the target brand ZELL. Fifth, participants evaluated
another brand entering the shampoo market – SHIKA – “well known for its good PH values”.
Finally, participants stated their age and gender and were thanked, debriefed, paid, and
dismissed.
We predicted that responses to benefits that are more accessible in memory (stronger)
will be faster than responses to benefits that are less accessible (weaker) in memory
(Zdravkovic and Till, 2012). Thus, we measured accessibility using response-time latencies.
To deal with the shortcomings of response-time data, we conducted a three-step procedure,
recommended by Fazio (1990). First, we recoded all responses shorter than 300 milliseconds
and longer than 3000 milliseconds into 300 milliseconds and 3000 milliseconds, respectively.
Second, we averaged and subjected the response-time latencies (RTs) to a logarithmic
14
transformation to meet the normality assumption for each benefit. Third, in a procedure
advocated by Priester et al. (2004), we calculated the logarithm of the average RTs for all the
statements in the filler task for each individual participant – called the baseline response-time
latency (see also Jewell and Unnava, 2003). This measurement serves as the individual
participant’s chronic and natural RT. Then, we subtracted the baseline log-RT from the
logarithm of the average of each brand association to construct an adjusted response-time
latency index for each association. Consequently, this procedure yields an index of the
average RT for each benefit on a participant-by-participant basis.
We assessed the defensive performance of the incumbent brand ZELL by measuring
the acceptance of the attacker, using an established brand equity measure. We used the four
item Overall Brand Equity (OBE) scale developed by Yoo and Donthu (2001) to measure the
evaluation of the new entrant SHIKA. The OBE score for SHIKA was measured for each
participant by averaging the agreement-disagreement responses on the four 7-point OBE
items: “It makes sense to buy SHIKA instead of any other brand, even if they are the same”,
“Even if another brand has the same features as SHIKA, I would prefer to buy SHIKA”, “If
there is another brand as good as SHIKA, I prefer to buy SHIKA”, and “If another brand is
not different from SHIKA in any way, it seems smarter to purchase SHIKA”
Study 1’s design utilized two fictitious brands – ZELL and SHIKA. Therefore, it is
not likely that any of the participants have attitudes towards the fictitious brands, prior to
exposure, that could influence the results. Essentially, the brands appear to be almost similar
– both focusing on PH values. However, ZELL in the broad condition has three associations
whereas SHIKA has only one association. This difference in the amount of information could
potentially influence the participants’ attitudes towards both ZELL and SHIKA. For example,
in the ELM literature (Elaboration Likelihood Model) it has been shown that the number of
arguments could influence attitudes in low-effort processing contexts (i.e., peripheral route to
15
persuasion – see Petty and Cacioppo, 1986). A second potential problem is what Pandelaere
et al. (2009) called the “first exposure effect”. Although, ZELL and SHIKA are both
fictitious brands, the order of presentation could influence attitudes. More specifically,
Pandelaere et al. (2009) demonstrated that first encountered stimuli may be more liked than
later encountered stimuli. In other words, “the first exposure effect” could favorably benefit
ZELL. These findings build on works concerning the pioneering advantage of brands that
enter markets first (Carpenter and Nakamoto, 1989).
Therefore, we used pre-attitudes toward ZELL (i.e., attitudes measured before
presenting the new competitor SHIKA) as covariates in the analysis. We measured brand
attitudes using three 7-point scales with instructions and scale anchors: “To what extent did
you find the brand… bad – good, negative – positive, unfavorable – favorable” (see
Haugtvedt et al., 1992). Pre-attitudes towards ZELL showed no significant differences
between conditions (F (1, 61) = 3.38, p= .07, η2 = .054).
3.1.3. Results
A one-way MANOVA showed that the target benefit – Has good PH values – was
significantly more accessible by participants exposed to the narrow condition than by those
exposed to the broad condition (MN = 1479 ms vs. MB = 1741 ms, F (1, 61) = 7.30, p < .05)
(see Table I). Furthermore, as expected, the two benefits provided only in the broad condition
were significantly more accessible by participants exposed to the broad condition than by
participants exposed to the narrow condition – Protects against dangerous UV rays (MN =
1786 ms vs. MB = 1461 ms, F (1, 61) = 5.60, p < .05) and More washes with less shampoo
(MN = 1747 ms vs. MB = 1328 ms, F (1, 61) = 9.50, p < .05).
Recall that we hypothesized that an attacker brand would receive more favorable
evaluation when attacking a broad brand than when attacking a narrow brand. ZELL’s brand
attitude index was constructed as the average scale of three highly interrelated attitude items
16
(Cronbach’s alpha = .964). The OBE scale was constructed by forming an averaged index
from the four highly interrelated scale items (Cronbach’s alpha = .946). The OBE index
served as the dependent variable in a between-subjects ANCOVA. Details of the ANCOVA
analysis are presented in Table I.
