The Disconnect Between
Marketed and Experienced 
Cannabis Aromas: A Need 
for Novel Approaches

Aroma marketing in cannabis has not yet become an effective branding and consumer expectation setting tool. Mismatches between the aroma that consumers expect and the aroma experience during consumption has prevented cannabis brands from using this quality as a product differentiator. These mismatches between promise and experience are partially driven by the use of terpenes to predict aroma of cannabis products. A simple experiment was designed to evaluate whether 
the quantity of individual terpenes of interest were, in fact, associated with the aroma observed by a trained sensory panel. The status quo assumption is that terpenes such as limonene and linalool are directly associated with aroma notes, such as “citrus” and “floral,” respectively. The results suggest that the content of individual terpenes is not strongly associated with the commonly assigned aroma (e.g., citrus from limonene) in cannabis buds. Further, use of total oils as an alternative to terpene testing to capture a wider array of aroma active compounds is discussed, along with post-hoc experiments to illustrate the variation in terpene content across matrices.

As the cannabis industry matures, it has become abundantly clear that there is a disconnect between marketed aromas of cannabis flower and the sensory experience of the user. The routine mismatch between observable aroma and expected aroma in cannabis depreciates the trust between brands and consumers and is likely to decrease the therapeutic effect of a cannabis consumption experience 1.  

‍

Much of the confusion surrounding cannabis aroma is derived from the common assertion that terpenes and terpenoids (terpene compounds) are the primary determinant of perceived aroma in cannabis. Terpene compounds are prominent compounds in essential oils (lipophilic plant extracts) and have long been used as flavorings as well as for various pharmacological applications 2. Some common terpene compounds seen in cannabis include linalool, geraniol, 
and limonene 3.

‍

Despite a lack of sufficient evidence that terpenes are the primary drivers of cannabis aroma, terpenes in cannabis are often quantified and used to present expected “aroma notes” for cannabis—for example, a testing lab may report a limonene-dominant cannabis cultivar as being “citrusy” in character—or even more presumptive, specific citrus fruits such as lemon and lime 4. Associations between predominant terpenes and aroma may seem intuitive at first, but when considering the vast number of chemical combinations, synergies, and masking effects, these simplistic derivations of aroma descriptors often sow more confusion than clarity.

‍

While terpene compounds are very likely to play a major role in defining the overall character and intensity of cannabis flower aroma, the extracted oil content (“essential oil”) of cannabis contains a significant non-terpene fraction, including esters, aldehydes, ketones, organic acids, pyrroles, and thiols. Often, non-terpenes contribute well over 50% of the weight of cannabis oil. This non-terpene fraction has been reported to play a significant role in overall aroma character of cannabis 5. In fact, cannabis’s distinct aroma may be predominantly derived from sulfurous compounds 5,6.

‍

Growers of hops for beer, which hold many similarities to the cannabis plant, were previously presented with a 
similar challenge. Early attempts to create standard measures for hop aroma (i.e., the “hop aroma unit”) were largely unsuccessful. More robust modeling was eventually adopted for hops that considered impact of non-terpene compounds 7. 
It’s possible that cannabis will benefit from similar holistic profiling of total oils content, but there is very little data published currently regarding the relationship between terpene and non-terpene content and the observed aroma of cannabis plants.  

‍

The overall objective of the current study was to assess whether terpene content has been overemphasized as a contributor to cannabis aroma. Pearson linear correlations were used to determine the strength of relationships between terpene content and aroma character in cannabis flower. The results will help clarify the validity of the status-quo practice of tying cannabis aroma notes to terpene content.

‍

Cannabis buds, for all studies, were presented blindly to trained panelists in opaque bags to prevent visual bias. 
The facilities and methods described previously were also utilized for these studies 8.

‍

Total volatile oils content in cannabis products is determined through steam distillation. This method involves adding the cannabis material (100g flower, 20g concentrates) with a large volume of water in a boiling flask. The mixture is then brought to a rolling boil, and the distillation process continues for 4–7 hours. During distillation, the essential oils are collected as they float on top of the aqueous condensate. The distillation rate is maintained at 25–35 drops per minute. After completion, the volume of the collected oil is observed and measured to the nearest 0.05 mL using a calibrated receiver. This method is adapted from the ASBC’s procedure for hops and hop pellets.

‍

Aroma character was evaluated by using check-all-that-apply (CATA) surveys based on our previously published aroma lexicon 8. Twelve (12) randomly selected cannabis samples were evaluated for this experiment, and each sample was evaluated by 8 trained panelists (n=8). For aroma character, a CATA questionnaire with hierarchical categories 
 was presented using custom software developed by Cambium Analytica according to methods described previously 8.

‍

Terpenes content was measured via GCMS using Agilent Intuvo GC (Agilent 9000) connected to MSD (Agilent 5977B) with EI Source Automatic Liquid Sampler (Agilent 7693A) Capillary Columns: two DB-624 30x0.25mmx1.4uL film thickness connected with midpoint back-flush module 
(Agilent p/n 122-1334UI-INT).

Correlation analysis was the primary statistical approach for analysis. CATA detection events were summed for each aroma attribute and correlations were established between the sum for each category with continuous quantitative terpene values using Pearson’s method. A confidence level of 95% was used to determine the significance for all correlations. Figure 3 illustrates associations (Pearson’s r) between measured terpenes and commonly used aroma descriptors associated with terpenes in cannabis.

