From the Lab, Resources for Roasters

Monday, November 4, 2019

I've Got 99 Flavors and "Currywurst" Ain't One

The flavor wheel might be elegant in its simple design, but the story of how-it-came-to-be is far more interesting than might be described in the gracefully curving lines of its figure.

Enter any one of the innumerable cupping labs across the world and, besides the requisite sample roasters, towering stacks of cups and shiny spoons you’ll probably find another similarity. It might take you a second to notice it, hanging inconspicuously on most of those rooms’ walls, blending into the background. But, eventually, the tastefully muted colors of the coffee flavor wheel will catch your eye and, taking a step closer to read the fine print, you’ll see tiers of descriptions gracefully parceled into tiny pie-slices. 

The flavor wheel might be elegant in its simple design, but the story of how-it-came-to-be is far more interesting than might be described in the gracefully curving lines of its figure.  

Sensory Science Begins  

Sensory science is a relatively new field. It arose in the late 1930s when industrialization and urbanization moved people further away from their food sources. Simultaneously, as more families left farms for city work, agriculture consolidated into larger farms (aided by emerging farm technology that was fuelled by petroleum) that shipped food further than ever before. In large city centers, many producers competed for the business of countless residents. Some competition involved reducing costs and cutting prices (with some even going so far as to adulterate products with cheaper alternatives) while other competition involved the emerging fields of marketing and branding.  

Previously, groceries such as flour and crackers were typically weighed from large, unmarked bags of bulk goods. Since 1916 when the world’s first self-service grocery, Piggly Wiggly, opened in Memphis, Tennesse, brand recognition became increasingly important: individual brands needed a way to make themselves stand out. From this rat race the field of sensory science emerged as a way to determine not only what flavors and combinations most appealed to the human palate but also how to market new tastes and products to curious, yet hesitant, consumers.   

The first iteration of a flavor wheel was created in the 1950s at Arthur D. Little Inc. by chemists developing the first-recorded descriptive sensory tool, the Flavor Profile Method. These chemists envisioned the first visual representation of all that information in the form of a linear rainbow where standard colors (red, orange, yellow and so on) represent the most general descriptors (such as nutty or fruity) and the subcategories (such as hazelnut and citrus) are shown in variations of these standard colors (such as  carmine and canary yellow). In this way, the entire spectrum of flavors can be represented as a spectrum of colors.  

Later versions of visual depictions of flavors adopted the wheel shape, beginning in 1979 with the beer flavor wheel by Dr. Morten C. Meilgaard and progressing with Dr. Anne C. Noble’s wine flavor wheel at University of California, Davis (UCDavis).  In these, and in subsequent versions, the original conception of flavor as a spectrum of colors has persisted. Using color on the wheel adds a visual appeal to the image, represents the connections between specific flavors and provides an additional way to commit the descriptors to memory.  

The first coffee flavor wheel, The Coffee Taster’s Flavor Wheel, created by Ted Lingle, was released by the former Specialty Coffee Association of America, [which later merged with the European branch to create the Specialty Coffee Association (SCA)] in 1995.  

Interestingly, when Lingle designed his wheel, he took the idea of the color scheme even further and used it to represent the weight of the molecules each color was meant to represent. Those in the enzymatic category, which contain a lighter molecular weight than those in the dry distillation category, are depicted as bright yellows and light greens, whereas the dry distillation terms are shown in dark purples and deep blues.  

The Coffee Flavor Wheel in its Infancy  

When Lingle created the first coffee flavor wheel, he based it off his previously published The Coffee Cuppers Handbook (1986). Though the wheel does not explicitly refer to the handbook, the book undoubtably helps clarify the more abstruse terms on the wheel.  

In fact, many of the terms on the original wheel were highly technical such as “turpeny,” “cineolic,” and “piquant.” And certain, fairly common terms didn’t appear at all, the most notable of which may be “strawberry” and “blueberry,” which are commonly found in natural processed coffees.   

The connection between terms in Lingle’s wheel is also more theoretical: the definitions in the aroma categories are connected to the way the coffee was grown, how sugars were developed in roasting and bean fiber. For instance, the term “coffee blossom” (under the enzymatic category) refers to the development, maturity and terroir evident in the cup. Likewise, “maple syrup” (under the sugar browning category) refers to the development of sugar. “Cloves” (dry distillation) is a reflection of the bean fiber.  

Meanwhile, as the categories of enzymatic, sugar browning and dry distillation have many sub-terms, the taste category does not offer the wide-reaching descriptions that are regularly found in cuppers’ notes. The terms for common experiences of sweet, sour, salty and bitter are limited in Lingle’s wheel while more theoretical concepts of maturation and terroir offer far more descriptors.  

New Advances in Descriptions of Flavor  

The next installment of our story begins in 2016 when the SCA, in collaboration with World Coffee Research (WCR), set out to revolutionize sensorial descriptions by utilizing a rigorous scientific process. This was, the WCR pointed out, the first time anyone had re-envisioned the coffee flavor wheel in over 20 years and the first time in sensory science history that such a methodical and demanding scientific method had been used to determine and organize the flavors on the wheel.  

Previous flavor wheels and sensory lexicons had been created by industry experts and had reflected their specific experiences and priorities. The new flavor wheel from SCA and WCR put the scientific method to work in an effort to create a more-objective lexicon that would enable conversation across locations, levels of expertise and backgrounds. 