The analysis showed that the new entrant SHIKA received a lower OBE score in the
narrow condition than in the broad condition (MN = 2.51 vs. MB = 3.16, F (1, 60) = 4.03, p <
.05, η2 = .056,) (see Table I). In other words, ZELL’s single benefit in the narrow condition
was more accessible in memory when we presented participants with SHIKA. Hence, ZELL
resisted attack from SHIKA better in the narrow condition than in the broad condition with
less accessible benefits.
- Table I about here
3.1.4. Discussion
The purpose of study 1 was to test whether a narrow brand positioning strategy
improves defensive brand performance. The results support this prediction. First, we found
that having fewer benefits increases the accessibility of a strategic target benefit. Participants
in the narrow strategy condition had significantly shorter RTs than did those in the broad
strategy condition. Furthermore, we found evidence a narrow positioning of the incumbent
brand also reduced the success of the attacker, hence showing a more effective defensive
performance. However, defensive brand performance is only one dimension of competitive
brand performance. Study 2 therefore focused on the second of these performance
dimensions – offensive performance.
17
3.2. Study 2
Brands can pursue growth in many ways. Among several alternatives, one important
growth strategy for brands is to extend the brand into new product categories and attack
established brands in these categories (Hem et al., 2014; Parker et al., 2018). Meyvis and
Janiszewski (2004) argue that the more accessible a relevant brand association x is, the more
likely it is that consumers evaluate the extension favorably. The association x might be
present in both the narrow and broad brand’s associative network, but due to the fan effect
association x is likely to be more accessible for the narrow than for the broad brand.
Therefore, we hypothesize that if a brand association is more accessible in consumers’
memories, it is more likely that an extension having this specific association will be favorably
evaluated, and thus, that a brand pursuing a narrow brand positioning strategy will show
better offensive performance than does a broad brand.
If our assumptions hold, we expect that a narrow brand positioning strategy will be
associated with higher levels of offensive performance than a broad brand positioning is.
Specifically, study 2 tests the offensive performance of brand A, pursuing either a narrow or
a broad brand positioning strategy, when extending the brand into a new product category.
3.2.1. Methodology
We recruited sixty-nine undergraduate business students (49.3% males; median age
22) from the same population as we used for the pretest and study 1. The first part of the
experiment was similar in all respects to the first part of the experiment in study 1. Then,
however, we told participants that ZELL had plans to extend the brand into a new product
category. Specifically, we exposed participants to information about the new ZELL sun lotion
– having optimal PH values. Hence, ZELL sun lotion based its fit with the original category,
shampoo, on the same benefit (i.e., PH values) in both conditions (see Aaker and Keller,
18
1990). Immediately after exposure, participants filled out the OBE scale for the extension,
stated their age and gender, and were thanked, debriefed, paid, and dismissed.
3.2.2. Results
We constructed RT indices following the same procedure as in study 1. A one-way
MANOVA on the adjusted RT index showed that the target association – Has good PH
values – was significantly more accessible by participants exposed to the narrow condition
than by those exposed to the broad condition (MN = 1526 ms vs. MB = 1716 ms, F (1, 67) =
5.06, p < .05; see Table II). These results replicated the results in study 1.
ZELL’s brand attitude index was constructed as the average scale of three highly
interrelated attitude items (Cronbach’s alpha = .897) and was used as a covariate in the
analysis. The dependent variable, OBE scale, was constructed by forming an averaged index
from the four highly interrelated scale items (Cronbach’s alpha = .871). This index served as
the dependent variable in a between-subjects ANCOVA.
The analysis showed that the brand extension (ZELL sun lotion) received a higher
OBE score in the narrow condition than in the broad condition (MN = 3.97 vs. MB = 3.22, F
(1, 66) = 6.18, p < .05, η2 = .070; see Table II). In other words, participants in the narrow
condition were more positive to the extension than were participants in the broad condition.
-Table II about here-
3.2.3. Discussion
The underlying rationale behind Study 2’s prediction was that stronger associations
are more accessible in memory. The analysis showed that in the narrow strategy condition,
the target benefit was more accessible than in the broad strategy condition. Furthermore, in
the narrow condition, the brand extension from shampoo to sun lotion, using the same target
19
association, received a higher brand equity score. These findings supported the prediction that
brands pursuing narrow brand positioning strategies enjoy higher levels of offensive
performance than do brands pursuing broad brand positioning strategies. Consequently, brand
extensions from narrow brands are more easily accepted.
