Starbucks Corporation

Trademark Trial and Appeal Board

Jan 17, 2019

No. 85792872 (T.T.A.B. Jan. 17, 2019)

Hearings: July 17, 2018 (Serial No. 86689423) Mailed: January 17, 2019
and November 21, 2018 (Serial No. 85792872)

UNITED STATES PATENT AND TRADEMARK OFFICE
Trademark Trial and Appeal Board
————
In re Starbucks Corporation
_____

Serial Nos. 85792872 and 86689423
_____

Julia Anne Matheson of Hogan Lovells US LLP
for Starbucks Corporation.

Mark Rademacher, Trademark Legal Policy Office,
Jennifer Chicoski, Administrator for Trademark Policy and Procedure.
_____
Before Adlin, Goodman and Heasley,
Administrative Trademark Judges.

Opinion by Adlin, Administrative Trademark Judge:

Applicant Starbucks Corporation seeks registration of two marks consisting of “a
green circle placed centrally on the front exterior side of a white cup.” Although the
two marks are not identical, and the grounds for their refusal are only partially
similar, the records, arguments and evidence in the two appeals overlap enough that
we decide both appeals in this single decision; “each appeal stands on its own merits,”
however. In re Hudson News Co., 39 USPQ2d 1915, 1916 & n.5 (TTAB 1996); see also
In re Supply Guys Inc., 86 USPQ2d 1488, 1490 (TTAB 2008).
This Opinion is Not a
Precedent of the TTAB
Serial Nos. 85792872 and 86689423
2
I. Background
Applicant seeks registration of the two marks summarized below:
Application
Serial No.
85792872
(the “’872 Application”)
86689423
(the “’423 Application”)
Drawing of Mark
Goods/Services “Coffee, tea, coffee and tea
based beverages, and cocoa”
in International Class 30;
and

“Restaurants, cafes,
cafeterias, snack bars and
coffee bars; carry-out
restaurant and food
preparation services” in
International Class 43.
“Coffee-based beverages;
cocoa; chocolate-based
beverages; tea and herbal tea;
tea and herbal tea-based
beverages” in International
Class 30; and

“Restaurant, cafe, cafeteria,
snack bar, coffee bar, tea bar,
tea house, and carry out
restaurant services; catering
services; preparation of food
and beverages; restaurant
services featuring a customer
loyalty program that provides
restaurant benefits to reward
repeat customers” in
International Class 43.
Description of Mark
in Application
“The mark consists of a green
circle placed centrally on the
front exterior side of a white
cup. The broken lines depict
the outline of the white cup
and show the relative
placement of the green circle.
No claim is made to the
portions of the drawing
represented by the dotted
lines nor the overall shape or
size of the cup.”
“The mark consists of a green
circle placed centrally on the
front exterior side of a white
cup. The colors green and
white are claimed as
elements of the mark. The
color black appears simply to
designate the shape of the
elements comprising the
mark and is not a part of the
mark.”

First Use Dates March 8, 2011 September 2, 2008
Serial Nos. 85792872 and 86689423
3
The Examining Attorney finally refused registration of the mark in the ’423
Application because he found that Applicant’s drawing of the mark ( ) is not a
“substantially exact representation” of the marks shown on Applicant’s two
specimens, depicted below:

the “Package Specimen” the “Counter Display Specimen.”
The Examining Attorney finally refused registration of the mark in the ’872
Application on several grounds. As with the ’423 Application, he found that
Applicant’s drawing of the mark in the ’872 Application ( ) is not a
Serial Nos. 85792872 and 86689423
4
“substantially exact representation” of the marks shown on any of Applicant’s six
specimens, depicted below:
1
2

1 Submitted with original application on December 3, 2012.
2 The four specimens following the photograph of an actual coffee cup with a green “siren”
logo in the center were all submitted with Applicant’s Office Action response of February 10,
2015.
Serial Nos. 85792872 and 86689423
5

3
In addition, the Examining Attorney refused registration of the mark in the ’872
Application because: Applicant did not describe the mark in the application as being
three-dimensional; “the mark is not sufficiently associated with the services in the
specimens of record;” and Applicant has not established that its mark has acquired
distinctiveness.4

3 Submitted with Office Action response of November 17, 2015.
4 The Examining Attorney also finally refused registration of the mark in the ’872 Application
on the ground that it is not inherently distinctive. Although Applicant appealed from this
refusal as well as the others, 21 TTABVUE 9, at the oral hearing it “conceded” on the question
of inherent distinctiveness. Thus, Applicant now relies solely on its claim that the mark has
acquired distinctiveness. In any event, as explained below in connection with Applicant’s
claim of acquired distinctiveness, the mark in the ’872 Application is not the type of mark
likely to be perceived as a source indicator, and is protectable only upon proof of acquired
distinctiveness. See e.g. In re American Academy of Facial Plastic and Reconstructive
Surgery, 64 USPQ2d 1748, 1753 (TTAB 2002) (“However, ordinary geometric shapes such as
circles, ovals, squares, stars, etc., are generally regarded as nondistinctive and protectable
only upon proof of acquired distinctiveness.”).
Serial Nos. 85792872 and 86689423
6
II. Analysis
At the outset, we reject Applicant’s argument that the Examining Attorney’s
refusals based on the Trademark Rules of Practice have “no basis in any statutory
provision (or the common law or another authority carrying the force of law).” 8
TTABVUE 7 (’423 Application); 15 TTABVUE 21 (’872 Application). In fact, Applicant
“shall comply with such rules or regulations as may be prescribed by the Director” of
the Office, which include the Trademark Rules applied by the Examining Attorney in
refusing registration of Applicant’s marks. 15 U.S.C. § 1051(a)(4). These rules and
regulations include Trademark Rule 2.51(a), at issue in both of the involved
applications. See e.g. In re Chemical Dynamics Inc., 839 F.2d 1569, 5 USPQ2d 1828
(Fed. Cir. 1988) (applying Trademark Rule 2.51(a) in considering an analogous
refusal of registration). We now turn to the substantive refusals.
A. The ’423 Application
The Examining Attorney argues that the “green circle” in Applicant’s drawing of
the mark in the ’423 Application is “featureless,” while the instant coffee Package
Specimen bears “an irregular green circle formed by paintbrush strokes”:

Serial Nos. 85792872 and 86689423
7
12 TTABVUE 3, 7.5 More specifically, the Examining Attorney argues that the circle
in the Package Specimen is “formed by paintbrush strokes, with the color green
ranging in intensity over different parts of the circle due to the thickness of the paint
applied to the surface.” Id. at 7. According to the Examining Attorney, the mark in
the Package Specimen is different than the mark in the drawing because it “conveys
the idea of … the whimsy and fluidity of fluid paintbrush strokes.” Id. at 8.
The Examining Attorney argues that the Counter Display Specimen “features a
white cup shown in perspective, suggesting three dimensions, and having accurate
and integral details, such as creases and an opening in the lid, depicted in green lines.
This second cup includes a solid green circle.” Id. at 7. Thus, according to the
Examining Attorney, the Counter Display Specimen is different from the drawing
because it “conveys the idea of a realistic cup in three dimensions ….” Id. at 8.
The Examining Attorney also contends that the mark in the drawing is not “part
of” the marks shown in the specimens and does not make a “separable commercial
impression” therefrom. Id.
Applicant contends that the Examining Attorney’s findings are inconsistent with
both precedential Board cases and the Office’s treatment of certain third-party
registrations that Applicant made of record.6 8 TTABVUE 13-20. It also argues that

5 Citations in this subsection are to the record in the ’423 Application.
6 Applicant argues in its briefs that the refusal should be reversed based on the results of Dr.
Leon Kaplan’s consumer survey to assess whether the marks in the ’872 Application and
Applicant’s related but uninvolved application Serial No. 85792857 had acquired
distinctiveness (“Kaplan Survey”). During the oral hearing, however, Applicant conceded
that the Kaplan Survey is irrelevant to the refusal to register the mark in the ’423
Application. Indeed, the survey does not address whether the drawing in the ’423 Application
Serial Nos. 85792872 and 86689423
8
the differences between the drawing and the specimens are “minute,” and that
“consumers confronted with [either specimen] come away with the impression of
[Applicant’s drawing].” Id. at 9, 11.
Turning first to the Class 30 goods, we find that Applicant’s drawing is a
“substantially exact representation of the mark as used” on the instant coffee Package
Specimen. Trademark Rule 2.51(a). In fact, the cup in the drawing has the same
shape as and appears identical to the cup in the specimen, as both cups are white and
bear “a green circle placed centrally on the front exterior side.”
While the Examining Attorney is correct that the green circle in the drawing is
“featureless,” while the green circle on the Package Specimen is not, the “features” of
the specimen’s green circle are inconspicuous, negligible, and subordinate to the
green circle in which they appear. In fact, unless a consumer is close to the green
circle depicted in the specimen, its “features” would be imperceptible, or barely
perceptible.7 To the extent that the “fluid paintbrush strokes” in the specimen
constitute “features,” they do not change the essence of the mark, which remains, at
bottom, a “green circle.” Similarly, while the specimen’s circle is imperfect, especially
on its left side, it remains a circle, and the degree of imperfection is minimal.
The “range in intensity” of the color green, caused by “the thickness of the paint
applied to the surface,” presents a closer question, at least with respect to the darker

