Content uploaded by Ruth Pogacar
Author content
All content in this area was uploaded by Ruth Pogacar on Aug 04, 2016
Content may be subject to copyright.
Sounds good: Phonetic sound patterns in top brand names
Ruth Pogacar &Emily Plant &Laura Felton Rosulek &
Michal Kouril
Published online: 1 April 2014
#Springer Science+Business Media New York 2014
Abstract Recent research has demonstrated that brand name sounds can influence
consumer behavior. Sound symbolism, the link between sound and meaning, can
convey product information, enhance affinity, and increase purchase intentions. This
study examines sound patterns of Interbrand top 100 brand names, including three
previously unexamined sound categories. Results show that top brand names have
different sound patterns than general brand names. The pattern of differences suggests
that sound symbolism may be one factor contributing to brand performance. Sounds
more frequent among top brand names have potentially brand enhancing properties,
while sounds less frequent may have the opposite effect. These findings should inform
best naming practices and strategies.
Keywords Brand names .Sound symbolism .Linguistics .Brand performance .
Phonetics
The competition for prime brand name real estate is increasingly fierce as
evidenced by the increase in trademark applications over the past decade (US
Patent and Trademark 2012). Mainstream faith in the importance of brand
Mark Lett (2015) 26:549–563
DOI 10.1007/s11002-014-9288-z
R. Pogacar (*)
Lindner College of Business, University of Cincinnati, 925 Campus Green Dr., Cincinnati, OH 45221,
USA
e-mail: ruth.pogacar@gmail.com
E. Plant
Department of Management and Marketing, University of Montana, 32 Campus Dr., Missoula, MT
59812, USA
e-mail: Emily.Plant@umontana.edu
L. F. Rosulek
Department of Linguistics, University of Montana, 32 Campus Dr., Missoula, MT 59812, USA
e-mail: Laura.FeltonRosulek@umontana.edu
M. Kouril
Department of Biomedical Informatics, Cincinnati Children’s Hospital Medical Center/University of
Cincinnati, 3333 Burnet Ave., Cincinnati, OH 45229, USA
e-mail: Michal.Kouril@cchmc.org
names with positive linguistic attributes is indicated by popular press reports
that “Prozac”was designed with a “z”to convey efficacy (Erlich 1995)andthe
“k”in Kraft imparts positive brand information (Klink 2009). Yet, securing a
desirable name is increasingly difficult due to the decreasing number of known
words available for trademark (Klink 2000).
Although boutique naming firms such as Lexicon, responsible for the $15 billion
dollar name Blackberry, have sprouted in response to increasing demand (Heath
and Heath 2011), the lack of scientific evidence for their methods begs further
examination of what brand name characteristics, if any, effectively contribute to
brand performance. Empirical evidence for the efficacy of naming practices has
never been more important as standard naming fees have soared to as much as
$50,000 in product categories such as automobiles, at a per-letter price of
5,000–$10,000 (Hoffman 2008).
This research documents the phoneme (i.e., sound) level linguistic attributes
of the top brand names using a new method and dataset. We use the data
derived from Interbrand’s list of top brand names (Interbrand 2011), identify the
frequencies of key sounds, and compare them with the corresponding frequen-
cies in a benchmark dataset of general brand names. We propose that sound
symbolism, or the link between sound and meaning, is one mechanism by
which positive linguistic attributes enhance brand names.
1Researchcontext
Brand names may increase awareness and generate favorable product perceptions
(Aaker and Keller 1990), because they convey information and meaning (Allen et al.
2008). Conversely, product failures such as the Ford Edsel have been partly attributed
to misnaming (Hartley 1992).
Research has demonstrated that sound symbolism can convey product infor-
mation independent of marketing communications (Klink 2000; Yorkston and
Menon 2004), thereby enhancing brand names. In the marketing domain, certain
sounds have been found to influence consumers’product evaluations (Lowrey
and Shrum 2007;YorkstonandMenon2004), price perceptions (Coulter and
Coulter 2010), and choices (Argo et al. 2010). Furthermore, consumers prefer
product names that employ sound symbolism to convey product information
(Klink 2001). Specifically, brand names are preferred when their sounds convey
preferred product attributes (Klink 2009;LowreyandShrum2007). For instance,
the experimental ice cream name “Frosh”was preferred over “Frish”because
the “o”sound is associated with the desirable attributes smoothness and creaminess
(Yorkston and Menon 2004).