3.3. Study 3
Romaniuk (2003) builds on Holden (1993) and identifies three specific benefit
categories: (1) product category benefits, (2) benefit attributes, and (3) situation-based
attributes. We obtained the results in both study 1 and study 2 in an experimental lab, using
fictitious brands and benefit attributes as types of associations. In the current research, we do
not differentiate between different categories of benefits. Instead we argue that the goal of
brand positioning is to influence consumers’ associative memories (Keller, 1993; Roedder
John et al., 2006), in which the speed of associative recall processes is influenced by brand
managers’ positioning strategies (Koll and von Wallpach, 2014). A logical extension of
studies 1 and 2 is therefore to test their predictions on real brands, and in addition use another
type of benefit – situation-based attributes – to further generalize the contribution of this
article. This is the purpose of study 3.
An important question is: Does a strong context match influence competitive brand
performance? Specifically, if an established brand has pursued a narrow brand positioning
strategy with a specific consumption situation strongly linked in its associative network, how
will consumers respond to a new entrant targeting the same consumption situation? A new
entrant would not have the same level of associative accessibility in the consumption
situation as the narrow brand would. Consequently, situational associative accessibility may
differ across brands in the same category, influencing defensive performance differently and
the likelihood of success for the new entrant.
20
3.3.1. Methodology
We conducted study 3 in two parts. First, we identified relevant brands, and found two
suitable brands in the Norwegian chocolate market. The first brand, Kvikk Lunsj, was
launched in 1938 and has for 80 years consistently been positioned as the best chocolate
brand to consume during recreational skiing and hiking activities. Kvikk Lunsj is part of
Norwegian heritage, and especially consumed during the Easter holiday. The second brand,
M, a chocolate-covered peanut candy, has been positioned in Norway as the “film chocolate”
since the 1980s. Many Norwegians associate M with film viewing, but most Norwegians still
consume a variety of other chocolates, popcorn, and candies in this situation. In addition,
activities like mountain hiking and skiing are much more specific and concrete consumption
contexts than is the film context. Consumers can enjoy films in the cinema, in the living
room, in airplanes, on electronic devices, etc., and therefore the film context is a much more
heterogeneous consumption context than is the hiking context. Consequently, Kvikk Lunsj
should have very accessible and strong associations connected to hiking. M, on the other
hand, should certainly be associated with film viewing, but most likely be weaker linked to
the film context than Kvikk Lunsj is to hiking. Hence, the first part of study 3 aimed to test
the accessibility of these context associations.
Second, in part two of study 3, we tested the brands’ defensive brand performance.
We introduced an attacker brand, a new fictitious chocolate brand, Bensdorp, in one of two
versions. Half the participants were told about ‘the new film chocolate BENSDORP’ and half
were told about ‘’the new hiking chocolate Bensdorp. Then participants provided responses
concerning their attitudes toward Bensdorp.
21
3.3.2. Participants, procedures, and measurements
Sixty-two undergraduate business students served as participants (46.8% males;
median age 22). They were recruited through advertising (i.e., posters) in the business school;
all participated voluntarily and received a gift certificate upon completing the test session.
Upon their arrival, we told participants that the purpose of the experiment was to conduct a
market survey on chocolate brands, hence disguising the experiment’s true purpose.
First, after participants read a short introduction text, we exposed participants to the
same true/false statements filler task used in studies 1 and 2 (Jewell and Unnava, 2003).
Second, we told participants that a series of statements regarding chocolate brands would
appear, one by one, on the screen, and that their task was to press one of two keys (true or
false) to indicate whether they agreed with the statements. In total, we randomly presented
them with fourteen statements. Four target statements (Kvikk Lunsj (M) is great when hiking;
M (Kvikk Lunsj) is great when viewing films) were presented among the statements appearing
on the screen. Specifically, these four statements tested the accessibility of M’s and Kvikk
Lunsj’s associations with the two consumption situations. As in the previous studies, we used
Fazio’s (1990) three-step methodology to adjust each statement’s RT to each individual’s
natural RT. Third, participants rated their attitudes toward a range of chocolate brands,
including Kvikk Lunsj and M, on three 7-point semantic differential scales similarly as done
in study 1 (Haugtvedt et al., 1992). Fourth, we exposed participants to a new chocolate brand
Bensdorp in one of two versions, 1. the new film chocolate or 2. the new hiking chocolate and
told them that this brand was about to enter the Norwegian chocolate market. We measured
attitudes toward the new entrant using the same attitude scales used for the other chocolate
brands. Finally, participants stated their age and gender, were thanked, debriefed, paid, and
dismissed.