“matches” the specimens of record. We discuss the survey below, however, in connection with
the refusal to register the ’872 Application.
7 We have considered the viewing distances from which typical coffee and restaurant service
consumers would likely encounter Applicant’s marks.
Serial Nos. 85792872 and 86689423
9
green on the left side of the green circle in the specimen. However, the paint is still
green, even if it is a darker green. We find that the shade of green in Applicant’s
drawing is a “substantially exact” representation of the green in the specimen,
including the darker green to the left of the circle. In so finding, we have taken into
consideration that Applicant claims the color “green” generally, rather than any
particular shade(s) of green. Cf. In re Cook Medical Technologies LLC, 105 USPQ2d
1377, 1384 (TTAB 2012) (in the context of a Section 2(d) refusal to register a mark
with a color claim, stating that the cited “mark is described only as the color ‘blue’
(applied to a certain part of the goods), and therefore we have considered the mark to
be for any shade that would fall under the general term ‘blue.’ That is, we decide this
ex parte appeal based on the information on the face of the cited registration; we do
not read in limitations.”); Amsted Ind. Inc. v. West Coast Wire Rope & Rigging Inc., 2
USPQ2d 1755, 1760 (TTAB 1987) (“the likelihood of confusion issue must be
determined based on the same presumption … that the yellow color sought to be
registered by West Coast Wire in respect of its wire rope can be any yellow color,
including that which is identical to the yellow color used by Amsted”); see also TMEP
§ 1202.05(e) (Oct. 2018) (“if a mark includes gradations of color, the description
should so indicate”). Here, Applicant has not claimed any particular shade of green.
Of course, Applicant’s drawing defines the scope of any rights arising from
registration, but in this case, we find that the green in Applicant’s drawing is a
“substantially exact” representation of the shades of green in the Packaging
Serial Nos. 85792872 and 86689423
10
Specimen.8
Turning next to the Class 43 services, we find that Applicant’s drawing, shown
below on the left, is a “substantially exact representation of the mark as used” on the
Counter Display Specimen, shown below on the right:

The Examining Attorney is correct that the specimen depicts the cup’s lid differently
than the drawing, perhaps “suggesting three dimensions.” Under the circumstances
of this case, however, we nevertheless find that the drawing is a substantially exact
representation of the mark shown in the specimen. Both feature white cups with a
“green circle placed centrally on the front exterior side,” and this time the green
circles appear identical, and at the very least have substantially exact features.
While the mark in the specimen has “creases and an opening in the lid, depicted
in green lines,” this distinction does not mean that the drawing is not a substantially

8 While Applicant and the Examining Attorney argue about whether the mark in the drawing
creates a “separable” commercial impression from the mark in the Package Specimen, and
both cite “mutilation” cases, neither argues that the Package Specimen presents a
“mutilation” issue. We agree. We address mutilation below in our discussion of the Counter
Display Specimen and the ’872 Application.
Serial Nos. 85792872 and 86689423
11
exact representation of the mark in the specimen. According to Applicant’s
description of its mark, “[t]he color black appears simply to designate the shape of
the elements comprising the mark ….” Here, the shape of the lid in the drawing is the
same as the shape of the lid in the specimen, even if the specimen lid depicts a
perspective view of creases and an opening and the drawing lid does not.
In any event, the absence of creases and an opening in the drawing is at most a
minor alteration. Indeed, the essence of the mark is a white cup with a green circle,
and the lid is but a minor detail, in both size and commercial impression. The creases
and opening on the lid in the specimen are but a minor detail of this minor detail.9
We recognize that this is perhaps not a typical mutilation case, as the creases and
opening on the lid are not analogous to a word mark, or to a typical feature of a design
mark. Cf. In re Frankish Enters., Ltd., 113 USPQ2d 1964 (TTAB 2015); In re Supreme
Steel Framing System Ass’n Inc., 105 USPQ2d 1385 (TTAB 2012); In re Big Pig Inc.,
81 USPQ2d 1436 (TTAB 2006); In re Miller Sports Inc., 51 USPQ2d 1059 (TTAB
1999). Nevertheless, Applicant still “has some latitude in selecting the mark it wants
to register.” In re 1175856 Ontario Ltd., 81 USPQ2d 1446, 1448 (TTAB 2006).
Furthermore, the cup depicted in Applicant’s drawing “creates a commercial
impression as an indication of origin separate and apart from” the creases and

9 In In re Brouwerij Bosteels, 96 USPQ2d 1414, 1420 (TTAB 2010), we observed that “the
‘classic glass Coca-Cola Bottle’ referenced in Wal-Mart [Stores, Inc. v. Samara Brothers, Inc.,
529 U.S. 205, 54 USPQ2d 1065 (2000)], would be no less source indicating when served as an
open container to a restaurant patron than when pulled as a closed container from a refrig-
erated display or store shelf by a consumer.”

Serial Nos. 85792872 and 86689423
12
opening in the lid in the specimen. In re Frankish, 113 USPQ2d at 1974. As indicated,
leaving the creases and opening in the lid out of the drawing is merely a “minor
alteration” which does “‘not create a new and different mark creating a different
commercial impression.’” Id. (quoting In re Schecter Bros. Modular Corp., 182 USPQ
694, 695 (TTAB 1974)). In fact, this case is analogous to Schecter Bros., in which we
found that the drawing of the mark (shown below on the left) is not a mutilation of
the mark in the specimen (shown below on the right):

Indeed, the shadow in the specimen in Schecter Bros. is much larger and more
distinctive, and thus a significantly greater alteration, than the creases and opening
in this case.
Finally, under the circumstances of this case, we do not share the Examining
Attorney’s apparent concern that the cup in the Counter Display Specimen is “shown
in perspective, suggesting three dimensions.” In fact, both the drawing and the
specimen are in two dimensions, even if they reveal depth as well as width and
height.10 Moreover, even if we agreed that the cup in the specimen is shown in

10 “No drawing of a trademark that is the subject of an application for registration is
presented in true three-dimensional form. Even an application to register a configuration of
a product depicts a mark in two-dimensional form, perhaps from a view that yields a
Serial Nos. 85792872 and 86689423
13
perspective suggesting three dimensions, that distinction between the specimen and
drawing would be irrelevant in this case. There is no evidence upon which we could
find that consumers would perceive a counter display bearing a two-dimensional
picture of a cup any differently than the same counter display consisting not of a
picture of the cup, but the cup itself. Nor is there any evidence or basis to find that
consumers would perceive a cup shown in perspective any differently than the same
cup shown without perspective. The commercial impression, in the context of coffee
beverages and restaurant and café services, is the same ─ a white cup with a green
circle placed centrally on the front exterior side.
For all of these reasons, we find that the drawing of the mark in the ’423
Application is also a substantially exact representation of the mark in the Counter
Display Specimen.11 The refusal to register the mark in the ’423 Application is
therefore reversed.
B. The ’872 Application
Before addressing the merits of the refusals to register this mark, we overrule the
Examining Attorney’s and Applicant’s objections to certain evidence.

