• And the Survey Says —

    by  • September 16, 2008 • trademark

    I had a great response to my survey; my thanks to everyone who participated and to those who drove traffic to the survey, particularly Marty at The Trademark Blog and John at The TTABlog. Here’s a link to the original post and here’s a pdf version of the blank survey questions, in case you missed it.


    The mark being tested was MOOSE TRACKS. 210 people answered the survey questions, 7 from outside the U.S. Here are the results for the U.S. in a snapshot:



    Term
    Percentage responding “brand”

    1.
    Cherry Garcia
    (brand)
    96.0%

    2.
    Blue Moon
    (generic)
    80.0%

    3.
    Founder’s Favorite
    (brand)
    70.0%

    4.
    Moose Tracks
    (test)
    64.2%

    5.
    Death by Chocolate
    (generic)
    61.5%

    6.
    Jamoca
    (brand)
    57.3%

    7.
    Rocky Road
    (generic)
    20.5%

    8.
    Cake Batter
    (brand)
    15.5%

    9.
    Vanilla
    (generic)
    1.5%

    10.
    Coffee
    (generic)
    1.0%


    The Story


    Sometimes you read a case thinking it’s a slam dunk and it comes out just the opposite of what you expected. I recently read Blue Bell Creameries, L.P. v. Denali Co., LLC, about the ice cream flavor “Moose Tracks.” My immediate reaction was “but that’s a generic flavor!” and I was surprised that genericism wasn’t raised as a defense. Moose Tracks ice cream is, to me, vanilla ice cream with a chocolate ribbon and peanut butter-flavored inclusions; the name comes from what a moose track looks like (you might not want to go there). This photo from my camera phone is what I think of (“Moose Tracks” immediately below “Lobster Tracks”):


    The decision was only a standard run through the likelihood of confusion factors and the four part test for preliminary injunction, ending with an injunction against DJ plaintiff Blue Bell’s use of MOOO TRACKS. Not only was there no mention of genericism, but the court said “there is no meaningful dispute that Denali’s registered trademark ‘Moose Tracks’ is a strong mark.”

    So I asked my hairdresser:

    (Me): Have you ever had moose tracks ice cream?
    (Him): Look at me, what do you think?
    (Me): What’s in it?
    (Him): Oh, chocolate chips, lots of stuff, depends on which one. Have you ever had Snickers?
    (Me): No.
    (Him): It’s good.
    (Me): Who makes moose tracks?
    (Him): Lots of companies, you know, it’s like all those kinds of ice cream.

    I live in New England, where moose are a common theme in advertising and naming, so maybe my reaction was a regional thing. I did some more research. I found many web sites for ice cream shops that list Moose Tracks as an available ice cream flavor with no suggestion it might be anything other than a flavor; I found Moose Tracks ice cream made with “famous Moose Tracks fudge.” I found many ice cream shop web sites that use the ® symbol with no suggestion that anyone but the store owned the trademark, but I also found a fair number that mentioned Denali.


    Denali owns a number of trademark registrations for variations of MOOSE TRACKS for ice cream and fudge, which it licenses to a number of ice cream manufacturers. From Denali’s web site, by my count “Moose Tracks” ice cream is sold under at least 60 different brand names. Here are some licensees:






    You can see the Denali moose logo on each container (superimposed over the photo of the ice cream), although there is nothing similar about the font for the words “Moose Tracks” or the trade dress of the cartons. Nevertheless by all indications this was not a “naked license” situation; the decision said “Denali licenses its registered trademarks and provides the formulas, or recipes, to the licensees. Denali has control over which companies receive licenses . . . . Under its licensing agreements, Denali has control over the quality of the product by requiring the licensees to buy ingredients from authorized vendors. Denali has control over the way its ‘Moose Tracks’ trademark is presented, by requiring that licensees obtain its approval for advertising and carton design.”


    I had some reason to have assumed that Moose Tracks was just a flavor and not a brand, but was I the only one? So, with a HT to my trademark professor, I decided to do a Teflon survey.