Schloss (1981) appears to be the first academic to analyze the linguistic
attributes of real brand names, finding that the top brands from 1975–1979
began more frequently with the letters “a,”“b,”“c,”“k,”“m,”“p,”and “s”
than did words in the dictionary. Vanden Bergh (1990) replicated Schloss’s
work and suggested that these letters are associated with brand success. How-
ever, it is impossible to draw conclusions about brand name sound patterns or
sound-symbolic effects from either study, because both examined letters rather
550 Mark Lett (2015) 26:549–563
than sounds. Many letters represent more than one sound (e.g., the “c”in
“Cisco”and “Cartier”) resulting in no direct one-to-one correspondence. Our
research contributes a linguistic analysis of sound patterns in brand names.
We take a distinct approach from both Schloss, who examined a list of brand names
based solely on advertising expenditures, and Vanden Bergh, who used an all-inclusive
trade name index. We examine Interbrand’s“Top 100”list, which is based on a
composite of brand performance factors (Interbrand 2011). Sounds that appear more
frequently among top brand names (compared with a sample of general brand names)
have potentially brand enhancing properties, while sounds appearing less frequently
may have the opposite effect (or may not have brand enhancing properties). The
findings from this research have potential to inform the best naming practices and
future research.
2 Linguistic analysis
2.1 Sound symbolism
Although the mapping of sound to object is assumed to be arbitrary in the
prevailing linguistic paradigm (Fromkin et al. 2010), mounting evidence in
linguistics, anthropology, psychology, and marketing suggests that sound symbolism
provides a mechanism for communicating meaning. Plato (1985) subscribed to
the theory that the “r”sound in words such as “river”and “road”conveys
movement. Another salient example is the Bouba/Kiki effect. When asked,
“which shape in Fig. 1is called ‘Bouba’and which is called ‘Kiki?’” 95 %
of the subjects replied that Kiki is the angular shape on the left and Bouba is
the round shape on the right. The effect was replicated with adults and
children, across languages including English, Swahili, and Bantu, demonstrating
that some sounds convey the meaning “round,”while other sounds convey the
meaning “angular”(Maurer et al. 2006).
Most sound symbolism research has focused on three primary categories:
vowels, fricatives, and plosives. In the following section, we summarize these
findings and introduce three new categories of sounds that have not been
previously examined in the marketing literature: nasals, affricates, and approximates.
By including these categories, we can examine all the sounds in the English
language for a more comprehensive survey of sound patterns in top brand
names.
2.2 Vowels
A robust literature has developed around vowel sound symbolism. Lowrey and Shrum
(2007) found that certain vowel sounds are perceived as generally positive or negative
by English speakers, regardless of the product category. For instance, the “o”sound in
“posh”was rated positively by the study participants, whereas the sound com-
bination “yoo”in “puke”was rated as generally negative. Similarly, the “u”
sound in “ugh”has historically been considered a negative sound (Jespersen
1922).
Mark Lett (2015) 26:549–563 551
More generally, front vowels, produced in the front of the mouth, like the “i”in
“Kiki,”convey the meanings small, feminine, fast, light, and angular. Back
vowels, produced in the back of the mouth, like the “ou”in “Bouba,”convey
the meanings large, masculine, slow, dark, and round (Hinton et al. 1994;Klink
2003). Consumers prefer brand names containing vowels associated with desirable
product category attributes. For instance, in the context of automobiles, partici-
pants preferred names emphasizing back vowels (e.g., “bromley”)forSUVs,and
names emphasizing front vowels (e.g., “brimley”) for two-seater convertibles
(Lowrey and Shrum 2007). This is because the back vowels convey largeness
while front vowels convey smallness.
2.3 Fricatives
Fricatives, such as the “s”and “z”sounds, are produced by restricting, but not
stopping, airflow through the mouth. There is reason to believe that some
fricatives have sound symbolic properties similar to vowels. However, research
is split on the positive versus negative connotations of fricatives. Fricatives
have been negatively associated with electability of political candidates, when
present in candidates’names (Smith 1998). Two studies also found that the
letter “s”appeared in disproportionately fewer brand names than would be
expected based on dictionary frequency (Schloss 1981;VandenBergh1990).
