Dr. Carla L.

Kuesten
Fast Moving Consumer Goods, USA

Exploring Sensory Brand Identity for Retail Coffees

Abstract Attribute Scales Linear discriminant analysis shows the coffees may • Five of the twelve brands showed similar
be correctly classified by brand membership using sensory brand identity for their products,
Sensory benchmarking was conducted using 40 Attributes are defined below:
the sensory attributes. regardless of form or caffeine level.
unflavored, U.S. retail coffees representing ground,
Color is a measure of the visual Sour Aromatics measures sour • Brands are identified by color with individual
whole bean, and instant coffees from the top 10 gauge of a coffee’s color intensity aromatics such as vegetable-like, • Remaining brands could improve sensory
leading brands of each category. The coffees covered ranging from light to dark as fermented, or fatty acid sour. The products represented by form and caffeine level. brand identity by delivering products that are
a range in degree-of-roast—light to dark and included perceived by the panelist. scale ranges from none to strong.
• The brand mean is represented by a diamond, more similar across different forms and caffeine
both caffeinated and decaffeinated varieties. Principal Cup Aroma is a measure of the Sweet intensity is a measure of
Component Analysis was applied to the data as well intensity of the aromatics present the level of sweet basic taste. The color coded by degree of roast. levels.
in the headspace of the cup using reference standards are sucrose
as linear Discriminant Analysis to determine whether a scale of none to strong. Panelists solutions. The scale ranges from • The ellipses are centered at the brand mean and • Some brands were notably less consistent
the sensory attribute ratings could be used to correctly describe the type of aromatics in none to strong.
identify the specific brand and type of coffee and the their comments. represent the average brand within product than other brands; some products could be
Sour measures the intensity of variation.
presence or absence of unique brand profiles. Balance is a measure of the blend sour basic taste. The reference confused with neighboring brands.
of the product. It takes into account standards are citric acid solutions.
the basic tastes, aromatics, and The scale ranges from none to • Coffees demonstrating a strong sensory brand • Decaffeination may change products enough
Introduction mouthfeels. The scale ranges from
unblended to blended.
strong. identity will fall within the brand ellipse. to be noticeably different from the rest of the
Bitter intensity is a measure of products within a brand.
The connection between sensory characteristics and Fullness measures the richness or the level of bitter basic taste. The Coffee Benchmarking Discriminant
body of flavor when tasting the Analysis Map
brands of retail coffees was explored. Brand identity, reference standards are caffeine
product. The scale ranges from solutions. The scale ranges from Brand Degree of Roast
• Brands that share a common sensory space,
a company’s competitive advantage and strategic thin to full. none to strong. A Light Roast
such as “light roast,” probably compete in the
B
asset, encompasses many factors—brand name, Coffee Aromatics Intensity is a
C
Light - Medium Roast

image, associations, personality, logo, symbols,

Mouthfeel is a measure of I J
D D Medium - Dark Roast
marketplace for the same consumer taste
Linear Discriminant 2

E
measure of the strength and type perceived amount of trigeminal F
Dark Roast

colors, jingles, etc.—that portray the perception of of coffee-like flavor aromatics. effects such as astringency or B
G segments; the consumer’s perception of quality
H
Panelists score this none to strong tannins usually associated with
the brand and what the brand stands for to the and describe their notes which can coffee. The scale ranges from none
A
GE
L
I
J
may be the final differentiator.
consumer. The sensory profile has a strong influence include descriptions of coffee roast to strong.
C K
L

