You are on page 1of 14

Case 2:14-cv-06856-WHW-CLW Document 10-1 Filed 11/07/14 Page 1 of 14 PageID: 69

Bruce P. Keller (bpkeller@debevoise.com)


David H. Bernstein (dhbernstein@debevoise.com)*
Michael Potenza (mpotenza@debevoise.com)
Jared I. Kagan (jikagan@debevoise.com)*
DEBEVOISE & PLIMPTON LLP
919 Third Avenue
New York, New York 10022
(212) 909-6000
Attorneys for Conopco, Inc.
*admitted pro hac vice
UNITED STATES DISTRICT COURT
DISTRICT OF NEW JERSEY
----------------------------------------------------------------------- x
CONOPCO, INC., doing business as UNILEVER,
:
:
Plaintiff,
:
:
-against: 14 Civ. 06856 (WHW)(CLW)
:
DECLARATION OF
HAMPTON CREEK, INC.,
:
MICHAEL MAZIS
:
Defendant.
:
----------------------------------------------------------------------- x
I, Michael Mazis, declare as follows:
1.

I submit this declaration on personal knowledge in support of Unilevers

application for a preliminary injunction against Hampton Creek, Inc. (Hampton Creek).
2.

I am Professor Emeritus of Marketing at American Universitys Kogod

School of Business and have conducted marketing research surveys for over 30 years. I
was a faculty member at American University for 28 years, and I served over 10 years as
chair of the marketing department. I have taught courses in consumer behavior,

Case 2:14-cv-06856-WHW-CLW Document 10-1 Filed 11/07/14 Page 2 of 14 PageID: 70

marketing research, marketing principles, marketing management, and marketing and


public policy. I have testified in Federal Court many times as an expert on consumer
perception of advertising and product labels, including on behalf of the Federal Trade
Commission. Additional information about my professional background is provided
below at 29-30, and in my resume, which is attached as Exhibit A hereto. A list of the
cases in which I have testified in the last four years is attached as Exhibit B hereto.
NATURE OF RETENTION AND SUMMARY OF CONCLUSIONS
3.

I was engaged by counsel for Unilever, the manufacturer of Hellmanns

and Best Foods mayonnaise, to conduct a survey among purchasers of mayonnaise,


mayonnaise dressing, or imitation mayonnaise. The purpose of the research was to
determine consumers perceptions of Just Mayo, a product distributed by Hampton Creek
Foods, Inc. As I understand it, because Just Mayo is egg-free, it cannot be called
mayonnaise under the Food and Drug Administrations standard of identity for
mayonnaise, which specifies that egg yolks are a necessary ingredient of mayonnaise.
Accordingly, I was asked to investigate whether consumers perceive Just Mayo as real
mayonnaise, as a spread or dressing that resembles mayonnaise but is not real
mayonnaise, or as something else.
4.

To determine whether consumers mistakenly believe that Just Mayo is

mayonnaise, I designed and conducted a survey. The findings of the survey, described
herein, demonstrate that an overwhelming majority (53.7%) of the respondents who saw
the Just Mayo label and product mistakenly perceive the product to be real mayonnaise.

Case 2:14-cv-06856-WHW-CLW Document 10-1 Filed 11/07/14 Page 3 of 14 PageID: 71

Adding the word Dressing to the Just Mayo label (Just Mayo Dressing) only
marginally diminishes, but does not eliminate, the propensity of consumers to identify
Just Mayo as real mayonnaise (41.5%). Therefore, to determine how much of the
confusion was attributable to the use of the word mayo and/or by the use of the word
just which fails to explain that the product is not real mayonnaise, I tested two
additional labels. One label used the name Just Delicious Dressing instead of Just
Mayo (replacing the alleged false use of the term Mayo), but was otherwise identical to
the Just Mayo label. The other label prominently changed the name from Just Mayo to
NOT QUITE MAYO (replacing the alleged false use of the term Just).
5.

Only 5.8% of the consumers who saw the NOT QUITE MAYO label

mistakenly believed that it was real mayonnaise. Subtracting the 5.8% from this group
who said the product was real mayonnaise from the 53.7% in the group who viewed the
Just Mayo label and selected this answer, yields 47.9% who identified Just Mayo as real
mayonnaise.1
6.