perspective of depth ….” Duramax Marine LLC v. R.W. Fernstrum & Co., 80 USPQ2d 1780,
1792 (TTAB 2006). The same is generally true of specimens, including in this case.
11 One could question the Examining Attorney’s assumption that the lid in the drawing does
not include creases or an opening. While it is true that Applicant has not claimed trademark
rights in creases or an opening, its drawing – which uses the color black only to designate
shape – is broad enough to encompass lids of the shape shown in the Counter Display
Specimen, whether or not those lids include openings or creases. Furthermore, takeaway
coffee cups are now ubiquitous (a development for which, the record suggests, Applicant
deserves much of the credit or blame), and we do not require evidence to recognize that many
takeaway coffee cup lids include creases or openings. Therefore, consumers could very well
assume from experience that the lid on the cup in the drawing in fact has creases and an
opening, even if they are not visible in the drawing.
Serial Nos. 85792872 and 86689423
14
1. Evidentiary Objections
The Examining Attorney objects to the Declaration of Ana B. Naydonov, an
attorney with Applicant’s former law firm, as untimely. While the Examining
Attorney is correct that “[t]he record in the application should be complete prior to
the filing of an appeal,” Trademark Rule 2.142(d), in this case, as Applicant points
out, the Examining Attorney failed to substantively address the Kaplan Survey for
almost four years after it was originally submitted with Applicant’s May 15, 2014
Office Action response. 13 TTABVUE.12 The Naydonov Declaration merely recounts
some of the history of Applicant’s efforts to direct the Examining Attorney to the
Kaplan Survey. While the declaration is ultimately irrelevant to our decision, we have
considered it, as its late filing was the result of the Office’s long delay in substantively
commenting on the Kaplan Survey until after the appeal was filed.13
The Examining Attorney also objects to third-party Registration No. 4072993,
because he contends that it was not properly introduced. However, it was timely and
properly submitted with Applicant’s November 21, 2017 Request for Remand. 11
TTABVUE 326-28; TBMP § 1209.04. The Examining Attorney’s objections are
therefore overruled.
Applicant’s objection to the portions of the Examining Attorney’s Appeal Brief that
criticize the Kaplan Survey are also overruled. The criticisms are not a new basis for
refusal, but rather provide more specific reasons for the refusal, and expand on

12 Citations in this subsection are to the record in the ’872 Application.
13 The Examining Attorney “recognizes that the prosecution of the application has been
complicated,” leading to “misunderstandings.” 18 TTABVUE 8.
Serial Nos. 85792872 and 86689423
15
criticisms originally provided on remand. 13 TTABVUE.
2. Must Applicant Indicate That the Mark is Three-Dimensional?
The Examining Attorney argues that Applicant’s mark is three-dimensional and
that under Trademark Rule 2.52(b)(2) “the applicant must indicate that the mark is
three-dimensional.” Applicant suggests in its Appeal Brief, 15 TTABVUE 15, and
argued at the hearing, that its mark is “both” two-dimensional and three-
dimensional, and that its description of the mark is sufficient. According to Applicant,
consumers are able “to recognize a single mark across more than one dimension,” and
the Office has issued a number of registrations which “do not contain a ‘dimensional’
limitation,” and “can be and are represented both two- and three-dimensionally in
the specimens.” Id. at 16-17.
Before addressing the merits, we should note that the Examining Attorney’s
position has been inconsistent. In the April 8, 2013 Office Action, the original
examining attorney14 found that functional elements of the mark (the cup), were
depicted in solid lines in Applicant’s original drawing, and required a new drawing
depicting unclaimed portions of the mark in broken lines and describing it as three-
dimensional. Applicant complied in its October 7, 2013 Office Action response. The
original examining attorney then indicated that “the mark does not include a three
dimensional configuration of a cup as this portion of the mark is now depicted in
dotted lines.” Office Action of November 17, 2013. Applicant accordingly amended the

14 The current Examining Attorney assumed responsibility for this application from the
original examining attorney in 2016.
Serial Nos. 85792872 and 86689423
16
description of the mark, no longer indicating that it is three-dimensional.
In any event, we do not share the Examining Attorney’s apparent concern that
Applicant, by virtue of its drawing and description of its mark, is seeking registration
for “more than one mark.” As Applicant points out, there are numerous examples of
marks which can be depicted in two-dimensional or three-dimensional form, without
constituting “more than one mark,” as shown by a sampling of registrations which
are included in the table below:15
Mark, Goods and
Services, Reg. No.
and Cite
Description Specimen

coffee

Reg. No. 3175941, 11
TTABVUE 46-51
(Applicant’s mark)
“Color is not claimed as a feature
of the mark. The mark consists of
a series of five concentric circles in
the colors green, white, green,
white, and black placed centrally
on the exterior of a cup. The
portions of the drawing
represented by dotted lines are
not claimed as elements of the
mark.”

batteries

Reg. No. 3251144, 11
TTABVUE 52-5716
“Color is not claimed as a feature
of the mark. The mark consists of
two parallel bands, which extend
around an electronic battery. The
dashes used to form the outline of
a battery are not part of the mark,
but are used to show the position
of the mark on the goods.”

15 While “consistency” arguments based on the Office’s treatment of allegedly similar marks
are generally entitled to little weight, we find these examples useful in placing the
circumstances of this case in context.
16 This registration was cancelled for failure to renew after Applicant filed its appeal.
Serial Nos. 85792872 and 86689423
17

Entertainment
services in the form
of professional
football games

Reg. No. 972622, 11
TTABVUE 75-80
None (however, “no claim to
exclusive rights in a descriptive
representation of a football
helmet per se is made apart from
the mark as shown”)

Franchising
services, namely,
offering technical
assistance in the
establishment
and/or operation of
real estate
brokerage firms

real estate
brokerage services
“The color(s) red, white, and blue
is/are claimed as a feature of the
mark. The mark consists of a
rectangular background used on
real estate listing signs and in
images thereof containing three
horizontal, rectangular bars. The
top rectangular bar is red and
contains the words ‘For Sale’ in
white, and the bottom rectangular
bar is blue. The middle bar is
white and is bounded on the left
and right by dotted lines to show
its location; the dotted lines are
not part of the mark and no claim
is made to such dotted lines.”
Serial Nos. 85792872 and 86689423
18

Reg. No. 1702048, 11
TTABVUE 81-86

This evidence makes clear that certain registered marks are sometimes depicted
two-dimensionally, and other times three-dimensionally. It is perhaps especially
noteworthy that in Registration No. 1702048, the description specifically indicates
that “the mark consists of a rectangular background used on real estate listing signs
and in images thereof containing three horizontal, rectangular bars” (emphasis
added).
Here, as described, the cup, or depiction of the cup, in each case having a centrally
located green dot, may identify the source of the coffee or restaurant and café services.
The mark remains the same, and it performs the same source-identifying function.
For all of these reasons, under the circumstances of this case we find that
Applicant’s description of its mark is acceptable.
3. Does the Drawing “Match” the Specimens?
The Examining Attorney argues that the mark in the drawing, featuring a solid
green circle, is not a “substantially exact” representation of the “siren” design mark
featured in some of the specimens. According to the Examining Attorney, “[f]ar from
being ‘minute’ or ‘inconsequential,’ the siren is a large and noticeable feature missing
from the drawing.” 18 TTABVUE 11. We agree.
In fact, the siren17 design creates the commercial impression of an

17 The siren is sometimes perceived as a mermaid. In fact, in its Registration No. 3428128,
Applicant’s description of the mark indicates that a “siren” is “a two-tailed mermaid.” Office
Serial Nos. 85792872 and 86689423
19
anthropomorphic female being or character with long hair, wearing a crown with a
star on top. While the background of the siren design is at least partially green and
at least partially circular, that is where the similarities between the siren design and
the mark in the drawing end. In short, no one would describe the siren design in two
of the specimens as a green circle, and no one would describe the green circle in the
drawing as a siren, or for that matter any other type of anthropomorphic being or
character.
Applicant points out that its Registration Nos. 3175941 (below on the left) and
307004218 (below on the right)

issued based on specimens which included the siren design rather than colored
circles. 15 TTABVUE 13; 11 TTABVUE 46-51. These two examples of inconsistent
treatment, one in a now-cancelled registration, standing alone, are unpersuasive.
“Neither the Trademark Examining Attorney nor the Board are bound to approve for
registration an Applicant’s mark based solely upon the registration of other
assertedly similar marks … having unique evidentiary records.” In re Datapipe, Inc.,
111 USPQ2d 1330, 1336 (TTAB 2014); see also In re Nett Designs, Inc., 236 F.3d 1339,
57 USPQ2d 1564, 1566 (Fed. Cir. 2001) (“The Board must decide each case on its own

