READING

The Science Behind the SCAA Flavor Wheel

The Science Behind the SCAA Flavor Wheel

flavor wheel specialty coffee association of america cafe cuppings taste sprudge

Four Barrel Coffee describes their Rwanda Simbi as “an old white Mustang convertible covered in bumper stickers with tangerine and nectarine peels under the passenger seat.” Unless every other roaster asks what Four Barrel is smoking and loads a bowl of it, there’s no way to speak the same language.

That’s where the Specialty Coffee Association of America’s Flavor Wheel comes in. Originally developed in rudimentary form in 1995, the SCAA released a new version at the beginning of the year that updates and codifies a lexicon of coffee flavors to help turn the subjective poetry of coffee descriptors into objectively agreeable prose.

flavor wheel specialty coffee association of america cafe cuppings taste sprudge

At this year’s SCAA Event, the association’s coffee science manager Emma Sage and World Coffee Research’s Hanna Neuschwander gave a deep dive into the science behind the new wheel. The end goal of the project is to be able to understand quality in an entirely different way than the cupping process, one that is quantifiable and beyond the palate discrepancies faced by even the most calibrated cuppers.

Despite the intense training that goes into cupping certification, an initial round of testing showed that a panel of sensory specialists actually out-tasted a panel of certified cuppers. Tasting a set of 13 Colombian coffees, the cuppers came up with 59 flavor terms, with only four terms universal to more than one cupper. The sensory panel arrived at 92 terms and were in 100% agreement. It proved that however valuable the cupping process, it doesn’t pass scientific muster.

flavor wheel specialty coffee association of america cafe cuppings taste sprudge

That set the groundwork for starting the real experiment. At Kansas State University, a panel of five trained sensory scientists spent 150 hours analyzing 105 coffee samples from 14 countries. In order to dodge the pitfalls of overly esoteric tasting notes, flavors were paired to specific reference points. For instance, “blackberry” wasn’t to be associated with berries plucked from the Argentinean hillside and eaten as an appetizer to a steak dinner paired with a reasonably priced Malbec, but rather Smuckers blackberry jam. The specific reference was kept on-hand such that tasters were able to gauge and quantify the comparative intensity of each flavor. They settled on 99 terms, which were confirmed by a second team of sensory scientists.

In order to find how these 99 terms could be oriented around a wheel, the SCAA and the University of California, Davis enlisted both sensory scientists and industry experts to cluster these terms into like groups. The result were nine overarching group terms (sweet, floral, fruity, etc.) that formed the interior of the wheel. Using a series of complicated graphs, each of these overarching groups was divided into more specific clusters that formed the second and third tiers of the wheel. To Four Barrel’s dismay, neither nectarine peels or bumper stickers made the cut.

flavor wheel specialty coffee association of america cafe cuppings taste sprudge

Although that seems like it uses a whole lot of science to take much of the fun out of describing coffee, the end game isn’t stifling poetic roasters. Codifying a flavor lexicon is just the first step. The goals of speaking the same language about flavor are to be able to better research coffee varieties and understand where quality comes from. It opens the doors for controlled tests to determine whether the chemistry inside certain varietals is responsible for specific flavors, or if environmental conditions have a more substantial impact.

Once the question of nature versus nurture is solved, it follows that this science could trickle down to the farm level to help farmers isolate specific flavor traits in their coffee plants, unlocking an entirely new approach to cultivating and sourcing. It has huge implications for the next few decades of the coffee industry, and before too long, roasters will actually be able to isolate the variety best suited for bringing out the subtle cracked leather aromas from the upholstery of a ’72 Mustang.

Dan Gentile is a freelance writer based in Austin, Texas. This is Dan Gentile’s first feature for Sprudge.

Photographs courtesy the Specialty Coffee Association of America.


RELATED POST

  1. Brett Whitman

    19 October

    We did indeed smoke a fat bowl of the sticky sweet Rwanda to come up with these tasting notes… Soon science will prove the value of this way of tasting. For now we will suffer the slings and arrows from the less progressives out there.