Only 10.2% of the consumers who saw the Just Delicious Dressing label

mistakenly believed that it was real mayonnaise. Subtracting the 10.2% from this
group who said the product was real mayonnaise from the 53.7% in the group who
viewed the Just Mayo label and selected this answer, yields 43.5% who identified Just
Mayo as real mayonnaise.
1

It is well accepted in the survey field that the gross level of confusion in a test cell
should be adjusted to reflect mis-measurement error, or noise, much like a
pharmaceutical trial accounts for noise through a placebo test.

Case 2:14-cv-06856-WHW-CLW Document 10-1 Filed 11/07/14 Page 4 of 14 PageID: 72

7.

The difference of 43.5% to 47.9% between confusion in the group that

was shown the Just Mayo label and the groups that were shown the Just Delicious
Dressing label (i.e., no mayo) and the Not Quite Mayo label (i.e., no just)
demonstrates that a very substantial percentage of consumers (more than 40 percent) are
confused by the name Just Mayo.
CONSUMER SURVEY
Data Collection and Universe
8.

I designed an online survey to determine whether consumers who

purchase mayonnaise, mayonnaise dressing, or imitation mayonnaise believe that Just


Mayo is real mayonnaise or some other type of spread or dressing. Decision Analyst, a
national marketing research firm based in Arlington, Texas was responsible for providing
the sample, hosting the survey, and collecting data for the research. In addition, under
my direction, Decision Analyst tabulated the data.
9.

Data for the survey were collected from August 13, 2014 to August 20,

2014. Decision Analyst used its American Consumer Opinion (ACO) panel to
complete the study. The ACO panel is a worldwide, online panel managed by
Decision Analyst; it reaches more than eight million consumers in the United States and
overseas.
10.

A total of 822 consumers participated in the research; all were adults 18

years of age or older and residents of the United States. In addition, all respondents had

Case 2:14-cv-06856-WHW-CLW Document 10-1 Filed 11/07/14 Page 5 of 14 PageID: 73

purchased mayonnaise, mayonnaise dressing, or imitation mayonnaise in the past six


months and also planned to purchase these products in the next six months.
Study Design
11.

The survey was comprised of two sections: a Screener designed to identify

consumers who qualified for the research based on purchase behavior and other criteria;
and a Main Questionnaire designed to assess perceptions of the product based on viewing
an image of the bottle and the product label.
12.

Respondents were assigned randomly to one of four groups: a group in

which respondents viewed the Just Mayo label and product and three additional groups
that saw images of the product with a modified label. Modifications involved:

Adding the word Dressing to the Just Mayo product label to assess whether
including this term materially altered consumers perceptions.

Changing the product name to Not Quite Mayo to assess the impact of a more
accurate product name on consumers perceptions that does not include the
allegedly false word Just (which means exactly or simply).

Changing the product name to Just Delicious Dressing to gauge how removing
the word Mayo and substituting another phrase (Delicious Dressing) in the
product name affected consumer perceptions.

Each respondent viewed only one image of the bottle and product label.
Screener Questions
13.

The survey was distributed to members of an online panel that had agreed

to be available to take internet-based surveys. The first three screener questions were
designed to ensure that the individual taking the survey was, indeed, the panel member.
If the participants responses were not consistent with the panel database, the participant
was not permitted to continue with the survey. These initial questions asked for the

Case 2:14-cv-06856-WHW-CLW Document 10-1 Filed 11/07/14 Page 6 of 14 PageID: 74

respondents gender, state of residence and year of birth. A PDF version of the
programmed Screen and Main Questionnaire is attached as Exhibit C.
14.

The next screener question was designed to eliminate respondents who

either worked in, or had a household member working in, an industry that might have
some specialized knowledge about the issues tested in the survey, such as marketing
research, advertising, or a company that manufactured, distributed, or sold food products.
This is done to allow the survey to focus on actual consumers.
15.

Next, respondents were asked:

Please review the list below and indicate if you have purchased any of the
following products in the past 6 months or expect to make such a purchase in the
next 6 months?