on how the brand is perceived and is often used to and mercaptans.
communicate the brand identity and position. Other Coffee Aromatics is a
Successful companies establish a sensory brand measure of the strength and type
of other coffee-related
identity. characteristics such as grainy, nutty,
brothy (hydrolyzed vegetable
protein) and burnt or green bean
Materials and Methods aromatics. The scale ranges from
none to strong.
Sample Preparation. The coffees were purchased Degree of Roast is a measure
at retail and prepared as follows: of the perceived amount of
roasting of the coffee character,
• All ground and whole bean coffee samples were from none to dark (espresso-like).
standardized using the same amount of coffee to Sweet Aromatics measures
water ratio in the same Bunn® pour-over machine. sweet aromatics such as
vanilla/vanillin, burnt sweet, and
caramelized sweet. The scale
• Whole bean coffees were ground using a spinning ranges from none to strong.
blade type grinder for a specified length of time.
• All instant coffee samples were standardized using
the same coffee to water ratio.
Results and Discussion
• All coffees were made using Poland Spring® water.
The flavor attribute data were summarized into two
• Coffees were presented in preheated cups.
flavor indices using Principal Component Analysis (PCA).
Protocol. The coffees were evaluated by a trained
Caffeine:
• Brands in a different sensory space, such as
Others is a measure of basic tastes Form Caffeinated Decaffeinated
or aromatics (often off-flavors) that
F
K
Whole
Bean
“dark roast,” focus on a very different consumer
are not measured in any of the H Ground
segment.
previous attributes. Panelists score Instant
this none to strong and describe
their notes which can include Wet • Manufacturers that produce more than one
Ash Tray(WAT), woody, musty, Linear Discriminant 1 brand may intentionally design their brands
Dirty Dishrag sour (DDR), and
rubbery sulfide.
Fig. 4. Linear Discriminant Analysis Bi-plot
to span different regions of the map, allowing
Aftertaste measures the intensity them to reach consumer taste segments with
of the basic tastes, aromatics or Results of the discriminant analysis shows:
mouthfeels still present in the
targeted, well-defined sensory brand identities.
mouth one minute after the last • The sensory attributes found to be most
swallow. Panelists score this none • The white space within the map suggests
to strong and describe the important in predicting the coffee brand and
opportunities between light and dark roast
characteristics in their comments. type include cup aroma, coffee intensity and
type, and sour aromatics. coffees to establish additional products with a
unique sensory profile.
• Evidence supports highly accurate brand
classification based on the sensory data alone,
though some misclassification for a few less Conclusions
discriminated brands exists.
Sensory benchmarking using a trained panel with
• This confusion is most readily evident in the principal component and discriminant analyses
lighter roasted coffees which deliver less flavor can be used for guidance in selecting and
Lower scores are... Higher scores are...

panel in a controlled environment as follows: intensity overall. maintaining a unique sensory brand identity in
Less degree of roast More degree of roast
Less coffee aromatics More coffee aromatics the marketplace.
• Profile Attribute Analysis (PAA), a quantitative Index 1 Thinner Fuller Attribute plots are useful for illustrating the
Less bitter More bitter
descriptive analysis method, was used to evaluate “Coffee Identity”
Less aftertaste More aftertaste similarity or differences in brand (or product)
the coffees using the scales defined below. Less cup aroma
Less blended
More cup aroma
More blended sensory profiles. References
• Coffee samples were randomized and coded for Dawidowska, Katarzyna. April 2002. Caffeine
blind evaluation by the panelists. More others/off-notes Less others/off-notes
Color
BA
BA Color DA
A D
Index 2 More other coffee aromatics Less other coffee aromatics
Cup Aroma
Balance BA
Cup Aroma
Balance A D Overload. American Demographics, Vol. 24:4, 16.
• Panelists tasted the same amount of each sample “Other
Characteristics”
(i.e. grain, nuts, burnt (i.e. grain, nuts, burnt Fullness
Coffee Aromatic Intensity
B
BA
Fullness
Coffee Aromatic Intensity
A
A
D
D
and evaluated the specified attributes at their or green beans) or green beans) Other Coffee Aromatics A Other Coffee Aromatics DA Coffee Recent Developments. 2001. R. J. Clarke
Attribute