However, in an atlas of 360 words rated by participants on a scale from good
to bad, the “s”sound was significantly more common among words rated as
good (Jenkins et al. 1958).
2.4 Plosives
Plosives are a class of sounds formed by a stoppage of air in the mouth. In
English, these includes “p,”“b,”“t,”“d,”“k,”and “g.”When a plosive is the
first sound in a word (i.e., word-initial, as in “Kraft”), brand name memory
(Lowrey et al. 2003), recognition, and recall (Cortese 1998) increase. This may
explain why word-initial plosives were the second most common attribute of
brand names identified by Vanden Bergh et al. (1987). Plosives also represent
three of the four most frequent word-initial sounds in positive English nicknames
(De Klerk and Bosch 1997).
Fig. 1 The Bouba/Kiki Effect. Source Köhler (1929)
552 Mark Lett (2015) 26:549–563
However, research is split into positive or negative attributes of plosives. Johnson
et al. (1964) found that plosives were more common among words rated unpleasant
than among words rated pleasant. Further, plosives were found to be significantly
overrepresented among “bad”words by Jenkins et al. (1958).
Sound symbolic characteristics of plosives have been documented as well. For
instance, in almost half of commonly used English words, plosive “g”followed by
“l”imparts visual connotations (e.g., glance, glare, gleam, and glitter) (Bolinger
1950).
2.5 Nasals
Nasal sounds, which include “m”,“n,”and the last sound in the word “Boeing,”
are produced by channeling airflow through the nasal cavity. According to
Johnson et al. (1964), nasals are more common among words rated pleasant
by the listeners. The “m”sound, in particular, was overrepresented among
words rated as good (Jenkins et al. 1958) and is the fourth most frequent
word-initial sound in positive English nicknames (De Klerk and Bosch 1997).
Furthermore, the letter “m”was the second most commonly occurring first letter of
brand names from 1975–1979 (Schloss 1981).
2.6 Affricates
Affricates have been less well studied than the previously described sound
categories. Affricates are a combination of two sound categories: stops and
fricatives. They include the “ch”sound in “Merrill Lynch”(“t”+“sh”)andthe
“dj”sound in “Gillette”(“d”+“zh”). Johnson et al. (1964) asserted that
affricates were more common among words rated unpleasant by the listeners;
however, they conflated affricates with stops in their analysis (for instance, “ch”
was placed in the same category with “t”).
2.7 Approximates
Approximates fall somewhere between sound categories—they are neither vowels nor
fricatives. These include “w”,“l”,“r”,and“y.”The literature does not suggest either
positive or negative attributes; one of the few studies to examine approximates pro-
poses that “w”may intensify the effect of the preceding fricatives. For example,
“thwack”represents a greater blow than simply “wack”(Hinton et al. 1994). This
sound category has not previously been examined in the context of marketing.
3 Predictions
Past research suggests that sound symbolism can influence consumer behavior. How-
ever, to the authors’knowledge, no study has previously examined sound patterns in
top brand names. If some sounds are more or less frequent in top brand names relative
to the benchmark, this may point to sound symbolic properties that contribute to
brand performance.
Mark Lett (2015) 26:549–563 553
Sounds significantly more common among top brand names may also reflect
preferences at the management level. For instance, if name inventors believe the “z”
in Prozac conveys efficacy due to popular press reports, then “z”may be dispropor-
tionately selected, regardless of consumer preferences. The mechanism by which
certain sounds become more common among top brand names may be through
management decisions, consumer behavior responses, or most likely, a combination.
Both consumers and managers are likely to respond similarly to names with positive or
negative sound symbolic properties.
3.1 Vowels
Based on Lowrey and Shrum (2007), we would expect to find the “o”sound in “posh”
overrepresented among top brand names. Conversely, we would expect to find the “u”
sound in “ugh”underrepresented based on Jespersen (1922). Regarding the remaining
vowels, there is not enough prior research to make predictions.