Ketchup or catsup
Mayonnaise, mayonnaise dressing, or imitation mayonnaise
Yellow mustard or brown mustard
Peanut butter or hazelnut spread
Horseradish or hollandaise sauce
None of these

This step was taken to disguise the purpose of the research and ensure respondents were
purchasers of the product category of interest. To qualify, respondents had to have
purchased mayonnaise, mayonnaise dressing, or imitation mayonnaise in the past six
months and had to have expected to make another purchase of one of these products in
the next six months.
16.

Respondents were then asked whether they needed corrective eyewear; if

so, they were instructed to wear their corrective eyewear for the duration of the survey.
Those who refused were not permitted to continue taking the survey. Finally,

Case 2:14-cv-06856-WHW-CLW Document 10-1 Filed 11/07/14 Page 7 of 14 PageID: 75

respondents were instructed to take the survey in one sitting and not to confer with
anyone else or view any other materials while completing the survey. Again, refusal to
agree to this condition was grounds for elimination from the survey.
Main Questionnaire
17.

Before answering questions about the various product labels, all survey

respondents were presented with:

A 360 degree image of the product that they could turn to view the entire bottle.
In addition, they could use a magnifier to hone in on any section of the 3D image
and enlarge that section of the image.

A 2-dimensional, enlarged version of the product label to facilitate reading the


printed label.
18.

Throughout the survey, respondents were allowed to refer back to the

2-dimensional product label, to answer the remaining survey questions. Copies of the
four labels that were used are provided on the following page and are included in Exhibit
D.

Case 2:14-cv-06856-WHW-CLW Document 10-1 Filed 11/07/14 Page 8 of 14 PageID: 76

Just Mayo Label

Not Quite Mayo Label

Case 2:14-cv-06856-WHW-CLW Document 10-1 Filed 11/07/14 Page 9 of 14 PageID: 77

Just Delicious Dressing Label

Just Mayo Dressing Label

19.

The first open-ended question asked How would you describe the product

you just saw? Please be as detailed as possible. A follow-up, open-ended question asked
if respondents had anything to add to their initial description of the product.
20.

The next question was closed-ended and central to the research issue. It

was:

Case 2:14-cv-06856-WHW-CLW Document 10-1 Filed 11/07/14 Page 10 of 14 PageID: 78

Which of the following phrases best describes the product you just viewed?

Real mayonnaise
A spread or dressing that resembles mayonnaise but is not real
mayonnaise
Tartar sauce
White horseradish sauce
Hollandaise sauce
Something else
I dont know

To avoid order bias, the first five response options were presented in random order.
21.

The final question asked why the respondent selected a specific response

to the previous closed-ended question.


FINDINGS
Product Perception (Closed-Ended Question)
22.

The operative question was closed-ended and, as noted above, pointedly

asked respondents to select the phrase that best described the product they had seen.
Which of the following phrases best describes the product you just viewed?

23.

Real mayonnaise
A spread or dressing that resembles mayonnaise but is not real
mayonnaise
Tartar sauce
White horseradish sauce
Hollandaise sauce
Something else
I dont know
Respondents who viewed Hampton Creeks Just Mayo product were

significantly more likely than those who saw the Just Delicious Dressing and Not
Quite Mayo labels to say it was real mayonnaise. As shown in the table below, over

10

Case 2:14-cv-06856-WHW-CLW Document 10-1 Filed 11/07/14 Page 11 of 14 PageID: 79

half (53.7%) the respondents who viewed the Just Mayo product, offered this response.
The full data set that includes the results from the survey is provided as Exhibit E.
Q4. Phrase That Best Describes Product Viewed

Real mayonnaise
A spread or dressing that resembles
mayonnaise but is not real
mayonnaise
Tartar sauce
Horseradish sauce
Hollandaise sauce
Something else
I don't know
n= (total 822)
24.