    First the disclaimer – it’s not a particularly reliable survey. For starters, the surveyed population is made up largely of trademark professionals, hardly a representative population for a genericism survey (albeit a population with a good appetite for ice cream). I asked questions that may not be kosher for an evidentiary survey, but that I hoped might provide some insight into people’s responses. I did not qualify respondents and anyone could have stuffed the ballot box. I’m making assumptions based on some fairly low numbers of responses where the margin of error is far larger than the percentage swing. My interest is in the conversation, not finding the “right” answer.


    For my survey I wanted a range of choices in the distinctiveness spectrum, so I looked for ice cream flavors, both trademarks and generic, that had lesser or greater descriptiveness (the references I used in deciding whether a flavor is a trademark or generic are hyperlinked):

    Test word:
    Moose Tracks
    Ingredient description only Reference to ingredient with another word No reference to ingredient
    Generic

    Vanilla
    Coffee

    Death by Chocolate (Wikipedia,Google)

    Trademark

    (® Cold Stone Creamery)

    (® Ben & Jerry’s)
    (® Baskin Robbins)

    Founder’s Favorite (® Cold Stone Creamery)



    I would put them in roughly this order along the Abercrombie spectrum, highest to lowest (you may disagree):


    1. Blue Moon
    2. Rocky Road
    3. Moose Tracks
    4. Founder’s Favorite
    5. Cherry Garcia
    6. Jamoca
    7. Death by Chocolate
    8. Cake Batter
    9. Coffee
    10. Vanilla

    Now compare the flavors in order of brand recognition based on the survey results with their spots on the spectrum:




    Term
    Brand name or generic term
    Rank in distinctiveness spectrum
    1.
    Cherry Garcia
    brand
    5
    2.
    Blue Moon
    generic
    1
    3.
    Founder’s Favorite
    brand
    4
    4.
    Moose Tracks
    test
    3
    5.
    Death by Chocolate
    generic
    7
    6.
    Jamoca
    brand
    6
    7.
    Rocky Road
    generic
    2
    8.
    Cake Batter
    brand
    8
    9.
    Vanilla
    generic
    10
    10.
    Coffee
    generic
    9


    You can see that the little-known, regional generic flavor “Blue Moon” rated very high in brand recognition, demonstrating that we do indeed attribute a lot of trademark significance to the arbitrariness of an identifier. But commercial strength pushed Cherry Garcia to the top of the list, even though it has some descriptive character to it.

    But I also asked whether people had encountered the flavors. The ice cream flavors from least to most often encountered were:


    Term
    Percentage responding “never seen”
    1.
    Blue Moon
    85.1%
    2.
    Founder’s Favorite
    85.1%
    3.
    Jamoca
    53.6%
    4.
    Cake Batter
    41.5%
    5.
    Death by Chocolate
    37.8%
    6.
    Moose Tracks
    34.9%
    7.
    Cherry Garcia
    4.6%

    Coffee (tie)
    4.6%
    9.
    Rocky Road
    2.6%
    10.
    Vanilla
    .5%


    I thought that if someone had never actually seen a flavor, his or her response would be more about the inherent distinctiveness of the mark because the person was, after all, just guessing based on the words alone. A comparison of the responses of those who had seen a particular flavor to those that hadn’t might give me a different insight into the genericism of the mark.