3.2 Fricatives
In the case of fricatives (such as the “s”and “z”sounds), generally, we would expect to
find underrepresentation based on Smith (1998). However, regarding the fricative “s,”
in particular, we examine competing predictions. Although the findings from Schloss
(1981)andVandenBergh(1990) would predict underrepresentation, Jenkins et al.
(1958) would predict overrepresentation. For the remaining fricatives, there is insuffi-
cient prior research to make predictions.
3.3 Plosives
In the category of plosives (“p,”“b,”“t,”“d,”“k,”and “g”), we examine
parallel predictions. Johnson et al. (1964)andJenkinsetal.(1958)would
predict that plosives, generally, will be underrepresented among top brand
names. However, Vanden Bergh et al. (1987), Lowrey et al. (2003), and Cortese
(1998) would predict that plosive sounds will be overrepresented in word-initial
position.
3.4 Nasals
Regarding nasals (“m,”“n”,and“ng”), if findings by Johnson et al. (1964) hold, nasals
will be more common among top brand names. Regarding the nasal sound “m,”in
particular, Schloss (1981), De Klerk and Bosch (1997), and Jenkins et al. (1958), all
would predict overrepresentation among top brand names.
3.5 Affricates and approximates
One might predict that affricates (“ch”and “dj”) would be underrepresented based on
Johnson et al. (1964); however, as previously mentioned, that study conflated affricates
with stops, whereas we distinguish between the two. We, therefore, make no predic-
tions regarding affricates or approximates (“w,”“l,”“r,”and “y”), for which there is
554 Mark Lett (2015) 26:549–563
insufficient information. Our hope is that this study will provide a starting point for
further examinations of these sound categories.
4 Study
Our dataset of top brand names is a composite list of unduplicated Interbrand “top
brands”from 2001 to 2010 (Table 1in Appendix section). Names originating from non-
English speaking countries of origin were excluded to control for language differences.
One name that occurred in the list more than once (Thomson Reuters/Reuters) was
analyzed using the longer version. The resulting list consisted of 88 brand names.
In the absence of a suitable published list of “weak”brand names, we created the
best possible proxy for benchmarking in the following way. A sample of 1,000 proper
nouns (the category of words that includes brand names) was randomly selected from
the Corpus of Contemporary American English (COCA), a respected linguistic data-
base with over 100 citations (Davies 2008,2009).
1
This diverse and well-rounded
collection of source material serves as a useful proxy for the English language,
capturing the frequency of word usage. We use English as a comparison metric
because it is the international language of business with the third largest number
of speakers worldwide (Lewis 2009). Furthermore, English is spoken in more
countries than any other language, including approximately 60 % of Interbrand’s
top brand countries of origin. Brand names are most often encountered in text,
and the use of text in brand name linguistic analysis is an accepted methodology
(Klink 2000).
Cross-checking this sample with the Hoovers.com business directory confirmed that
all the proper nouns in our benchmark dataset belong to business names. The presence
of 13 % of the Interbrand top brand names in the benchmark dataset provides evidence
that it captures the relevant category of words. Duplicates and Interbrand top brand
names were removed, leaving 884 proper nouns belonging to a sample of general
names (Table 2in Appendix section). This is a conservative benchmark in that it is a
random sample of general names rather than less successful or weak brand names.
Despite the differences between the datasets, such as the frequency of acronyms among
top brand names (e.g., UBS, UPS, and HSBC), the benchmark provides a useful
comparison.
Two independent coders with formal linguistic training transcribed the Interbrand
and benchmark brand names into the International Phonetic Alphabet so that each
sound was represented by one symbol and each symbol represented only one sound.
Transcriptions were based on official firm pronunciations. A third party reconciled
discrepancies.
2
For each brand name in the Interbrand and benchmark datasets, sound frequencies
were calculated by multiplying the number of occurrences in each word by the
1
We use the 95 million words COCA magazine database which references almost 100 popular magazines
such as Tim e,Men’s Health,Good Housekeeping,Cosmopolitan,andSports Illustrated for the years 1990–
2012 (Davies 2008). The database draws from publications in a wide variety of domains such as news, health,
home and gardening, women's, financial, religion, and sports, resulting in a balanced portrayal of the English
language.