Attribute

Degree of Roast BA Degree of Roast A D

appropriate temperatures, aroma between The “Coffee Identity” index includes attributes that are closely associated with coffee,
Sweet Aromatics BA Sweet Aromatics DA and O. G. Vitzthum, editors. Blackwell Science
Sour Aromatics AB Sour Aromatics DA
150º–160ºF, and flavor between 140º–150ºF. i.e., Degree of Roast, Coffee Aromatics, Fullness, Bitterness, and Aftertaste. Sweet AB Sweet D
A Ltd.
Sour AB Sour AD
The “Other Characteristics” Index includes off-notes and other coffee aromatics (such Bitter BA Bitter A D
• Panelists used spring water and unsalted crackers as grain, nuts and green or burnt beans) Mouthfeel BA Mouthfeel D
A Heath, Robert. 1999. Just popping down to the
to rinse their mouths between samples and waited Fig. 2. PCA Indices
Others/Off-Notes
Aftertaste
AB
BA
Others/Off-Notes
Aftertaste
A
D
AD
approximately 5 minutes between samples. 1.0 2.0 3.0 4.0 5.0 6.0 7.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 shops for a packet of image statements: A new
Rating Rating

Descriptors Scale The 33 retail coffees selected for this study show a theory of how consumers perceive brands.
Fig. 5. Sensory Attribute Line Plots
Type Attributes Recorded 1 2 3 4 5 6 7 wide range of flavor characteristics. Journal of Market Research Society, London.
Less others/off-notes 8.0
Appearance Color Light Dark Less other coffee
25
Vol. 41:2, 153–169.
aromatics 26 Several key learnings from this study are
Aromatic Cup Aroma None Strong 28 1 4
summarized below: Building Strong Brands. 1996. David A. Aaker.
“Other Characteristics”

32 8
7.0 29
6
Flavor Balance Unblended New York: The Free Press.
Flavor Index 2

Blended 31 30
Fullness Thin Full
33 27 • Locations within the map represent different
24 12
Coffee Aromatics None Strong 6.0 20 7 10 18
21 sensory profiles and possibly different consumer Marketing Aesthetics: The Strategic Management
19
Other Coffee Aromatics None Strong 16 23
17 taste segments. of Brands, Identity and Image. 1997. Bernd H.
Degree of Roast None Dark 22
Sweet Aromatics None 5.0 • Cup aroma, coffee intensity and type, sour Schmitt and Alex Simonson. The Free Press.
Strong 10 Least
0.84
Sour Aromatics None Strong
Significant
Difference aromatics, balance, and bitterness are the
Sweet None More others/off-notes Anonymous. January 2000. Which brew for you?
Strong More other coffee
3.14
sensory attributes most important in
Sour None Strong
aromatics 4.0
Consumer Reports, Vol. 65:1, 29–33.
Bitter None
15
Less degree of roast
20 25 30
More degree of roast
distinguishing and classifying brands.
Strong Less coffee aromatic intensity More coffee aromatic intensity
Mouthfeel None Strong
Thinner
Less bitter Flavor Index 1
Fuller
More bitter • Consistency is key to ensuring sensory brand Watkins, Carolyn. May 2000. Coffee service:
less aftertaste More aftertaste
Others None Strong
“Coffee Identity”
A whole latte opportunity. Food Management,
Aftertaste None
Less cup aroma
Less blended
More cup aroma
More blended identity (illustrated by the relative size of
Strong Vol. 35:5, 66–72.
KEY:
Grd = Ground Decaffeinated - empty
ellipses).
Panelists record word descriptors of the perceived effect for the WB = Whole Bean Caffeinated - filled
indicated attributes. Ins = Instant

Fig. 1. Coffee Attribute Scales Fig. 3. PCA Flavor Map

15 Acorn Park Contact: Dr. Carla Kuesten

Cambridge, MA 02140 Tel: 617-498-5181
www.tiax.biz Email: kuesten.carla@tiax.biz