22
3.3.3. Results
Table III shows the results of the paired samples tests. Kvikk Lunsj had significantly
shorter RTs in the hiking situation (MKvikkL_hiking = 1591 ms, Std. Dev. = 605 ms) than in the
film situation (MKvikkL_film = 1956 ms, Std. Dev. = 703 m), t (61) = 5.28, p < .001, η2 = .31).
Furthermore, Kvikk Lunsj also had significantly shorter RTs in the hiking situation than M
had in that situation (MKvikkL_hiking = 1591 ms, Std. Dev. = 605 ms vs. MM_hiking = 1785 ms,
Std. Dev. = 693 ms, t (61) = -2.645, p < .005, η2 = .10). These results support the prediction
that Kvikk Lunsj is strongly associated with the hiking consumption situation.
M, on the other hand, did not appear to belong more strongly to the film than to the
hiking situation (MM_hiking = 1785 ms, Std. Dev. = 693 ms vs. MM_film = 1822 ms, Std. Dev. =
595 ms, t (61) = 1.54, p = .129, η2 = .04). Furthermore, there was no significant difference in
RTs between Kvikk Lunsj and M in the film situation (MM_film = 1822 ms, Std. Dev. = 693 ms
vs. MKvikkL_film = 1956 ms, Std. Dev. = 703 ms, t (61) = .24, p = .814, η2 = .00).
- Table III about here
The three brand attitude items (i.e., attitudes toward BENDSDORP) were all highly
interrelated (Cronbach’s alpha = .977) and collapsed into an average index serving as a
dependent variable in a between-subjects ANOVA. The ANOVA produced the following
results: BENSDORP positioned as a film chocolate was significantly more favorably
evaluated than was Bensdorp positioned as a hiking chocolate (MBensdorp_film = 3.72 vs.
MBensdorp_hiking = 2.46, F (1, 60) = 8.05, p < .01, η2 = .03). Hence, Kvikk Lunsj was more able
to protect itself from competition and performed better than M, and these results supported
our predictions.
23
Differences in pre-attitudes toward Kvikk Lunsj and M are a potential alternative
explanation for the evaluation of the new competitor. To test this possibility, we included
attitudes toward Kvikk Lunsj and M as covariates in an ANCOVA (MM = 4.33, Std. Dev. =
1.90; MKvikkL = 5.31, Std. Dev. = 1.66). Two indexes were constructed as the average indexes
of the three highly interrelated attitude items of, respectively, Kvikk Lunsj and M
(Cronbach’s alpha (Kvikk Lunsj) = .977; Cronbach’s alpha (M) = .986). None of the
covariates had significant effects on the dependent variable (Brand Attitude Index (Kvikk
Lunsj), F (1, 58) = 1.30, p = .258; Brand Attitude Index (M), F (1, 58) = 1.01, p = .32),
hereby strengthening our conclusion that the better defensive performance of Kvikk Lunch as
compared to M, vis-à-vis the new entrant Bensdorp, is due to its narrow situational
positioning.
3.3.4. Discussion
Study 3 conceptually replicates study 1 in a real-world context and generalizes the
pattern of results shown in studies 1 and 2 with other types of benefits (i.e., situation-based
attributes). However, the difference of associative accessibility between conditions in study 3
is not because of differences in manipulations of a newly formed associative network for a
fictional brand, but rather because of actual differences in contextual associative strength
between the real-life brands Kvikk Lunsj and M. A priori attitude differences between the
two brands also do not explain the results.
4. General discussion
We introduced this paper with a typical brand manager’s dilemma. Should brand
managers go for a broad or a narrow positioning of their brands and what is the role of market
dynamics? Across three studies, we have investigated how different brand positioning
strategies influence competitive brand performance. We have shown how brands pursuing a
24
narrow brand positioning strategy outperform broad brands in two important dynamic market
situations: narrow brands are better protected against a new entrant (study 1 and study 3), and
enjoy a higher likelihood of success with a brand extension (study 2). Furthermore, we
suggested that these results can be explained by accessibility of relevant brand associations,
and we indeed found that consumers access associations faster for narrow brands than for
broad brands.
4.1.Theoretical implications
The branding literature (e.g., Böger et al., 2017; French and Smith, 2013; Keller,
2012) has highlighted that the strength, favorability, and uniqueness of associations are
important characteristics of a brand’s associative network. Yet, few efforts have been made to
test how increasing associative strength influences competitive brand performance. The first
theoretical contribution of the present research is therefore to show the importance of
association accessibility in more dynamic market conditions, where new brands are launched
and attack the incumbent brands in the market. In these situations, consumers evaluate the
favorability of the new entrant directly compared to an incumbent brand. Since narrow brand
positioning strategies make specific strategic benefits more accessible, the competitive brand
performance of such brands appears superior in these strategic scenarios.