    Comparing the difference in people’s responses based on this distinction showed some interesting results. First, the percentage of “don’t know” responses dropped; if people had seen it they had formed an understanding. Taking “Blue Moon” again, if people had actually seen the flavor their understanding that it was a generic flavor name increased fairly dramatically:


    Blue Moon (generic) Brand Common Don’t Know
    All responses 80.0% 9.5% 10.5%
    Had seen it (29 responses)
    69.0%
    (-11.0%)

    31.0%
    (+20.5%)

    0.0%


    The same pattern was true in the opposite direction for the brand Founder’s Favorite; for the most part it was the brand recognition that increased for those who had seen it:


    Founder’s Favorite (brand) Brand Common Don’t Know
    All responses 70.0%
    13.5%
    16.5%
    Had seen it (29 responses)
    79.3%
    (+9.3%)

    13.8%
    (+0.3%)

    6.9%


    Cake Batter showed the expected pattern for a brand; brand recognition increased and identification as “common” decreased among those who had seen it (although brand recognition is still very low):

    Cake Batter (brand) Brand Common Don’t Know
    All responses 15.5%
    77.0% 7.5%
    Had seen it (113 responses)
    17.7%
    (+2.2%)

    76.1%
    (-0.9%)

    6.2%


    Generic Death by Chocolate was as predicted; generally the recognition of it as a generic flavor increased:

    Death by Chocolate (generic) Brand Common Don’t Know
    All responses 61.5%
    29.0%
    9.5%
    Had seen it (120 responses)
    62.5%
    (+1.0%)

    33.3%
    (+4.3%)

    4.2%


    But two flavors didn’t behave as one might expect; first, the brand Jamoca appears to be losing recognition amongst those who had seen it:

    Jamoca (brand) Brand Common Don’t Know
    All responses 57.3%
    26.1%
    16.6%
    Had seen it (88 responses)
    56.8%
    (-0.5%)

    39.8%
    (+13.7%)

    3.4%



    Second, for Moose Tracks both identifications as brand and common increased:


    Moose Tracks (test) Brand Common Don’t Know
    All responses 64.2%
    22.9%
    12.9%
    Had seen it (127 responses)
    66.9%
    (+2.7%)

    27.6%
    (+4.7%)

    5.5%


    The next survey question might provide some clues why. I also asked people to name the company that makes each flavor. It was Denali’s practice of licensing a flavor name to different manufacturers that I found curious. I had no problem with the concept of a flavor as a brand indicator, but only if it was always associated with the same house brand, like Ben & Jerry’s (although of some interest, “Cherry Garcia” is actually a licensed mark, originally owned by Ben & Jerry’s but assigned in 1997 to the estate of Jerry Garcia). Can people recognize that, even though the ice cream itself was made by different companies, the flavor was from a single source?

    According to the Denali web site, using Ohio as an example, these brands of ice cream are sold in a “Moose Tracks” flavor at the listed stores:


    Ruggles, sold at Meijer, Giant Eagle, Discount Drug Mart, and Acme;
    Pierre’s, sold at Giant Eagle, Marc’s, Acme, Tops, Heinan’s, and Dave’s;
    Reiter, sold at Giant Eagle and Discount Drug Mart;
    Giant Eagle Creamery Classic, sold at Giant Eagle;
    Tofts, sold at Discount Drug Mart; and
    Meijer, sold at Meijer.


    So in a Meijer store you can buy two different brands of ice cream in Moose Tracks flavor, in a Giant Eagle store you can buy it in four brands, and so on. What ice cream brands did the survey show people associated with Moose Tracks ice cream?

    Ben & Jerry’s is the epitome of successful marketing and commercial strength. Of the 154 people who both identified Cherry Garcia as a brand and provided an answer on the “what company makes it” question, four people didn’t know, one person said Baskin Robbins, one said “t&j” and 148 said Ben & Jerry’s. In fact, people’s association of Ben & Jerry’s with ice cream names is so strong that for every flavor, vanilla included, someone named Ben & Jerry’s as the manufacturer.


    Contrast the result in Cherry Garcia with the generic but more distinctive Blue Moon. Of the 50 people who both identified Blue Moon as a brand and provided an answer on the “what company makes it” question, 44 couldn’t name a company, one said Hudsonville, one said Maggie Moo, three identified it as a beer (Marc, I think all yours), and one said the ever-present Ben & Jerry’s. So while the responses still tended to “brand” (69%), they actually couldn’t come up with a maker.