2
The coding of these brand names and the benchmark dataset is available from the first author upon request.
Mark Lett (2015) 26:549–563 555
frequency of the word in the COCA database. This allowed us to examine the relative
frequencies of sounds rather than simply presence versus absence.
3
We examined differences in sounds frequencies between datasets using independent
sample ttests. Due to the large number of comparisons (41 different phonemes) and the
inflated risk of type I error, tests were conducted using a conservative Bonferroni
adjusted alpha level of .0012. This calculation is based on Dunn (1961), who recom-
mends dividing the standard alpha of .05 by the number of comparisons (.05/41). In
four cases (“th”as in “the,”“sh”as in “shine,”“zh”as in “Beijing,”and “oo”as in
“Nook”), analyses were not done because total observations among top brand names
were zero.
5 Analysis and results
5.1 Vowels
All but two vowel sounds occur in different proportions between top brand names and
the benchmark. Of these, five are overrepresented and four are underrepresented
(Fig. 2).
As predicted, the “o”sound in “posh”is overrepresented among top brand names
(e.g., “Starbucks”), consistent with Lowrey and Shrum (2007) (Recall that this “o”
sound is a back vowel with the sound symbolic meaning large, masculine, slow, dark,
and round) (Hinton et al. 1994). Also as predicted, the “u”sound is underrepresented
among top brand names (e.g., “Budweiser”), consistent with Jespersen (1922).
The most common vowel sound among top brand names, which also varies the most
between data sets, is the “e”in “eBay.”The “a”in “AOL,”“e”in “Estee Lauder,”and
“a”in “Kraft”are also overrepresented in top brand names. These sounds are all front
vowels, which carry the well established cross-cultural sound symbolic meanings
small, feminine, fast, light, and angular.
Vowel sounds underrepresented among top brand names include the “i”in “Intel,”
“a”in “Amazon,”“au”in “Ralph Lauren,”and “u”in “Budweiser.”Two sound s—the
“o”in “Mobil”and the “oo”in “Yah oo,”—were not significantly different between data
sets.
It is notable that only one front vowel was underrepresented, whereas only one back
vowel was overrepresented. The only back vowel more common among top brand
names is the “o”sound in “posh,”which as Lowrey and Shrum (2007) discovered, is so
positively perceived that its presence in experimental brand names can “trump”a
discordant factor such as incongruent product category.
5.2 Fricatives
Five of the six fricatives examined differ between data sets (Fig. 3). Of these, all are
more common among top brand names, with one exception, “th”as in “Three M.”One
fricative (“f”as in “Ford”) is not significantly different. This is contrary to Smith
3
One brand name (Gap) was discarded because its frequency could not be determined since it is indistin-
guishable from the common noun “gap.”
556 Mark Lett (2015) 26:549–563
(1998), who found that names with fricatives are negatively associated with electability
of political candidates.
The fricative “s”sound is significantly overrepresented, contrary to Schloss (1981)
and Vanden Bergh (1990), but consistent with Jenkins et al. (1958).
5.3 Plosives
Contrary to predictions based on Jenkins et al. (1958)andJohnsonetal.
(1964), we find that every plosive is overrepresented among top brand names
(Fig. 4a).
Consistent with expectations based on Vanden Bergh et al. (1987) and Lowrey et al.
(2003), in word-initial position, four of six plosives differ between data sets, and these
are all overrepresented (Fig. 4b). Interestingly, this pattern reverses when non-English
country of origin brand names are included in the analysis; then only two word-initial
plosives are significantly different between data sets (“b”and “k”), and these are under
rather than overrepresented, suggesting that the pattern of differences in this sound
category is language specific.
0%
10%
20%
30%
40%
50%
60%
Proportion of Names with Sound
Interbrand
Benchmark
Fig. 2 Frequencies of vowel sounds in top brand names. ***Significant at the 0.001 level. **Marginally
significant at the 0.01 level
0%
5%
10%
15%
20%
25%
30%
35%
th*** f v*** s*** z** h***
Proportion of Names with Sound
Interbrand
Benchmark
Fig. 3 Frequencies of fricatives in top brand names. ***Significant at the 0.001 level. **Marginally
significant at the 0.01 level
Mark Lett (2015) 26:549–563 557
5.4 Nasals
Of the three nasal sounds (“m”,“n,”and “ng”), “m”(as in “Marriott”) is overrepre-
sented, consistent with predictions based on Schloss (1981) and Jenkins et al. (1958).