The current results can be contrasted with previous research showing how having
many benefits (broad brand positioning) makes a brand more salient for consumers in usage
situations, and thus why such brands are more favorably evaluated or chosen (Alba and
Marmorstein, 1987; Romaniuk, 2003; Romaniuk and Sharp, 2003). However, these results
are obtained in static market conditions, that is, in conditions where the supply of brands is
stable, and researchers measure consumer choices among existing alternatives in the market.
The second major theoretical contribution is that we show how brand positioning strategies
influence competitive brand performance when the market changes, when new brands attack
25
the current market alternatives, and when a brand extends to another product category.
Acknowledging the difference in market conditions is essential to understanding why the
results of the current research diverge from earlier findings. We argue that association
accessibility is important in both static and dynamic market conditions, but that there are
important differences in the characteristics of this general accessibility. In static markets it is
important to have accessible brand associations jointly covering broad parts of the market,
usage situations, and consumer needs. The brand must be broadly relevant (e.g., Romaniuk,
2003; Romaniuk and Sharp, 2003). In dynamic market conditions, however, a general level
of high association accessibility is less important. Instead, high levels of accessibility of
specific associations, that might be attacked by a new entrant or can be used as a basis of fit
to extend the brand into new product categories, are important. As we have shown in the
current research, narrow brand positioning is a better strategy to achieve this goal.
4.2.Managerial implications
Brand managers are concerned with brand positioning. Two alternative strategies are
broad or narrow brand positioning strategies – focusing on few vs. many benefits – and thus
correspondingly, brand associations in consumers’ memories. Across three studies, we have
shown that managers concerned with protecting their brands against new competitors or
focusing on competing with their brands in new categories would benefit from a narrow
brand positioning strategy. In dynamic strategic scenarios, brands with few, but strong and
accessible associations would display better competitive brand performance. Hence, an
important contribution of the current article is that it provides evidence that associative
accessibility influences competitive brand performance and evidence that, contrary to many
well-founded assumptions and tendencies in business practice, it is better to focus on a
narrow brand strategy and a few strong associations than on a broad brand strategy and a rich
26
associative network. In essence, the current article provides empirical evidence favoring the
USP school of thought used by many marketing practitioners.
Another aspect of the current research that has managerial implications is that it
demonstrates the use of response-time latency (RT) in brand management. The branding
literature traditionally determines strength of brand associations by asking consumers to
indicate the associations’ subjective strength (see Roedder John et al., 2006), by the order of
mentioning (top-of-mind), or by the frequency of mentioning the associations (Böger et al.,
2017; Keller, 2012; Teichert and Schöntag, 2010). The RT measure, on the other hand, offers
an alternative methodology, in which participants are unaware that associative strength is
measured, and offers a procedure more consistent with the associative network model of
human memory. Specifically, the measurement technique taps directly into the actual time it
takes a consumer to connect the brand with a specific association in memory. As such, the RT
measurement should be both a superior and a more practical measurement of associative
accessibility. It can be used to evaluate the effectiveness of communication designed to
establish or strengthen such connections.
4.3 Future research
The current research offers several possibilities for future research. First, we
conducted the three studies in a controlled lab environment to maximize internal validity. In
the real world, a whole range of variables might moderate the results and can be included in
future research. One example of such a moderator is the diagnosticity or decision relevance
of an association, or the extent to which the association allows to differentiate between choice
options (Lynch et al., 1988). In the current research, we showed the effects by increasing the
accessibility of moderately diagnostic benefits, established by a pretest. We deliberately
opted for a moderately diagnostic benefit to avoid ceiling or floor effects, but the implication
is that we kept diagnosticity constant to focus on accessibility. To draw a more general
27
conclusion about the relative strengths of narrow positioning strategies future research should
examine whether increasing or reducing the decision relevance of chosen benefits (Miniard et
al., 2020) would have moderated these findings. We would speculate that the advantage of
the narrow positioning strategy increases monotonically with the diagnosticity of the
attributes on which the target brand and its competitor are evaluated. On the other hand, one
should also consider that in many situations the relevance and accessibility of a brand
association will covary. Relevant associations are more often used and will therefore also
tend to be more accessible. The current research does not address this issue.
Second, the current research conceptualizes two types of competitive brand
performance in terms of protection from new competitors and in terms of the ability to use
the brand in growth strategies (e.g., brand extensions). Previous research outlines many other
indicators of defensive and offensive performance (e.g., lower price elasticity, increased
customer loyalty, higher market shares, effectiveness in marketing communication, etc. – see
Keller (2012) for a list of potential indicators). Future research should investigate the effect
of narrow and broad brand positioning on other forms of competitive brand performance.