    “Moose Tracks” had the second highest number of responses to the “who” question after Cherry Garcia, with 58 answering both “brand” and responding on the “who” question. But unlike the universal association between “Cherry Garcia” and “Ben & Jerry’s,” 37 responses named 21 different companies as the manufacturer of the Moose Tracks flavor. “Moose Tracks” or “Denali” were listed six times, 9 different licensees (Alaskan Classics, Cascade, Gifford’s, Hershey’s, Kemp’s, Meijer, Pierre’s, Spartan, and Velvet) were listed 9 times, and 22 times people listed 11 companies that are not licensees and who do not make a Moose Tracks flavors (Baskin Robbins, Ben & Jerry’s, Blue Bell (oops, the DJ plaintiff), Blue Bunny, Breyer’s, Dryer’s, Eddie’s, Edy’s, Emack & Bolio’s, Friendly’s and Perry’s). Twenty people couldn’t name a brand, and one said “all.”

    One more statistic, I had asked about packaging. My prediction was that people who more often ate ice cream from an ice cream shop (like the top photo) would be less inclined to identify “Moose Tracks” as a brand. The results were the opposite:

    Moose Tracks (test) Brand Common Don’t Know
    Packaged (88 responses)
    63.6%
    29.6%
    6.8%
    Unpackaged (28 responses)
    78.6%

    21.4%

    0.0%



    We know that people don’t have to be able to name the company for a mark to have trademark significance, but is it relevant that when they can name names, they name so many different companies, including so many wrong companies? If a flavor has brand significance when I see it in association with one brand of ice cream, what happens when I go to the store down the street, or look in the next case over, and see another brand with the same flavor? With different trade dress? Is this why those who ate packaged ice cream had a lower “brand” response? The licensing relationship between Denali and its licensees seems to be escaping consumers, why? My ice cream shop photo above is of a licensee’s brand, but “Maine Lobster Tracks” is a registered trademark of the licensee Gifford’s, not Denali – how does that affect the brand significance? Should it matter in a trademark lawsuit whether the consumers are identifying a common company as the manufacturer or many different ones, or is it simply enough that consumers think it’s a brand? Is the identification of many manufacturers evidence that the flavor is going down the genericism road but hasn’t reached the legal identity of “generic” yet?


    Even though it was a question specifically about flavor, people overwhelmingly named an ice cream manufacturer in response. Can an ice cream flavor be a trademark and yet associated with many manufacturers? If one is going to use flavors as trademarks, do all flavors associated with the manufacturer have to be distinctive? Are there some kinds of marks, like a flavor, a slogan or distinctive packaging, that can function as a trademark when sold with a primary indicia, i.e., manufacturer brands, but can’t live alone? If so, where does that fit into our standard confusion analysis?

    Epilogue

    Thanks to the seven of you outside of the U.S. who responded. The only flavors the majority of respondents had seen were vanilla, coffee, Death by Chocolate and Cherry Garcia, and the flavors fell pretty much in the order of inherent distinctiveness (Founder’s Favorite at 89.7%, Blue Moon, Jamoca and Cherry Garcia tied at 71.4%, Death by Chocolate and Rocky Road tied at 57.1%, Moose Tracks and Cake Batter tied at 42.9%, and vanilla and coffee tied at 0.0%).

    A few random comments on the survey results. The majority of the responses to “how often do you eat ice cream” was “monthly,” yet about 3/4 of the respondents buy their ice cream in packaging – so hats off to those of you who can manage to buy ice cream at the grocery store and not eat it daily.

    Poor Cake Batter ice cream. A teaching moment – proof that you can’t turn a generic name into a trademark, even if you get it registered. And you gotta love trademark lawyers; I had several people make comment along the lines of “I know Cold Stone makes Cake Batter, but I still think it’s generic” (and they responded “common” on the survey).

    I’d love to hear from the person who commented “A very misleading survey.” How come?

    Thanks to all who helped on this exercise; comments enthusiastically invited. Let it rip.