  2. Ryan Ahn

    18 October

    …wait, where was the science in this article?

    1. The chemistry of certain varietals is responsible for specific flavors. All flavor and aroma comes from the underlying chemical matter. Environmental conditions change the underlying chemical makeup of the beans (things like weather, season, etc.), which in-turn have a substantial impact on coffee flavor. Environmental conditions and the innate underlying chemistry of certain varietals are not mutually exclusive when it comes to flavor.

    2. So let me get this straight: a team of five ‘experts’ spent 150 hours tasting coffee with ‘references’ and determined that these 99 flavors were the most important reference flavors in the 105 coffee samples that were in front of them. This might be the dumbest thing I’ve ever heard. Let me elaborate:

    a) Smucker’s® Blackberry Jam is not a representative reference for what blackberries actually taste like. Which begs the question: what other ‘standards’ were they using? Do you know what might be the best flavor reference for blackberries? Blackberries!

    b) There are thousands of compounds that make up coffee flavor and aroma. There are significantly more than 99 recognizable reference flavors that remind us of things we have tasted or smelled in the past (such lemons, chocolate, leather, etc). There are more than 50 countries that produce and grow coffee, there are multiple green processing methods, hundreds of coffee varietals, near-infinite possible roast time-temperature curves, and a multitude of distinct brewing methods that ALL influence flavor and aroma. Surely among all these factors, a distinguishing characteristic between coffee A and coffee B is that one has more of a Smucker’s® blackberry jam note and that the other resembles that of blackberries which are sun-ripened on the hills of the French countryside. In short, limiting the ‘science’ to 99 distinct reference flavors is not enough to account for possible perceptual differences that arise from the vast amount of possible flavor variation in coffee.

    c) A team of five experts is not representative of the general population – the people that will actually be consuming the coffee. These flavors might be most impactful to the experts, because they are the people thinking about flavor for a living. But are these relevant to what Joe Shmoe enjoys in his morning cup o’ joe? I don’t think so. There is a question of relevancy in this flavor wheel and perceptual bias imposed by the coffee industry. It’s the case of our good friend Joe getting cognitively ‘incepted’ by the marketing team and saying, “This coffee _is_ pretty good.” He picks up the package, inscribed, “Hints of fresh lilac, French vanilla, and muddled elderberry.” Joe thinks to himself, “Oh, I didn’t taste any of this before, but that vanilla flavor does taste great!” Or worse, he is incepted negatively! Perhaps Joe hates vanilla, but now that he has been _told_ that’s what he is _supposed_ to taste in this coffee, he actually does taste it and end up enjoying it even less.

    The description and communication of flavor should be data-driven, and in terms that the general market knows, not by an _imposed_ wheel of reference flavors developed by five people.

    3) An objective fact: Four Barrel is awesome. The Four Barrel flavor notes create a better feeling for what can actually be expected while that individual is tasting the coffee, and from a flavor profile communication perspective, they’re doing better in my opinion.

    The SCAA is far away from quantifying flavor as a meaningful metric

  3. Steve Smith

    19 May

    “Tasting a set of 13 Colombian coffees, the cuppers came up with 59 flavor terms, with only four terms universal to more than one cupper. The sensory panel arrived at 92 terms and were in 100% agreement. It proved that however valuable the cupping process, it doesn’t pass scientific muster.”

    I think you’d have to admit that the conclusion stated above does not follow from the information preceding it.

    I get many of the limitations of the cupping process, as well as the relationship between subjectivity and objectivity in sensory evaluation. I also appreciate the balance struck by Sprudge between providing entertainment and information, but I think this is over the top and doesn’t reflect the care Sprudge often displays in trying to clarify the conversation around an important coffee topic.

    I love Sprudge, this is a great topic, just sayin’.

  4. I got similar notes for Four Barrels Rwanda; hoof of old white mustang.

    • Brett

      19 October

      Hey Chris Happy you enjoyed the coffee… Sure those tasting notes fit right in over at your place ;)

INSTAGRAM
Follow us on Instagram