Third, our studies only considered a few product categories, all in the FMCG domain.
Further research should examine whether our conclusions are bounded by the properties of
these specific categories and could look at products versus services and at nondurables versus
durables. We did use a category that is relative homogeneous (shampoo) and one that is
relatively more heterogeneous (candy bars), but future research could systematically look at
category heterogeneity as a boundary condition. One speculation could be that in very
heterogeneous product categories, the narrow brand positioning might have less of an
advantage because the combination of more benefits creates a brand that is more easily
differentiated from competitors. In product categories with more homogenous products (e.g.,
shampoos) a narrow strategy would be better because the stronger and more accessible
28
associations increase the likelihood that consumers identify the brand. Future research should
therefore extend the current research and investigate boundaries on the superiority of narrow
brand positioning.
5. References
Aaker, D. A., and Keller, K. L. (1990), “Consumer Evaluations of Brand Extensions”,
Journal of Marketing, 54 (January), 27–41.
Alba, J. W., and Marmorstein, H. (1987), “The Effects of Frequency Knowledge on
Consumer Decision Making”, Journal of Consumer Research, 14 (June), 14–25.
Bohner, G., and Wänke, M. (2002), Attitudes and Attitude Change. Hove, UK: Psychology
Press.
Böger, D., Kottemann, P., Meissner, M., and Decker, R. (2017), “A mechanism for
aggregating association network data: An application to brand concept maps”, Journal
of Business Research, 79, 90–106.
Carpenter, G.S., and Nakamoto, K. (1989), “Consumer Preference Formation and Pioneering
Advantage”, Journal of Marketing Research, 26 (August), 285–298.
Cobb-Walgren, C. J., Ruble, C. A., and Donthu, N. (1995), “Brand equity, brand preference,
and purchase intent”, Journal of Advertising, 24 (3), 25–40.
Dens, N., and De Pelsmacker, P. (2010), “How Advertising Strategy Affects Brand and USP
Recall for New Brands and Extensions”, International Journal of Advertising, 29 (2),
165–194.
Dimitriu, R., Warlop, L., and Samuelsen, B. M. (2017), “Brand extension similarity can
backfire when you look for something specific”, European Journal of Marketing, 51
(5/6), 850–868.
29
Fazio, R. H. (1990), “ A Practical Guide to the Use of Response Latency in Social
Psychological Research” , Hendrick, C. and Clark, M.S. (Eds.), Review of Personality
and Social Psychology, 11, 74–97, Beverly Hills, CA: Sage.
Fazio, R.H., Chen, J-M., McDonel, E.C., and Sherman, S.J. (1982), “Attitude Accessibility,
Attitude-Behavior Consistency, and the Strength of the Object-Evaluation
Association”, Journal of Experimental Social Psychology, 18, 339–357.
Frazer, C. F., Sheehan, K. B., and Patti, C. H. (2002), “Advertising strategy and effective
advertising: Comparing the USA and Australia”, Journal of Marketing
Communications, 8(3), 149–164.
French, A., and Smith, G. (2013), “Measuring brand association strength: a consumer based
brand equity approach”, European Journal of Marketing, 47 (8), 1356–1367.
Haley, R. I. (1968), “Benefit Segmentation: A Decision-Oriented Research Tool”, Journal of
Marketing, 32 (3), 30–35.
Haugtvedt, C. P., Petty, R. E., and Cacioppo, J. T. (1992), “Need for Cognition and
Advertising: Understanding the Role of Personality Variables in Consumer
Behavior”, Journal of Consumer Psychology, 1 (3), 239–260.
Heckler, S. E., Keller, K. L., Houston, M. J., and Avery, J. (2014), “Building brand
knowledge structures: Elaboration and interference effects on the processing of
sequentially advertised brand benefit claims”, Journal of Marketing Communications,
20 (3), 176–196.
Hem, L. E., Iversen, N. M., and Olsen, L. E. (2014), “Category characteristics’ effects on
brand extension attitudes: A research note”, Journal of Business Research, 67, 1589–
1594.
30
Higgins, E. T. (1996), “Knowledge Activation: Accessibility, Applicability and Salience”,
Higgins, E.T. and Kruglanski, A.W. (Eds.), Social Psychology: Handbook of Basic
Principles, pp. 133–168. New York, NY: Guilford Press.
Hoeffler, S. and Keller, K. L. (2003), “The Marketing Advantages of Strong Brands”,
Journal of Brand Management, 10 (6), 421–445.
Holden, S. J. S. (1993), “Understanding brand awareness: let me give you a c(l)ue!”,
Advances in Consumer Research, 20, 383–388.