Action response of October 7, 2013.
18 This registration was cancelled after Applicant filed its appeal.
Serial Nos. 85792872 and 86689423
20
merits … Even if some prior registrations had some characteristics similar to Nett
Designs’ application, the PTO’s allowance of such prior registrations does not bind
the Board or this court.”). There is apparently only one remaining example of the
Office determining that the siren design is a “substantially exact” representation of
a colored circle. Applicant’s consistency argument is therefore not persuasive on this
point, as the Office’s “practice” is not illustrated by a mere one or two examples.19 In
any event, we find that the solid green circle is not a “substantially exact”
representation of the siren design.20
To the extent this case implicates “mutilation,” we find that essential and integral
subject matter – the siren design in both the original specimen and the additional
specimen submitted on November 17, 2015 – is missing from Applicant’s drawing.
See TMEP § 807.12(d) (Oct. 2018). Of course, “applicant may apply to register any
element of a composite mark if that element, as shown in the record, presents a
separate and distinct commercial impression which indicates the source of applicant’s
goods or services and distinguishes applicant’s goods or services from those of others.”
In re Miller Sports, 51 USPQ2d at 1060-61 (citing, e.g., In re Chemical Dynamics, 5

19 These rare examples contrast to the Office’s repeated allowance of (at times) three-dimen-
sional marks which are not so described in the application/registration, which Applicant es-
tablished with numerous examples, a mere sampling of which is discussed above.
20 Even if Applicant is correct that when viewed at a distance the siren design will be
perceived as a green circle, Applicant has not established that any particular viewing
distance is part of the “substantially exact” test. In comparing the drawing to the siren design
mark in the specimens, we have considered the various viewing distances from which typical
coffee and restaurant service consumers would likely encounter Applicant’s marks. At those
distances, consumers would perceive the siren as a siren and the solid green circle as a solid
green circle.
Serial Nos. 85792872 and 86689423
21
USPQ2d at 1828 and Institut National des Appellations D’Origine v. Vintners Int’l
Co., Inc., 958 F.2d 1574, 22 USPQ2d 1190 (Fed. Cir. 1992)). Here, however, the green
circle in Applicant’s drawing does not present a “separate and distinct commercial
impression” from the siren design.
To the contrary, as explained above, consumers viewing the siren design will
perceive an anthropomorphic female rather than any type of green circle. In other
words, it is not a composite mark consisting of a siren design and a green circle (used
as background or otherwise), let alone a green circle which presents a separate and
distinct commercial impression from the siren. The green circle and siren design are
simply not “interrelated elements of a single unified design.” Cf. In re Chemical
Dynamics, 5 USPQ2d at 1830.
To the extent consumers perceive the outer boundary of the siren design as
circular, it is not a green circle. It is only partially green, and it is not even clear that
it is mostly green; it might be more white than green. Furthermore, the bottom of the
siren’s torso, and other features which reach the circle’s boundary, are white, so the
siren design is not even enclosed within a green circular outline, but is instead
depicted under a green half-circle at the top, with the bottom half-circular outline
being largely white.
Notwithstanding Applicant’s suggestion to the contrary, the green circle in the
drawing is anything but a “minor alteration” of the siren design mark in two of
Applicant’s specimens. See Schecter Bros., 182 USPQ at 695. Indeed, in In re Univ. of
Miami, 123 USPQ2d 1075, 1077 (TTAB 2017) we found the drawing on the left below
Serial Nos. 85792872 and 86689423
22
to be merely a minor alteration of the specimen on the right below (and similar
specimens)
.
Similarly, in In re Frankish Enters., 113 USPQ2d at 1964, we found the drawing on
the left below to be merely a minor alteration of the specimen on the right below (and
other specimens):

The difference between this case and those is obvious when we compare Applicant’s
drawing on the left below to Applicant’s siren design specimen(s) on the right below:

In contrast to Univ. of Miami and Frankish Enters., there are wholesale, rather than
merely minor, differences between the mark in Applicant’s drawing and the mark in
its siren specimens.
Turning to Applicant’s other specimens, which feature cups with solid green
Serial Nos. 85792872 and 86689423
23
circles, we disagree with the Examining Attorney’s finding that they are unacceptable
because “each does not include an actual cup as one might expect for a trade dress
specimen.” 18 TTABVUE 17. For the reasons stated above in reversing the refusal
based on Applicant’s failure to identify its mark as three-dimensional, the Examining
Attorney’s apparent requirement that the specimens “include an actual cup” is
misplaced. There is no evidence that consumers would perceive a difference between
the mark in Applicant’s drawing, and the mark Applicant’s remaining
specimens:
.
Each of the specimens includes a two-dimensional depiction of a white coffee cup with
a centrally-placed green circle on the front exterior. Neither the drawing, nor any of
the specimens, are actual cups. Whether the mark is placed on an actual three-
dimensional cup or is presented as a two-dimensional depiction of an actual cup is
Serial Nos. 85792872 and 86689423
24
irrelevant given Applicant’s identification of goods and services.21 Even if the two-
dimensional depictions of the cups in the specimens were replaced with actual coffee
cups, the commercial impressions would be no different.
Finally, to the extent the Examining Attorney argues that the specimens are
unacceptable because they do not show “three-dimensional features of the mark,”
TMEP § 1202.02(c)(iv), we disagree. Each specimen shows three-dimensional
features because in the first specimen, the lid reveals three-dimensionality, and in
the rest, the liquid and whipped cream contents of the cups reveal three-
dimensionality. That is, a lid would not rest on something in only two dimensions,
and nothing in two dimensions could hold liquid contents.
In short, we find that Applicant’s drawing is not a substantially exact
representation of the siren design mark specimens, but is a substantially exact
representation of the mark in the remaining specimens.
4. Do the Specimens Show Use of the Mark in Association With the
Services?
A service mark is “any word, name, symbol, or device, or any combination thereof
… [used] to identify and distinguish the services of one person … from the services of
others and to indicate the source of the services, even if that source is unknown.”
15 U.S.C. § 1127. A service mark must be “used in such a manner that it would be

21 We do not understand what the Examining Attorney means when he argues that “the cup
and circle designs shown in these specimens are not depictions of actual objects like the mark
in the drawing.” 18 TTABVUE 17. There is no evidence that Applicant’s drawing depicts an
“actual object” either; nor is there any evidence that the specimens do not depict an “actual
object,” i.e. a specific, existing coffee cup which could have served as a model for the artist
who drew the specimens. The point is that in both the specimens and the drawing, the mark’s
commercial impression is the same.
Serial Nos. 85792872 and 86689423
25
readily perceived as identifying” the services, which is “determined by examining the
specimens of record in the application.” In re Moody’s Investors Svc. Inc., 13 USPQ2d
2043, 2047 (TTAB 1989); see also In re Volvo Cars of North America Inc., 46 USPQ2d
1455, 1458 (TTAB 1998) (a mark “must be used in a manner calculated to project to
purchasers or potential purchasers a single source or origin” for the services, but mere
intent that it function as a mark is not sufficient); In re Duratech Industries Inc., 13
USPQ2d 2052 (TTAB 1989). “At a minimum, the specimen must show a direct
association between the services and the mark sought to be registered.” In re
Osmotica Holdings Corp., 95 USPQ2d 1666, 1668 (TTAB 2010); see also, In re
Advertising & Marketing Development, 2 USPQ2d at 2014. That is, “[a] specimen that
shows only the mark with no reference to, or association with, the services does not
show service mark usage.” In re DSM Pharm., Inc., 87 USPQ2d 1623, 1624 (TTAB
2008).
Here, the Examining Attorney argues that none of Applicant’s specimens “show
the use of the mark in the drawing displayed in the sale or advertising of the services
or the rendering of the services ….” 18 TTABVUE 24. Applicant counters that its
specimens show the mark in connection with its rendering of the services.
Because we have found that Applicant’s drawing is not a “substantially exact”
representation of the siren design specimens, we focus on the specimens featuring a
solid green circle. Applicant refers to these three specimens as “point-of-sale signage”:
Serial Nos. 85792872 and 86689423
26
.
and to this specimen as “retail store window signage”:
.
Office Action response of and substitute specimens submitted on February 10, 2015.
The “point of sale” signage is displayed in the rendering of the services, i.e. at the
counter where customers place orders and cashiers accept payment for Applicant’s
“coffee bar,” “carry-out restaurant” and other services.22 The “retail store window”
signage advertises Applicant’s services.
In considering whether Applicant’s specimens show use of the green circle mark
in connection with Applicant’s services, our review is not limited to the specimens