The SCA and WCR accomplished this by working with five trained sensory scientists at Kansas State University. These scientists spent over 150 hours analyzing 105 coffee samples that came from 14 different origins. Through their analysis, they created 110 terms to describe coffee flavors. Each term had a unique reference point that could be recreated from brand-specific materials such as Smucker’s blackberry jam or particular McCormick spices. A second team of scientists then verified their results.  

One potential (and revealing) limitation of the new project, however, was ‘points of reference.’ The reference brands that scientists used were almost entirely continental United States-specific brands, making it difficult for non-CONUS based coffee professionals to verify their results with the same products described in the lexicon.   

In a 2016 post from the WCR upon the release of the wheel, the writers acknowledged that the wheel was not truly global. However, they argued both for its continued relevancy and its potential to be widely utilized. They wrote, “While we aim for the lexicon to be a universal tool for coffee scientists and industry, in its present form, it is not truly global. The references used in the lexicon are only widely available in the Unites States, where the lexicon was developed and where World Coffee Research will do most of its sensory evaluations in the near future. We hope to adapt the lexicon for other places with locally available references and appropriate translations, given adequate funding and partnerships. But this limitation of the lexicon doesn’t mean it isn’t globally relevant or useful. The research being conducted using the lexicon will be used to study and improve coffee from every part of the world.” 

Organizing the New Wheel  

After the sensory scientists had compiled their list of 110 terms, UCDavis and the SCA brought together more sensory scientists and coffee industry experts to engage in a study of how to group those terms into categories that made the most sense. You can read an in-depth explanation of this study here. The end results of the study were nine primary groups that were broken down into more and more specific descriptors.  

The groups were then oriented in three concentric circles on the wheel representing three tiers of descriptors: general terms, umbrella terms and specific descriptions. The general terms make up the innermost circle while the specific descriptors are thinner segments on the outermost circle. The idea is that people can begin with a general term, such as “fruity” and progress to an umbrella term, like “tropical fruit” and then choose a specific description such as “papaya” or “pineapple.” 

One way the new wheel is distinct from the previous wheel by Lingle is that it does not include many taste descriptors commonly used to describe defective coffees. Whereas Lingle’s wheel devotes an extensive slice to flavor attributes caused by defects, the new wheel was developed by assessing only non-defective coffees. Such flavors as “musty,” “skunky” and “meaty brothy” are included in the wheel only because they were found in small amounts in the coffees that were evaluated. Other, more severe, defects, such as “medicinal” (which appears on Lingle’s wheel) do not appear. It simply wasn’t encountered when developing the lexicon! In the same 2016 post by the WCR mentioned above, the writers say that, “it’s possible the lexicon will expand in the future to provide more comprehensive coverage of attributes commonly considered to be defects.”  

Using the Wheel to Drive the Industry Forward  

In 2016, when the scientists and the coffee professionals gathered to cup 13 Colombian coffees, the coffee professionals identified 59 terms to describe the coffee but only four of those terms were used by multiple people. The sensory scientists identified 92 terms and every single term was used by more than one person. This demonstrates that while cupping (at the time) was useful in the coffee world, it was not scientifically rigorous in its ability to be translatable across the varied experiences of coffee tasters and drinkers. Thus, despite the minor limitations of the new wheel, most coffee professionals agree that the new wheel is a huge step forward. 

As Susan Wilcox, Coffee Quality Controller and Q Grader at Sucafina North America, said, “The wheel helps calibrate cuppers worldwide, encouraging a shared lexicon across cultural differences and personal preference. It allows supplier and buyer, roaster and consumer to effectively communicate. In partnership with the Q scoring system, the Wheel provides a framework from which we can distill the mystery and complexity of specialty coffee down into a more coherent, quality-driven product.”  

Danner Friedman, Specialty Trader at Sucafina North America, echoes this sentiment. “Using the same terminology to describe a coffee is key to buying and selling specialty coffee.  It takes the guesswork for the buyer out of purchasing a coffee without sampling it if they are calibrated with their seller." 

Across the supply chain, the accessibility of the wheel enables buy-in on creating quality coffee throughout the chain. For Juan Andres Gutierrez, Specialty Supply Chain Coordinator at Sucafina Colombia, using the descriptions on the wheel helps him connect with producers, demonstrating the importance of specific farming practices and helping them understand the value of their coffee. “With producers the wheel has helped, in my opinion, engage the producer in the development and improvement of their farming practices and processing methods,” he says. “It helps them understand how each and every step affects their final product. It sparks a curiosity in them when you describe their coffee with attributes and motivates them to keep on striving to reach that profile or add attributes to their coffee. I find that by describing the coffee, in the case of specialty producers, it makes the value of the coffee more tangible and understandable to them – they understand better why the roaster is paying a premium for that coffee.” 

The Future of the Coffee Wheel 

At the end of the day, we do have to acknowledge that the current flavor wheel is not perfect. There is always room for improvement based on further research as well as user feedback. Even as the wheel and lexicon continue to grow and improve, the wheel will continue to fuel conversation, thoughtfulness and research in the specialty coffee sector. We look forward to seeing the advancements, discoveries and debates that emerge from the industry’s use of the coffee flavor wheel and WCR flavor lexicon.