However, one is underrepresented (“n”as in “Nike”), and one does not differ signif-
icantly between top brand names and the benchmark (“ng”as in “Boeing”). Therefore,
our prediction that nasal would be overrepresented, generally, based on Johnson et al.
(1964), is only partially supported (Fig. 5).
5.5 Affricates and approximates
Only one affricate is statistically significant: “dj”as in “Gillette”is underrepre-
sented (Fig. 6). (The “ch”as in “Merrill Lynch”shows no significant difference
between datasets.)
Finally, whereas one of the approximates is overrepresented (“l”as in
“Levi’s”), two are underrepresented (“r”as in “Ralph Lauren”and “w”as in
“Wall Street Journal”). One (“y”as in “Yahoo!”) is not significantly different
between datasets (Fig. 7).
0%
5%
10%
15%
20%
25%
30%
35%
p* b*** t*** d*** k*** g**
Proportion of Names with Sound
Interbrand
Benchmark
0%
5%
10%
15%
Proportion of Names with Sound
Interbrand
Benchmark
a
b
Fig. 4 a Frequencies of plosives in top brand names. ***Significant at the 0.001 level. **Marginally
significant at the 0.01 level. *Marginally significant at the 0.03 level. bFrequencies of word-initial plosives
in top brand names. ***Significant at the 0.001 level
558 Mark Lett (2015) 26:549–563
6 General discussion
This research contributes to the literature by examining sound patterns in top brand names,
including the previously unexamined sound categories of affricates, approximates, and
nasals. We introduce a new method of analysis by using phonetic transcription to compare
sound frequencies between top brand names and a sample of general brand names, based
on a well-established corpus. Our results show that top brand names display different
sound patterns from general brand names. Knowledge of the sounds more prevalent in top
brand names has the potential to improve best naming practices and naming strategy.
Managers should consider sounds significantly more common among top brands
when crafting new names, such as the plosives “p,”“b,”“t,”“d,”“k,”and “g.”
Similarly, “l,”“m,”“v,”“s,”“z,”and “h”may confer positive sound symbolic
properties to brand names. The vowel sounds “e”in “eBay,”“a”in “AOL,”“e”in
“Estee Lauder,”“a”in “Kraft,”and “a”in “Starbucks”are also potentially positive
sounds for brand names. The sounds overrepresented among the top brand names
0%
10%
20%
30%
40%
50%
m*** n***
Proportion of Names with Sound
Interbrand
Benchmark
Fig. 5 Frequencies of nasals in top brand names. ***Significant at the 0.001 level
0%
5%
10%
15%
ch dj***
Proportion of Names with Sound
Interbrand
Benchmark
Fig. 6 Frequencies of Affricates in Top Brand Names. ***significant at the .001 level
Mark Lett (2015) 26:549–563 559
largely correspond with what is known about sound symbolism, suggesting that sound
symbolism may play an important role in brand name attractiveness. Managers and
name inventors should therefore consider sound symbolism when crafting new names.
Overrepresented sounds, including the “m”in “Marriott”and “s”in “Sprite,”should
be tested experimentally for positive associations. Similarly, sounds underrepresented
among top brand names, including the “u”in “Budweiser,”should be tested for
negative associations.
Although some of the findings are consistent with expectations based on the
previous research, others are not. This suggests the presence of unexamined moder-
ators. Word position (i.e., word-initial, word-middle, or word-final) is one likely
moderator (Smith 1998;KlinkandWu2013). Product category is also a likely
moderator of positive and negative sound perceptions (Lowrey and Shrum 2007;
Yorkston and Menon 2004). For instance, front vowels are more common among
electronics in top brand names than back vowel sounds, possibly because they
convey smallness and fastness, which are desirable attributes for electronics. While
the sample size of brand names of electronics companies is too small to draw
definitive conclusions, examining differences in linguistic characteristic among
product categories represents a potentially rich area for future research.