22 A presumably typical point of sale for Applicant’s services is depicted in the final siren
design specimen submitted with the Office Action response of November 17, 2015.
Serial Nos. 85792872 and 86689423
27
themselves. “Both precedent and examination guidance make clear that in assessing
the specimens, consideration must be given not only to the information provided by
the specimen itself, but also to any explanations offered by Applicant clarifying the
nature, content, or context of use of the specimen that are consistent with what the
specimen itself shows.” In re Pitney Bowes, Inc., 125 USPQ2d 1417, 1420 (TTAB 2018)
(citing DSM Pharm., 87 USPQ2d at 1626 (“In determining whether a specimen is
acceptable evidence of service mark use, we may consider applicant’s explanations as
to how the specimen is used, along with any other available evidence in the record
that shows how the mark is actually used.”)).
The advertising specimen displayed in retail store windows does not refer to the
coffee bar, carryout or restaurant services provided inside Applicant’s retail stores. It
is therefore unacceptable. In fact, “[f]or advertisement specimens … ‘[i]n order to
create the required ‘direct association,’ the specimen must not only contain a
reference to the service, but also the mark must be used on the specimen to identify
the service and its source.’” Pitney Bowes, 125 USPQ2d at 1419 (quoting Osmotica
Holdings, 95 USPQ2d at 1668); see also In re Adair, 45 USPQ2d 1211, 1214-15 (TTAB
1997); In re Metriplex, Inc., 23 USPQ2d 1315, 1317 (TTAB 1992). Here, the cup with
a green circle in the advertising specimen does not create a “direct association” with
Applicant’s services, as it does not refer to or identify them. Rather, consumers would
likely view the specimen as akin to a menu listing that promotes and displays the
coffee and oatmeal, rather than Applicant’s service of providing the coffee and
oatmeal to the customer. See In re El Torito Rest. Inc., 9 USPQ2d 2002 (TTAB 1988).
Serial Nos. 85792872 and 86689423
28
The situation is different, however, with respect to the point-of-sale signage
specimens. In fact, a specimen need not explicitly refer to the services if it shows the
mark used in “rendering” the services. In re WAY Media, Inc., 118 USPQ2d 1697,
1698 (TTAB 2016) (citing In re Metriplex, 23 USPQ2d at 1316-17 and In re Johnson
Controls, 33 USPQ2d 1318, 1320 (TTAB 1994)). Nevertheless, there must still be
“something which creates in the mind of the purchaser an association between the
mark and the service activity.” In re Johnson Controls, 33 USPQ2d at 1320.
We find that Applicant’s point-of-sale specimens display Applicant’s mark in the
rendering of the identified services.23 Indeed, in In re Metriplex, the specimens
showed “the mark as it appears on a computer terminal in the course of applicant’s
rendering of the service,” and were found acceptable. 23 USPQ2d at 1316-17. This
case is analogous, as Applicant’s mark appears on the counter where Applicant’s
services are requested, provided and purchased.
The Examining Attorney’s reliance on El Torito is misplaced. There, “table tents”
bearing the mark were placed on restaurant tables, and the table tents and mark
identified only food items. In large part because there was “no evidence that the mark
is used in the promotion of applicant’s services,” we found that the mark only
identified food items. 9 USPQ2d at 2004. Similarly, in In re Brown & Portillo Inc., 5
USPQ2d 1381, 1384 (TTAB 1987), we found, based on “the only evidence of record,”
that the mark in question identified only food products. In doing so, however, we

23 This is a separate inquiry from whether Applicant’s mark has acquired distinctiveness,
which we address below. In this section we assess only whether Applicant’s specimens show
use of its mark in association with its services.
Serial Nos. 85792872 and 86689423
29
discussed our similar decision affirming a refusal to register in In re McDonald’s
Corp., 230 USPQ 210 (TTAB 1986), another case involving a mark used only for food
items. Our decision in McDonald’s was ultimately reversed by the Federal Circuit in
an unpublished decision, however. In Brown & Portillo we distinguished the Federal
Circuit’s decision in McDonald’s, pointing out that although the specimens in
McDonald’s identified only food products, there was also an “extensive” record in that
case of additional material “consisting of a variety of advertisements, television
storyboards and an affidavit of the vice president of McDonald’s who indicated, in
part, that the purpose of the advertising of McDonald’s is to promote not only the sale
of breakfast food sandwiches, but also to promote the restaurant services in order to
increase sales and promote the image of McDonald’s.” In re Brown & Portillo, 5
USPQ2d at 1383-84. In this case, in contrast to Brown & Portillo and El Torito, but
much like the Federal Circuit’s McDonald’s decision, Applicant has established that
a coffee cup with a green circle is extensively used to identify Applicant’s services.
For example, the following, which displays the mark in question, was used in a
joint promotion between Applicant and Target launched in 2014:

Declaration of Russ Jacobs, Applicant’s Director, Corporate Counsel, submitted with
Office Action response of February 10, 2015 (“Jacobs Dec.”) ¶ 2. Applicant used the
Serial Nos. 85792872 and 86689423
30
mark in this billboard advertising, also from 2014:

Id. ¶ 5. Applicant uses the mark with its mobile app and gift cards, as well as video
advertisements:

Supplemental Declaration of Katherine Seawell, Applicant’s Vice President, Espresso
& Brewed (“Supp. Seawell Dec.”), submitted with Applicant’s Office Action response
of May 15, 2014 ¶¶ 14-15. There is also evidence that the coffee cup with a green
circle is recognized as a mark for Applicant’s restaurant and café services. For
example, a newspaper article, discussing Applicant’s service of coffee, states: “I can
remember a flash of pleasure when I drove into a small town … and saw that familiar
green circle.” Id. at ¶ 11. The cup with a green circle has been featured on the landing
Serial Nos. 85792872 and 86689423
31
page for customers accessing Applicant’s public wifi network:

Second Supplemental Declaration of Katherine Jean Seawell (“Second Supp. Seawell
Dec.”), submitted with Applicant’s November 17, 2015 Request for Reconsideration
¶ 12. Perhaps most importantly, Applicant has used the cup with green circle mark
on its menu boards, not to identify menu items, but instead to identify Applicant’s
restaurant and café services, in over 10,000 locations:

Id. ¶ 15.
This evidence differentiates this case from Brown & Portillo and El Torito, and
brings it closer to the Federal Circuit’s decision in McDonald’s. That is, after
considering Applicant’s explanations and evidence “clarifying the nature, content, or
context of use of the specimen,” In re Pitney Bowes, 125 USPQ2d at 1420, including
Applicant’s other uses of its mark inside and outside its retail stores, online and
Serial Nos. 85792872 and 86689423
32
otherwise, we find that Applicant’s point-of-sale specimens submitted on February
10, 2015 establish a direct association between Applicant’s services and the mark in
Applicant’s drawing. The refusal of registration based on the specimens not showing
use of the mark for Applicant’s services is therefore also reversed.
5. Has Applicant’s Mark Acquired Distinctiveness?
Because Applicant has conceded that the mark in the ’872 Application is not
inherently distinctive, it must establish that the mark has acquired distinctiveness.
The Examining Attorney concedes that Applicant is a “commercially successful
enterprise,” in fact a “juggernaut.” 18 TTABVUE 20. He argues, however, that
Applicant’s evidence of widespread general commercial success does not necessarily
imply public recognition of the specific mark in question, for the specific goods and
services identified in the application. Id. He also takes issue with various aspects of
the “format and/or the method” of the Kaplan Survey, and concludes that Applicant
has not established acquired distinctiveness.
Applicant suggests, for the most part without supporting evidence, that its
secondary meaning evidence primarily concerns the mark for which it seeks
registration. It also relies heavily on the Kaplan Survey, pointing out that the
percentage of respondents identifying Applicant as the source of goods sold under the
applied-for mark is greater than is typically required to establish acquired
distinctiveness.
Applicant bears the burden of establishing acquired distinctiveness. Real Foods
Pty Ltd. v. Frito-Lay N. Am., Inc., 906 F.3d 965, 128 USPQ2d 1370, 1374 (Fed. Cir.
2018); Yamaha Int’l Corp. v. Hoshino Gakki Co., 840 F.2d 1572, 6 USPQ2d 1001,
Serial Nos. 85792872 and 86689423
33
1006-07 (Fed. Cir. 1988). In this case, its burden is quite high. Indeed, “[t]he kind and
amount of evidence necessary to establish that a mark has acquired distinctiveness
in relation to goods or services depends on the nature of the mark and the
circumstances surrounding the use of the mark in each case.” In re Chevron
Intellectual Property Grp. LLC, 96 USPQ2d 2026, 2030 (TTAB 2010). Here,
Applicant’s mark ─ consisting of a common basic shape (a circle), in a particular color
(green), placed on a common basic object (a coffee cup) ─ is not inherently distinctive,
as Applicant conceded at the oral hearing. It is not the type of mark likely to be
perceived as a source indicator. See e.g. Yamaha Int’l, 6 USPQ2d at 1008; Seabrook
Foods, Inc. v. Bar-Well Foods Ltd., 568 F.2d 1342, 196 USPQ 289 (CCPA 1977); In re
Serial Podcast, LLC, 126 USPQ2d 1061, 1073 (TTAB 2018); In re Lorillard Licensing
Co., 99 USPQ2d 1312, 1316-17 (TTAB 2011); In re Chevron, 96 USPQ2d at 2030; In
re American Academy, 64 USPQ2d at 1753 (“However, ordinary geometric shapes
such as circles, ovals, squares, stars, etc., are generally regarded as nondistinctive
and protectable only upon proof of acquired distinctiveness.”); Guess? Inc. v.
Nationwide Time Inc., 16 USPQ2d 1804, 1805 (TTAB 1990); J.T. McCarthy,
MCCARTHY ON TRADEMARKS AND UNFAIR COMPETITION, § 7:33 (5th ed. Nov. 2018
update) (“Ordinary geometric shapes such as circles, ovals, squares, etc., even when
not used as a background for other marks, are regarded as nondistinctive and
protectable only upon proof of secondary meaning.”) (emphasis added).
Moreover, “[t]he fact that applicant’s [circle] is green does not change the standard
by which these types of marks are judged.” In re Benetton Grp. S.p.A., 48 USPQ2d
Serial Nos. 85792872 and 86689423
34
1214, 1216-17 (TTAB 1998). Nor does it matter that Applicant’s mark is
unaccompanied by words or other indicia of origin.
Applicant has provided no authority for the proposition
that simple geometric shapes (such as circles, squares and
equilateral triangles) should, when used by themselves, be
regarded as inherently distinctive. It simply makes no
sense to say that an equilateral triangle, when used in
association with a word, is inherently non-distinctive, and
then to say that an equilateral triangle, when used by
itself, is inherently distinctive.