Jewell, R. D., and Saenger, C. (2014), “Associative and dissociative advertising strategies in
broadening brand positioning”, Journal of Business Research, 67, 1559–1566.
Jewell, R. D., and Unnava, H. R. (2003), “When competitive interference can be beneficial”,
Journal of Consumer Research, 30 (2), 283–291.
Keller, K. L. (2012), Strategic Brand Management: Building, Measuring, and Managing
Brand Equity 4th Edition, Upper Saddle River, NJ: Prentice Hall.
Keller, K. L., and Lehmann, D. R. (2003), “How Do Brands Create Value?”, Marketing
Management, May/June, 46–53.
Keller, K. L., Sternthal, B., and Tybout, A. M. (2002), “Three Questions You Need to Ask
About Your Brand”, Harvard Business Review, 80 (9), 80–81.
Kent, R.J. and Allen, C.T. (1994), “Competitive Interference Effects in Consumer Memory
for Advertising: The Role of Brand Familiarity”, Journal of Marketing, 58, 97–105.
Koll, O., and von Wallpach, S. (2014), “Intended brand associations: Do they really drive
consumer response?”, Journal of Business Research, 67, 1501–1507.
Krishnan, H. S. (1996), “Characteristics of Memory Associations: A Consumer-Based Brand
Equity Perspective”, International Journal of Research of Marketing, 13 (4), 389–
405.
31
Lei, J., Dawar, N., and Lemmink, J. (2008), “Negative Spillover in Brand Portfolios:
Exploring the Antecedents of Asymmetric Effects”, Journal of Marketing, 72 (3),
111–123.
Lynch, J. G. (2006), “Accessibility-Diagnosticity and the Multiple Pathway Anchoring and
Adjustment Model”, Journal of Consumer Research, 33 (1), 25–27.
Lynch, J. G., Marmorstein, H., and Weigold, M. F. (1988), “Choices from Sets Including
Remembered Brands: Use of Recalled Attributes and Prior Overall Evaluations”,
Journal of Consumer Research, 15,169–184.
Meyvis, T., and Janiszewski, C. (2004), “When Are Broader Brands Stronger Brands? An
Accessibility Perspective on the Success of Brand Extensions”, Journal of Consumer
Research, 31, 346–357.
Michel, G., and Donthu, N. (2014), “Why negative brand extensions evaluations do not
always negatively affect the brand: The role of central and peripheral brand
associations”, Journal of Business Research, 67, 2611–2619.
Miniard, P. W., Alvarez, C. M., and Mohammed, S. M. (2020). “Consumer acceptance of
brand extensions: Is parental fit preeminent?” Journal of Business Research, 118,
335-345.
Nayakankuppam, D., and Mishra, H. (2005), “The Endowment Effect: Rose-Tinted and
Dark-Tinted Glasses”, Journal of Consumer Research, 32 (December), 390–395.
Niu, Y, and Wang, C. L. (2016), “Revised unique selling proposition: Scale development,
validation, and application”, Journal of Promotion Management, 22 (6), 874–896.
Palmeira, P., Lei, J. and Valenzuela, A. (2019), “Impact of vertical line extensions on brand
attitudes and new extensions”, European Journal of Marketing, 53 (2), 299–319.
Pandelaere, M., Millet, K., and Van den Bergh, B. (2009), “First Is Best: First Exposure
Effects in Aesthetic Judgments”, Advances in Consumer Research, 36, 753.
32
Parker, J. R., Lehmann, D. R., Keller, K. L., and Schleicher, M. G. (2018), “Building a multi-
category brand: when should distant brand extensions be introduced?”, Journal of the
Academy of Marketing Science, 46 (2), 300–316.
Petty, R.E., and Cacioppo, J.T. (1986). Communication and Persuasion: Central and
Peripheral Routes to Attitude Change. New York, NY: Springer Verlag.
Priester, J. A., Nayakankuppam, D., Fleming, M. A., and Godek, J. (2004), “The A2sc2
Model: The Influence of Attitudes and Attitude Strength on Considerations and
Choice”, Journal of Consumer Research, 30 (March), 574–587.
Pullig, C., Simmons, C.J., and Netemeyer, R.G. (2006), “Brand Dilution: When Do New
Brands Hurt Existing Brands?”, Journal of Marketing, 70 (April), 52–66.
Reeves, R. (1961), Reality in Advertising. New York, NY: Knopf.
Roberts, J.H. (2005), “Defensive Marketing: How a Strong Incumbent Can Protect Its
Position”, Harvard Business Review, November, 150–157.
Roedder John, D., Loken, B., and Joiner, C. (1998), “The Negative Impact of Extensions:
Can Flagship Products Be Diluted?”, Journal of Marketing, 62 (January), 19–32.