One limitation of the present research is the difference in name types between top
brands and the benchmark dataset, such as the frequency of acronyms among top brand
names. Future research should examine frequencies and consumer preferences for
different name types. Similarly, although the current research only examines English
speakingcountryof origin names, differencesacross languages should be consideredin
future studies.
Changes in sound patterns among top brand names over time may also be illumi-
nating. If, for instance, front vowels have become more prevalent among top brands as
consumer electronics shifted toward smaller and faster devices, this might represent
shifting sound patterns that mirror shifting consumer preferences. This, and future
research on the moderating roles of word position and product category, will lead to
a greater understanding of the marketing impact of brand name sound attributes.
Acknowledgement Many thanks to James Kellaris, Jakki Mohr, and Mary Steffel for their feedback,
Jackelyn Torres Van Buren, Kelsey Fanning, and Julie Workman-Hank for their assistance with data analysis,
Joel Steckel for his editorial guidance, and the anonymous reviewers for their valuable comments.
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
w*** l*** r*** y
Proportion of Names with Sound
Interbrand
Benchmark
Fig. 7 Frequencies of approximants in top brand names. ***Significant at the 0.001 level
560 Mark Lett (2015) 26:549–563
Appendix
Tab l e 1 2001–2010 Interbrand top brands used in analysis
3m Coca-Cola Intel MTV
Accenture Colgate J.P.Morgan Nike
Adobe Compaq Jack Daniels Oracle
AIG Dell Johnny Walker Pampers
Amazon.com Disney Johnson & Johnson Pepsi
American Express Duracell Kellogg's Pfizer
AOL eBay KFC Pizza Hut
Apple Estée Lauder Kleenex Polo Ralph Lauren
Avo n Fedex Kodak Smirnoff
Bacardi Financial times Kraft Sprite
Barbie Ford Levi's Starbucks
Barclays GE LG Sun
Blackberry Gillette Marlboro Texas Instruments
Boeing Goldman Sachs Marriott Thomson Reuters
BP Google McDonald's Tiffany & Co.
Budweiser Harley-Davidson Merck Time
Burberry Heinz Merrill Lynch UPS
Burger King Hertz Microsoft Visa
Campbell's Hilton Mobil Wall Street Journal
Caterpillar HP Morgan Stanley Wrigley
Cisco HSBC Motorola Xerox
Citi IBM Intel Yahoo!
Tabl e 2 Sample from benchmark
dataset (complete dataset includes
884 randomly selected names
available from authors)
Atlantic Bailey Baker Bronx
Brooks Bush Chrysler Dakota
Dodge Edwards Ellis Foster
GM Holland Hughes Jackson
Jupiter Kansas Klein Lakers
Lee Mac Marc Nato
Neptune Nixon Orion Pearl
Reno Rush Saturn Sears
Ted Temple Union Village
Walker Wendy Yankees Young
Mark Lett (2015) 26:549–563 561
References
Aaker, D. A., & Keller, K. L. (1990). Consumer evaluations and brand extensions. Journal of Marketing,
54(January), 27–41.
Allen, C. T., Fournier, S., & Miller, F. (2008). Brands and their meaning makers. In C. Haugtvedt,
P. Herr, & F. Kardes (Eds.), Handbook of consumer psychology (pp. 781–822). New York:
Psychology Press.
Argo, J., Popa, M., & Smith, M. C. (2010). The sound of brands. Journal of Marketing, 74(4),
97–109.
Bolinger, D. L. (1950). Rime, assonance, and morpheme analysis. Word , 6 ,117–136.
Cortese, M. J. (1998). Revisiting serial position effects in reading. Journal of Memory and Language, 3(9),
652–665.
Coulter, K. S., & Coulter, R. A. (2010). Small sounds, big deals: phonetic symbolism effects in pricing.
Journal of Consumer Research, 37(2), 315–328.
Davies, M. (2008). The Corpus of Contemporary American English: 450 million words, 1990-present.http://
corpus.byu.edu/coca/. Accessed 24 Mar 2014.
Davies, M. (2009). The 385+ million word corpus of contemporary American English (19902008+): Design,
architecture, and linguistic insights. International Journal of Corpus Linguistics, 14(2), 159–190.