Guess? Inc., 16 USPQ2d at 1805.
a. Sales, Advertising and Unsolicited Media Attention
Applicant relies on testimony about its revenues and advertising expenditures, as
well as unsolicited media attention relating to its marks. The problem, however, is
that for the most part it is not clear that this evidence relates to the mark in the ’872
Application.
Applicant has taken the position, throughout prosecution, that the drawing of its
mark is a “substantially exact” representation of the siren design
, a position with which we disagree, as indicated above. As a result,
when Applicant’s testimony refers to Applicant’s “mark,” or “logo,” we cannot
determine if the reference is to the green circle design shown in the drawing, or to
the siren design depicted in some of the specimens, because Applicant views these as
essentially the same. To the extent that Applicant’s secondary meaning evidence
Serial Nos. 85792872 and 86689423
35
relates to the siren design ─ and that extent is unknown ─ it has at best minimal
probative value to our determination regarding whether the green circle mark in the
drawing has acquired distinctiveness. While the cups bearing the siren design share
some features with the cups bearing the green circle design, the evidence is too
uncertain and unreliable to warrant the conclusion that consumer association of one
design with Applicant carries over to the other.
Applicant’s conflation of evidence pertaining to the siren design with evidence
pertaining to the green circle design is obvious from the prosecution history.
Applicant originally depicted the mark in the ‘872 Application as the siren design and
referred to it as the “Starbucks Cup Mark:”

Declaration of Katherine Jean Seawell (“Seawell Dec.”) submitted with Applicant’s
Office Action response of October 7, 2013 ¶ 3. This discrepancy is perhaps best
illustrated by Applicant’s testimony that customers bought hundreds of millions (and
in 2012 over one billion) beverages served in cups bearing the “Starbucks Cup Mark.”
Seawell Dec. ¶ 11. At the oral hearing, however, Applicant conceded that it has not
Serial Nos. 85792872 and 86689423
36
actually sold any beverages in cups bearing the green circle design. Rather, it sells
beverages in cups bearing the siren design. Thus, the astronomical figures in the
hundreds of millions, and billions, which Applicant submitted during prosecution,
apparently do not pertain to the mark in question. We have therefore focused on
Applicant’s secondary meaning evidence that clearly refers to the mark in question,
the mark in the ‘872 Application, the green circle design mark.
We acknowledge that after its original Office Action response, Applicant’s
declarations started defining the mark in the ‘872 Application as the “Green Circle
Design,” and seemed to be referring to and providing evidence about that applied-for
mark, at least in part. However, the extent to which this more recent testimony and
evidence is reliable and relevant remains unclear. For example, in her original
declaration, Ms. Seawell testified that “Customers in the United States of America
have purchased the Beverages bearing the Starbucks Cup Mark [defined as the siren
design mark as set forth above] from STARBUCKS stores, both company-owned and
licensed locations, in the following quantities”:

Seawell Dec. ¶ 11. In her next declaration, which was clearly referring to the green
circle design rather than the siren design, Ms. Seawell testified that “Customers in
the U.S. have purchased over 3 billion Beverages in white cups bearing the Green
Circle Design from STARBUCKS stores, both company-owned and licensed
locations,” in the following amounts:
Serial Nos. 85792872 and 86689423
37

Supp. Seawell Dec. ¶ 7. While it would appear from Ms. Seawell’s declarations that
Applicant sold beverages in billions of cups bearing the siren design, and billions of
cups bearing the green circle design, Applicant specifically denied at the oral hearing
that it sold any beverages in cups bearing the green circle design. Therefore, the
figures in the more recent Seawell Declarations appear to be incorrect. Furthermore,
in her next declaration, Ms. Seawell changed her testimony. Specifically, she
indicated that “Customers in the U.S. have purchased the Beverages in cups bearing
the Starbucks Cup Mark [defined in her original declaration as the siren design
mark] from STARBUCKS stores, both company-owned and licensed locations, in the
following quantities”:

Second Supp. Seawell Dec. ¶ 8. Because the number of siren design cups listed for
2011-2013 in this declaration is identical to the number of cups listed for those years
in the previous declaration, we infer that the numbers refer to the same type of cup,
whether the green circle design or the siren design, and that the discrepancy is merely
a typo or an oversight in how the mark is described. The problem, however, is that it
Serial Nos. 85792872 and 86689423
38
is not clear, from this and other inconsistencies in the declarations, which of the two
marks are being discussed, the siren design or the green circle design. Again,
Applicant effectively indicated at the oral hearing that the numbers refer to the siren
design rather than the green circle design.
Similarly, much of the unsolicited media attention upon which Applicant relies
refers to Applicant’s “green logo” (sometimes referred to as “familiar” or “iconic”), but
“green logo” could just as easily describe the siren design as the green circle design.
Supp. Seawell Dec. ¶ 11 and Ex. B; Second Supp. Seawell Dec. ¶ 9 and Ex. B. In the
same way, media references to Applicant’s “white-and-green coffee cups,” or the
equivalent, could just as easily refer to cups bearing the siren design mark as the
green circle design mark. It is therefore not clear whether these articles are referring
to the mark in question, or the siren design mark. We therefore now turn to only the
evidence that clearly refers to the green circle design mark rather than the siren
design mark.24
Applicant offers the “Starbucks Card,” described as a “loyalty card” that “allows
customers to pay for their items at company-owned and licensed STARBUCKS
stores,” which bears the green circle design:

24 While we have attempted to give Applicant “credit” for all documented uses of the green
circle design, some of the photographs submitted are too unclear, or too small, to determine
whether the mark on the cups is the siren design or the green circle design. See e.g. Supp.
Seawell Dec. ¶ 9.
Serial Nos. 85792872 and 86689423
39

Seawell Dec. ¶ 20 and Ex. L. Applicant has not indicated how many consumers’
loyalty cards bear the green circle design, however. Similarly, while Applicant has
used the green circle design on coffee packaging, such as the specimen submitted and
discussed above in connection with the ’423 Application, Applicant has not indicated
how much coffee bearing this mark has been sold. Applicant also uses the green circle
design mark on mobile apps, but does not indicate how many people downloaded or
use the apps:

Supp. Seawell Dec. ¶ 14. Applicant used the mark in connection with its “Tweet-a-
Coffee” campaign with LinkedIn “that alone generated almost $180,000 in sales in
several months”:

Serial Nos. 85792872 and 86689423
40
Id. ¶ 16.
Some of the unsolicited media attention Applicant introduced also appears to re-
late to the green circle design:
“Starbucks also serves pretty good coffee. I can remember
a flash of pleasure when I drove into a small town … and
saw that familiar green circle.” Jon Carroll, We Cheer the
March of the Peet’s, San Francisco Chronicle, Dec. 15,
2009. Id. ¶ 11 and Ex. B.