Roedder John, D., Loken, B., Kim, K., and Monga, A. B. (2006), “Brand Concept Maps: A
Methodology for Identifying Brand Association Networks”, Journal of Marketing
Research, 43 (November), 549–563.
Romaniuk, J. (2003), “Brand Attributes – “distribution outlets” in the mind”, Journal of
Marketing Communications, 9, 79–92.
Romaniuk, J., and Sharp, B. (2003), “Measuring Brand Perceptions: Testing Quantity and
Quality”, Journal of Targeting, Measurement and Analysis of Marketing, 11 (3), 218–
229.
Romaniuk, J., and Nenycz-Thiel, M. (2013), “Behavioral brand loyalty and consumer brand
associations”, Journal of Business Research, 66, 67–72.
33
Samuelsen, B. M., and Olsen, L. E. (2012), “The Attitudinal Response to Alternative Brand
Growth Strategies”, European Journal of Marketing, 46 (1/2), 177–191.
Smith, E. R., and Queller, S. (2001), “Mental Representations”, Tesser A. and Schwarz, N.
(Eds.), Blackwell Handbook of Social Psychology: Intraindividual Processes, 111–
133. Hoboken, NJ: Wiley-Blackwell.
Sohn, M. H., Anderson, J. R., Reder, L. M., and Goode, A. (2004). Differential fan effect and
attentional focus. Psychonomic Bulletin & Review, 11(4), 729-734.
Teichert, T. A., and Schöntag, K. (2010), “Exploring Consumer Knowledge Structures Using
Associative Network Analysis”, Psychology and Marketing, 27 (4), 369–398.
Van Osselaer, S.M.J., and Janiszewski, C. (2001), ”Two Ways of Learning Brand
Associations”, Journal of Consumer Research, 28, 202–223.
Völckner, F., and Sattler, H. (2006), “Drivers of Brand Extension Success”, Journal of
Marketing, 70, 18–34.
Yoo, B., and Donthu, N. (2001), “Developing and Validating a Multidimensional Consumer-
Based Brand Equity Scale”, Journal of Business Research, 52, 1–14.
Zdravkovic, S., and Till, B. D. (2012), “Enhancing brand image via sponsorship – Strength of
association effects”, International Journal of Advertising, 31 (1), 113–132.
Narrow
(one benefit)
Broad
(three benefits)
OBE Index (SHIKA)
N
32
31
Mean
2.51a
3.16a
Std. Error
.225
.228
Response times in
milliseconds
(mean values)
“Has good PH values”
1479 (M)
465.01 (SD)
1741 (M)
546.41 (SD)
“Protects against
dangerous UV rays”
1786 (M)
566.81 (SD)
1461 (M)
570.98 (SD)
“More washes with
less shampoo”
1747 (M)
583.64 (SD)
1328 (M)
391.44 (SD)
Table I
Study 1 – ANCOVA and descriptive statistics
a. Covariates appearing in the model are evaluated at the following value:
Attitude Index (ZELL) = 4.1746.
2
Narrow
(one benefit)
Broad
(three benefits)
OBE Index (Zell Extension)
N
36
33
Mean
3.97a
3.22a
Std. Error
.204
.214
Response times in
milliseconds
(mean values)
“Has good PH values”
1526 (M)
526.39 (SD)
1716
525.20 (SD)
“Protects against
dangerous UV rays”
1933 (M)
631.00 (SD)
1578 (M)
433.84 (SD)
“More washes with
less shampoo”
1910 (M)
556.17 (SD)
1625 (M)
499.00 (SD)
Table II
Study 2 – ANCOVA and descriptive statistics
a. Covariates appearing in the model are evaluated at the following value:
Attitude Index (ZELL) = 4.2415.
3
Paired Samples Test. Dependent variable: Adjusted RT
Mean
Std.
Deviation
t
df
Sig.
(2-tailed)
Eta squared
(η2)
Pair 1
KvikkL_Hiking – M_Film
Pair 2
KvikkL_Hiking – M_Hiking
Pair 3
KvikkL_Hiking – KvikkL_Film
Pair 4
M_Hiking – KvikkL_Film
Pair 5
M_Hiking – M_Film
Pair 6
M_Film – KvikkL_Film
-.182
-.103
.194
.091
.080
.012
.372
.306
.289
.247
.408
.386
-3.855
-2.645
5.282
2.909
1.539
.237
61
61
61
61
61
61
.000
.010
.000
.005
.129
.814
.196
.103
.314
.122
.037
.001
Table III
Study 3 – Results of the paired samples t-tests
4
Figure 1
The research model