Just
Mayo
53.7%

Just
Delicious
Dressing
10.2%

Not
Quite
Mayo
5.8%

Just
Mayo
Dressing
41.5%

45.9%

81.6%

90.3%

57.6%

0.0%
0.0%
0.0%
0.0%
0.5%
205

1.0%
0.5%
0.0%
5.3%
1.5%
206

0.0%
0.0%
0.0%
3.4%
0.5%
206

0.0%
0.0%
0.0%
0.5%
0.5%
205

Respondents who saw the product name, Not Quite Mayo, were least

likely to say the item was real mayonnaise only 5.8% selected this response. Nine in
ten respondents who viewed this label (90.3%) indicated the best description was a
spread or dressing that resembles mayonnaise but is not real mayonnaise, showing that a
prominent disclaimer that the product is not quite, rather than just, can be effective at
communicating to consumers the truthful nature of the product.
25.

The 5.8% who viewed the Not Quite Mayo label and who said the product

was real mayonnaise may be considered noise or guessing. Even when people were
told the product was not quite mayonnaise, a certain percentage (5.8%) still responded
that it was real mayonnaise, indicating that they were not as attentive as other
respondents to the name on the product label. Subtracting the 5.8% from this group who

11

Case 2:14-cv-06856-WHW-CLW Document 10-1 Filed 11/07/14 Page 12 of 14 PageID: 80

said the product was real mayonnaise from the 53.7% in the group who viewed the Just
Mayo label and selected this answer, yields 47.9% who identified Just Mayo as real
mayonnaise, after adjusting for noise or guessing attributable to the use of the word
just instead of not quite.
26.

The Just Mayo Dressing label tested whether adding the word Dressing

to Just Mayo would affect consumers perceptions of the product. That addition
marginally diminishes, but does not eliminate, the propensity of consumers to identify
Just Mayo as real mayonnaise (41.5%). I, therefore, tested the Just Delicious Dressing
label to determine the effect of completely eliminating the word mayo. In this group,
where Delicious Dressing was substituted for Mayo on the product label (Just
Delicious Dressing), 81.6% of respondents said the product was a spread or dressing
that resembles mayonnaise but is not real mayonnaise and only 10.2% said the product
was real mayonnaise. Subtracting the 10.2% from this group who said the product was
real mayonnaise from the 53.7% in the group who viewed the Just Mayo label and
selected this answer, yields 43.5% who identified Just Mayo as real mayonnaise. Thus,
I conclude removing the word mayo from the product name and replacing it with
Delicious Dressing substantially improves respondents understanding that the product
is a spread or dressing resembling mayo.
27.

To further assess the level of noise or guessing, I am currently testing an

additional label that replaces Just Mayo with Just Dressing, and will include the results
from that testing in a supplemental declaration to be submitted with Unilevers reply
brief.

12

Case 2:14-cv-06856-WHW-CLW Document 10-1 Filed 11/07/14 Page 13 of 14 PageID: 81

CONCLUSION
28.

Based on responses to the closed-ended question about the nature of the

product, it is apparent that a majority of consumers (53.7%) who saw the label Just Mayo
mistakenly believed the product was real mayonnaise. Product perception was
substantially altered only when a name was used that more accurately described the
product (Not Quite Mayo) or when the word mayo was changed entirely (Just
Delicious Dressing). In the group that viewed the Not Quite Mayo label, only 5.8% of
respondents said the product was real mayonnaise; among respondents who viewed the
Just Delicious Dressing label, 10.2% offered this response. Subtracting these levels of
confusion from the 53.7% level of confusion in the group that viewed the Just Mayo label
yields a difference of about 43.5% to 47.9%. These differences demonstrate that
consumers are confused by the name Just Mayo.
ADDITIONAL BACKGROUND INFORMATION
29.

I received my B.S. degree in Economics from the University of

Pennsylvania, my M.B.A. degree from New York University, and my Ph.D. degree in
Business Administration from Pennsylvania State University. I was editor of the Journal
of Public Policy & Marketing from 1992 to 1995, and I was Associate Editor of The
Journal of Consumer Affairs from 1998 to 2001.
30.

From 1976-79, I served as an in-house marketing expert at the Food and

Drug Administration (FDA) and at the Federal Trade Commission (FTC), where I
evaluated consumer perception of advertising and product labels and designed and

13

Case 2:14-cv-06856-WHW-CLW Document 10-1 Filed 11/07/14 Page 14 of 14 PageID: 82

You might also like