“Often consumers favor a coffee retailer based on the brand
of the cup, such as the green circle emblem at Starbucks,
he said.” Jenny Kincaid Boone, The Roanoke Times, Oct. 7,
2007. Id.

However, Applicant did not provide circulation figures or other evidence from which
we can determine the extent of consumer exposure to these articles.
b. The Kaplan Survey
The Kaplan25 Survey was intended “to measure the degree to which members of
the relevant population associate the mark/design in question [the mark in the ‘872
Application] with one or more than one source.” Office Action response of May 15,
2014 (Kaplan Survey at 5). Dr. Kaplan identified the “relevant population” as
“individuals 18 and above who in the past 30 days had purchased, for takeout either

25 We find that Dr. Kaplan is qualified to provide an expert opinion. He has a Masters Degree
from Purdue University, for which he did coursework in industrial psychology, consumer
behavior and psychological measurement; a Ph.D from Purdue for which he did coursework
in industrial psychology and social research methods; and an MBA from The Wharton School.
He has over 40 years of professional experience in consumer and marketing research,
including surveys for litigation, and has published extensively on consumer psychology and
marketing. He has qualified as an expert in trademark litigation. Office Action response of
May 15, 2014 (Kaplan Survey Ex. A).

Serial Nos. 85792872 and 86689423
41
for themselves or for someone else, coffee, tea, a coffee or tea-based beverage, or cocoa
or hot chocolate ─ or were likely to do so in the next 30 days.” Id. at 6. The survey
was conducted via the Internet, with Dr. Kaplan pointing out, based on a third-party
survey, that approximately 85% of American adults use the Internet and opining that
“nothing about Internet usage should adversely impact the representativeness of” the
surveyed relevant population. Id. Potential survey participants were “randomly
drawn” such that the selection was “unbiased” and “representative of households in
the 50 United States.” Id. at 7.
Potential participants were first asked “screening” questions to ensure the sample
was representative of the relevant population, geographically, and by age and gender.
Id. at 8. Respondents who worked, or lived with someone who worked, in the
advertising, marketing research, food or beverage or legal industries were excluded.
Id. at 10. Respondents who accessed the survey on a cellular phone or any device
besides a desktop, laptop or tablet computer were also excluded. Id. The survey was
“double blind,” meaning that participants did not know “the identity of the client, why
the research was being conducted, or that it might be used in a legal proceeding,” and
“[c]are was taken throughout the design of the study, to not use leading or suggestive
wording.” Id. at 13. Ultimately, those selected based on their responses to the
screening questions proceeded to answer the main questionnaire.
The main questionnaire displayed “a cup for coffee, tea, or cocoa,” and participants
were asked to “look at it as you would if you were given the cup or you saw a person
carrying or drinking from it.” Id. at 11. Some participants, those in one of the “test”
Serial Nos. 85792872 and 86689423
42
cells, were shown the mark in the drawing of the ’872 Application.26 Others, those in
the “control” cell, were shown a “generic light blue cup”:

Id. at 12. The purpose of a control such as this is “to estimate and correct for error in
the test cell results. This error can have many sources including respondent guessing,
etc.” Id.
Participants who responded that they associate the cup shown with one company
were then asked:
“What, if anything, about this cup makes you say that?”;

“And if you have an opinion, what company do you
associate this cup with?”;

“If you have an opinion, what brand name or names, or
product name or names, if any, are used by that
company?”; and

“If there is anything else you can think of about that
company or that would help identify it please write that
here.”

Id. at 12-13.
According to the Kaplan Survey, 69% of participants shown the mark in the ’872

26 As indicated, there were two “test” cells, one which considered the mark in the ’872
Application, and another which considered another of Applicant’s marks which is displayed
on coffee cups (Application Serial No. 85792857).
Serial Nos. 85792872 and 86689423
43
Application associated it with one company. In the text of his report, Dr. Kaplan
indicated that “only about one in 11 associated the Control Design with one company
… When the percent associating the Control Design with one company is used to
correct for noise (by subtracting it from the percent associating each of the ‘test’
designs with one company),” 60% of survey participants associated the mark in the
’872 Application with one company. Id. at 15.27 As Applicant points out, results at
this level are typically found sufficient to establish that a mark has acquired
distinctiveness, all else being equal. 15 TTABVUE 18-19.
Here, however, we have questions about the survey’s probative value. One major
issue is the cup Dr. Kaplan chose to use as the “control.” Unlike the “test” cups, the
control cup has no markings or identifying features, other than perhaps its
unremarkable light blue color.28 Indeed, Dr. Kaplan himself refers to the “control”
cup as “generic.” Office Action response of May 5, 2014 (Kaplan Survey at 12).
Consumers presented with Dr. Kaplan’s chosen “generic” control cup would be
much less likely to identify it with one company than a cup bearing some type of
marking or other identifying feature, such as a colored geometric shape. Indeed, a
small percentage of participants associating the control cup with one company is the

27 Dr. Kaplan’s report is internally inconsistent on this point. Immediately below this
narrative discussion in the Kaplan report, a chart identified as “Table 1” appears which
purports to represent the “Associations with Each Cup.” Office Action response of May 5,
2014 (Kaplan Survey at 15). However, the chart is inconsistent with the narrative
immediately above which it purports to summarize. Specifically, the chart indicates that only
“4%,” rather than “one in 11,” associated the control cup with one company. This
inconsistency is unexplained and detracts from the survey’s probative value.
28 There is no evidence or basis upon which to believe that light blue is an identifying feature.
Indeed, all non-transparent cups will have an identifiable color.
Serial Nos. 85792872 and 86689423
44
exact result that would be expected when consumers are presented with a “generic”
cup. It would have been more logical, and reliable, for the “control” cell to be shown a
cup more analogous to Applicant’s mark. For example, the control cup would have
been more effective if it was beige (or even white) and featured, for example, a blue
triangle placed centrally on the front exterior side. By controlling for “noise” this way,
the survey would have more reliably revealed why participants identified a cup with
one company, and whether Applicant’s mark has truly acquired secondary meaning.
As designed, however, the survey is not convincing that consumers’ identification of
the test cup with one company was the result of the mark in the ’872 Application. J.T.
McCarthy, MCCARTHY ON TRADEMARKS AND UNFAIR COMPETITION, § 32:187 (“The
general principle for choosing an appropriate control is easily stated: It should share
as many characteristics with the experimental stimulus as possible, with the key
exception of the characteristic whose influence is being assessed.”) (quoting S.S.
Diamond, Control Foundations: Rationales and Approaches in Trademark and
Deceptive Advertising Surveys 210 (ABA, Eds. Diamond & Swann 2012)).
On the other hand, we acknowledge that 67% of those who associated the test cup
with one company “specifically associated” that cup with Applicant, while only 3% of
those who associated the control cup with one company specifically associated that
cup with Applicant. This persuades us that many in the survey’s sample (apparently
only slightly less than 50% of those in the test cell for the mark in the ’872
Application) associated the cup they were shown with Applicant. Nevertheless, this
evidence is less significant than it appears on its face, for several reasons.
Serial Nos. 85792872 and 86689423
45
First, the “data set” attached to the Kaplan Survey as Exhibit E appears to include
all survey participants, those in the control group and both test groups. One of the
test groups was shown Applicant’s related but uninvolved mark which includes not
only the green circle on the front but markings on the back of the cup:

If there is a way to distinguish the information in the data set provided by the test
cell for this uninvolved mark from the information provided by the test cell for the
involved mark in the ’872 Application, that is not explained or apparent from the
record. Thus, we do not know from this record which data provided is from survey
participants shown the mark in question, making it more difficult to assess the data
set and the survey as a whole.29
While Dr. Kaplan was apparently able to distinguish one test cell from another
and appears to have tallied the results for only those survey participants shown the
mark in question, i.e. the mark in the ’872 Application, we are unable to examine the
data set for only those survey participants in the relevant test cell. Instead, we can