De Klerk, V., & Bosch, B. (1997). The sound patterns of English nicknames. Language Sciences, 19(4), 289–
301.
Dunn, O. J. (1961). Multiple comparisons among means. Journal of the American Statistical Association, 56,
52–64.
Erlich, J. (1995). Giving drugs a good name, The New York Times Magazine, September 3, 36–37.
Fromkin, V., Rodman, R., & Hyams, N. (2010). An introduction to language (9th ed.). Boston: Wadsworth.
Hartley, R. F. (1992). Marketing mistakes (5th ed.). New York: John Wiley & Sons.
Heath, D. & Heath, C. (2011). How to pick the perfect brand name, Fast Company, January 3.
Hinton, L., Nichols, J., & Ohala, J. (Eds.). (1994). Sound symbolism.Cambridge:CambridgeUniversityPress.
Hoffman, G. (2008). New car names - how they get 'em, CNN, March 7.
Interbrand (2011). Methodology. Accessed 20 Oct 2011. http://www.interbrand.com.Accessed24Mar2014.
Jenkins, J. J., Russell, W. A., & Suci, G. J. (1958). An atlas of semantic profiles for 360 words. The American
Journal of Psychology, 71(4), 688–699.
Jespersen, O. (1922). Language: its nature, development and origin.London:Allen&Unwin.
Johnson, R. C., Suzuki, N. S., & Olds, W. K. (1964). Phonetic symbolism in an artificial language. Journal of
Abnormal and Social Psychology, 69(2), 233–236.
Klink, R. R. (2000). Creating brand names with meaning: the use of sound symbolism. Marketing Letters,
11(1), 5–20.
Klink, R. R. (2001). Creating meaningful new brand names: a study of semantics and sound symbolism.
Journal of Marketing Theory and Practice, 9(2), 27–34.
Klink, R. R. (2003). Creating meaningful brands: the relationship between brand name and brand mark.
Marketing Letters, 14(3), 143–157.
Klink, R. R. (2009). Gender differences in new brand name response. Marketing Letters, 20(3), 313–326.
Klink, R. R. & Wu, L. (2013). The role of position, type and combination of sound symbolism imbeds in
brand names. Marketing Letters.
Köhler, W. (1929). Gestalt psychology. New York: Liveright.
Lewis, P. (ed.). (2009). Ethnologue: Languages of the World, Sixteenth edition. Dallas, Tex.: SIL International.
http://www.ethnologue.com.Accessed24Mar2014.
Lowrey, T. M., & Shrum, L. J. (2007). Phonetic symbolism and brand name preference. Journal of Consumer
Research, 34,406–414.
Lowrey, T. M., Shrum, L. J., & Dubitsky, T. M. (2003). The relation between brand-name linguistic
characteristics and brand-name memory. Journal of Advertising, 32(3), 7–17.
Maurer, D., Pathman, T., & Mondloch, C. J. (2006). The shape of boubas: Sound-shape correspondences in
toddlers and adults. Developmental Science, 9(3), 316–322.
Plato. (1985). In E. Hamilton & H. Cairns (Eds.), The collected dialogues of plato including the letters.
Princeton: Princeton University Press.
Schloss, I. (1981). Chickens and pickles. Journal of Advertising Research, 21,47–49.
Smith, G. W. (1998). The political impact of name sounds. Communication Monographs, 65(2),
154–172.
562 Mark Lett (2015) 26:549–563
U.S. Patent and Trademark website accessed 4/19/2012. http://www.uspto.gov/dashboards/trademarks/main.
dashxml.Accessed24Mar2014.
Vanden Bergh, B. (1990). The rekurring kase of the special K. Journal of Advertising Research, 30(5), RC9–
RC12.
Vanden Bergh, B., Adler, K., & Oliver, L. (1987). Linguistic distinction among top brand names. Journal of
Advertising Research, 27(4), 39–44.
Yorkston, E., & Menon, G. (2004). A sound idea: Phonetic effects of brand names on consumer judgments.
Journal of Consumer Research, 31(1), 43–51.
Mark Lett (2015) 26:549–563 563