29 Some participants identified Applicant as the “one company” identified by the cup because
of the “boxes,” “squares” or “markings” on “the back” of the cup, or the “tabs for requests.”
These responses seem to make clear that these participants were shown the uninvolved
mark, but the rest of the data set is unclear as to which mark or control was shown to which
participant.
Serial Nos. 85792872 and 86689423
46
assume only that by referring to Applicant they are from one of the two test cells.30
Moreover, some of the participants who specifically identified Applicant as the
“one company” identified by the cup mark apparently did so not because of the green
circle, but because of features unclaimed in or absent from the involved ’872
Application. For example, some participants indicated that the cup’s “shape” is the
reason they identified Applicant, but Applicant does not claim rights in the shape of
the cup in the ’872 Application. There is no indication that these or similarly
irrelevant results were excluded from Dr. Kaplan’s tally.
More importantly and commonly, a large fraction of participants who identified
Applicant as the source of one of the test cups did not do so in a way that indicates
that their identification was the result of the mark on the test cup shown to them.
While some participants who indicated that the cup was associated with one
company, and specifically identified Applicant, stated that the reason for associating
the cup with one company was the “green circle” or “green logo” or the equivalent,
others gave more ambiguous answers, such as “green equals starbucks” or “Color,
gradations.”
A large percentage of participants did not specify why they associated the cup
with Applicant. For example, many participants answered simply “Starbucks” or the
equivalent. Answers like these may be the result of the mark on the cup the

30 Even this assumption might not be safe. The record makes clear that Applicant is one of if
not the largest and most well-known coffee and café brands. It would only be natural for
consumers to identify Applicant more often than other less successful brands under these
circumstances.
Serial Nos. 85792872 and 86689423
47
participant was shown, but they may not. Similarly, other participants provided the
following answers, which do not suggest that the mark on the cup was the reason for
associating the cup with Applicant or one company:
“looks like it would be a logo brand for just 1 company;”

“looks very familiar;”

“I am so familiar with the cup since I drink it almost
every day;”

“Starbucks is a frequent stop, and is convenient;”

“Looks like a Starbucks cup;”

“I know whose it is” and “go there often;”

“It is from Starbucks. I spend a few hundred dollars a
month there so I know. I have been a gold member
since 2009” and “If you cut me I would bleed coffee from
starbucks;”

“design;”

“Design;”

“Market is well dominated by this brand;”

“Familiarity” and “Use to seeing;”

“Over priced” and “Good coffee everything else is not
good;”

“nothing;”

“I am familiar with this cup as I have been to this coffee
shop many time.;”

“N/A” and “no;”

“Idk” and “no;”

Serial Nos. 85792872 and 86689423
48
“none” and “nine;”

“coffee” and “starbuck;”

“The design is familiar to me;”

“Looks like that shape;”

“plain;”

“Looks like starbucks shape;”

“looks like a cup from coffee chain;”

“Starbucks!!!!!!!!!!!” and “Because they dominate
everything!!!!!!!!;”

“look the same;” and

“looks.”

Office Action response of May 15, 2014 (Kaplan Survey Ex. E).
We also question the survey’s chosen universe. By only questioning those who
purchased relevant beverages in the last 30 days, or plan to do so in the next 30 days,
the survey is underinclusive. More occasional consumers are also relevant, and less
likely to be familiar with the players in the industry or their cups or trademarks. In
fact, if the survey was conducted of only daily purchasers of takeout coffee, the
percentage associating the test cups with one company or a specific company would
probably be higher, while if the survey was conducted of only biannual purchasers of
takeout coffee, the percentages would likely be lower. Here, there is no reason
provided for selecting the specific universe chosen. As we held in an analogous
situation, involving sandwiches rather than coffee/beverages:
… the universe of respondents is too narrow. The
Serial Nos. 85792872 and 86689423
49
description of goods in the application is “sandwiches,
excluding hot dogs.” As indicated above, the relevant public
comprises ordinary consumers who purchase and eat
sandwiches. Relevant consumers are not limited to recent
and frequent patrons of fast food restaurants and sandwich
shops; they include inter alia patrons of eat-in and sit-
down restaurants, delicatessens, and supermarkets that
sell sandwiches. Because the application does not limit the
channels of trade or potential customers for applicant's
sandwiches, we must presume that they are sold in all
normal channels of trade and to all of the usual customers
for such goods. Octocom Sys. Inc. v. Houston Computers
Servs. Inc., 918 F.2d 937, 16 USPQ2d 1783, 1787 (Fed. Cir.
1990); In re Elbaum, 211 USPQ 639, 640 (TTAB 1981)
(citing Kalart Co., Inc. v. Camera-Mart, Inc., 258 F.2d 956,
119 USPQ 139 (CCPA 1958)). Accordingly, the survey
results may be skewed because the survey did not include
all of the potentially relevant consumers.

Sheetz of Del., Inc. v. Doctor’s Assocs. Inc., 108 USPQ2d 1341, 1361 (TTAB 2013).
Here, as in Sheetz, Applicant’s identification of goods and services is broad, and
includes coffee, cafeterias and restaurants, and these goods and services may be
purchased by frequent or infrequent consumers. Moreover, as in Sheetz, limiting the
universe, in this case to those purchasing takeout coffee (or tea, cocoa or hot
chocolate), is underinclusive because the record indicates that purchasers of
Applicant’s goods and services sometimes consume the products or receive the
services inside cafes or restaurants, including Applicant’s retail locations, or in their
own homes. Takeout coffee is but a portion of Applicant’s business.
Relatedly, while we recognize that Internet surveys are becoming increasingly
common, and do not suggest that they are invalid or too unreliable to be probative as
a general matter, in this case the survey method raises questions. For example, while
the Kaplan Survey indicates that “85% of American adults use the Internet,” this
Serial Nos. 85792872 and 86689423
50
represents a particular subset of the population as a whole, a subset that excludes
some purchasers of coffee, an inexpensive item. See J.T. McCarthy, MCCARTHY ON
TRADEMARKS AND UNFAIR COMPETITION, § 32:165.25 (“Potential drawbacks of an
internet survey are population bias if the target universe consists of low-income,
rural, or elderly persons, because those groups have Internet usage in lower
percentages than other groups.”). This problem is compounded here not only because
of the inexpensive and ubiquitous nature of the goods and services, which are enjoyed
by an exceedingly high and diverse fraction of the population as a whole, but because
the Kaplan Survey excluded participants who access the Internet on smart phones.
This exclusion exacerbated the survey’s underinclusivity.
When we consider the survey and the rest of the record in its entirety, we find
that Applicant has failed to establish that the mark in the ’872 Application has
acquired distinctiveness. Applicant’s siren design is not at issue here, only the green
circle design is. As indicated, and as Applicant admitted at the oral hearing,
Applicant does not sell coffee in cups bearing the green circle design. While it uses
the green circle design in a number of other ways, perhaps quite extensively, for the
most part Applicant has not quantified its use of the green circle design specifically.
Indeed, at times during prosecution of the ’872 Application, Applicant appears to have
conflated evidence about the siren design mark with evidence concerning the green
circle design, compounding the problem. Furthermore, while the Kaplan Survey has
some probative value, it also has problems, and raises a number of questions, all of
which lead us to question whether it overstates the relevant public’s association of
Serial Nos. 85792872 and 86689423
51
the mark in the ’872 Application with a single source. Indeed, Dr. Kaplan used an
inappropriate control, selected an underinclusive universe and designed a survey and
questions (and provided a data set) that do not reliably establish an association
between the mark in question and Applicant. While the result here might very well
be different if the survey and other evidence were more specifically targeted to
assessing the mark in the ’872 Application specifically, on this record, we find that
Applicant has not established that the mark has acquired distinctiveness. The refusal
to register the mark in the ’872 Application on the ground that it is not inherently
distinctive and has not acquired distinctiveness is therefore affirmed.31
III. Conclusion
The refusal to register the mark in the ’423 Application is reversed, because
Applicant’s drawing of that mark is a substantially exact representation of the mark
in the Package Specimen and the Counter Display Specimen.
The refusals to register the mark in the ’872 Application because it is not described
as being three-dimensional, because the drawing and specimens do not “match” and
because the specimens do not show use of the mark in connection with the identified
services are all reversed. The refusal to register the mark in the ’872 Application
because it is not inherently distinctive and has not acquired distinctiveness is
affirmed.

31 The ’423 Application included a Section 2(f) claim when it was originally filed. The
Examining Attorney never addressed the Section 2(f) claim, nor was it an issue on appeal.
Serial Nos. 85792872 and 86689423
52
Decision: The refusal to register the mark in Application Serial No. 86689423 is
reversed. The refusal to register the mark in Application Serial No. 85792872 is
affirmed.