The data reveals that simple associations between a single terpene compound and a characteristic aroma are not reliable. For example, limonene appeared to have no relationship with detection of lemon, orange, or generic citrus aromas. The best predictor for citrus aroma was ɑ-terpinene for generic citrus (r=0.62) and lemon (r=0.69). For orange, ɑ-pinene (r=0.69) was the best predictor—yet even these associations may be entirely contextual to the unique combination of terpene and non-terpene compounds in the samples. Linalool, often purported to smell distinctly like lavender flowers, was not associated with any floral note, but rather leafy greens (r=0.79), dried herbs (r=0.69), and parsley (r=0.63).

‍

Further, key aroma categories that are commonly used to describe cannabis were not correlated with any terpene measured—sulfurous aromas such as cheesy and skunky showed no significant associations. Three examples of partial mismatch between expected and perceived cannabis aroma profiles in this sample set are presented in Table 1. In example C, while black pepper was a highly cited aroma by the trained sensory panel, the most prevalent terpene was limonene. This may because the “top terpenes” are not weighted for relative aroma strength, as is commonly the case when testing labs report expected aromas based on terpenes. Conversely, in sample B, citrus notes and even specific citrus fruits were highly cited by panelists, but terpenes such as limonene are not present in high quantities. Using the top terpenes to predict aroma created an incomplete picture of the aroma experienced by the panelists, and this phenomenon is likely to 
exist for consumers who purchase aroma marketed cannabis products as well.

As a whole, these data suggest that attempts to classify cannabis aroma based on terpene content, while common 10,11, are not likely to succeed. The limited relationship between terpenes and cannabis aroma is partially driven by substantial variation in composition of the total mass of aroma active oil in cannabis. The non-terpene fraction of cannabis volatiles can compose the majority of the total volatile oil content at times, with this difference being exacerbated upon processing. Importantly, the combination of terpenes and other non-terpene aroma compounds will drive overall perceived aroma character. Production of terpenes occurs primarily in the trichomes of the plant (along with cannabinoids), which has resulted in breeding efforts focused on promotion of trichome density. However, these efforts may inadvertently produce cannabis buds that do not appeal to a significant fraction of the consumer base.

‍

A post-hoc experiment was conducted on hops, cannabis flower, and two types of cannabis extracts produced from the same cannabis crop to illustrate this trend (Figure 4). The variation in the terpene/non-terpene ratio is further complicated by differences between different extraction methods (i.e., resin or rosin). With rosins, which tend to have slightly greater mass of total oils compared to resins, an even greater variance in terpene /non-terpene ratio can be seen—which may suggest a more complete picture of the total aroma compounds.

Novel approaches to rapidly characterizing cannabis aroma are needed to provide more truthfulness in cannabis marketing. One alternative approach that may provide better alignment between expected and perceived cannabis aroma is to routinely collect total volatile oil content and then characterize relative quantity of aroma compound classes within the total oils. These classes may include esters, aldehydes, ketones, pyrroles, organic acids, and thiols 6,12,13—but more research is needed to determine the aroma classes that are meaningfully impactful to the consumer experience. This approach has worked well for hops 14 and, given the similarities between the plants, a total oils-based approach to measuring and marketing cannabis warrants further study.

Considering alternatives that may be superior to the use of terpenes alone to predict cannabis aroma, the use of total oils presents several marketing advantages that may promote its adoption across brands. One reason terpenes may have become attractive to consumers is that it is a single value (% terpenes) that is easy for consumers to understand (presented as; more terpenes, better quality) while making purchasing decisions. Total oils, as a value, is most often 50–75% higher than the terpene value alone, which may help differentiate products during initial adoption. In the same way, total oils is a single value that is easily interpreted and understood by purchasers. Further, % total oils may be positioned as a representation of overall quality and increased production of both oil soluble terpenes and cannabinoids in the plant.

‍

Secondly, total oil reporting escapes market-wide aroma character expectations, allowing brands to avoid innately developed expectations by consumers and to market an aroma that is established by empirically using sensory analysis. Variance amongst the aroma-active profile of cannabis buds (Figure 5., 44.26±10% terpene fraction in total oil) limits the use of a single aroma compound class to create generalized categories of cannabis aroma—but total oils, while not necessarily representing aroma character, may be a superior alternative for the consumer who make be seeking cannabis with potent or subtle aromas. However, a relationship between total oils and aroma intensity has not been established and future research is needed to support this.

Finally, brands that set accurate expectations regarding the aroma properties of their cannabis product are likely to engender repeat purchasing over time as consumers develop trust in the brand. This is perhaps the most valuable benefit of total oils, as expectation setting is a vital component of long-term brand success.

‍

References

1. Plumb, J.; Demirel, S.; Sackett, J. L.; Russo, E. B.; Wilson-Poe, A. R. The Nose Knows: Aroma, but Not THC Mediates the Subjective Effects of Smoked and Vaporized Cannabis Flower. Psychoactives 2022, 1 (2), 70–86. https://doi.org/10.3390/psychoactives1020008.
2. Masyita, A.; Mustika Sari, R.; Dwi Astuti, A.; Yasir, B.; Rahma Rumata, N.; Emran, T. B.; Nainu, F.; Simal-Gandara, J. Terpenes and Terpenoids as Main Bioactive Compounds of Essential Oils, Their Roles in Human Health and Potential Application as Natural Food Preservatives. Food Chem. X 2022, 13, 100217. https://doi.org/10.1016/j.fochx.2022.100217.
3. Hanuš, L. O.; Hod, Y. Terpenes/Terpenoids in Cannabis: Are They Important? Med. Cannabis Cannabinoids 2020, 3 (1), 25–60. https://doi.org/10.1159/000509733.
4. Home. PhytoFacts. https://www.phytofacts.com (accessed 2024-06-06).
5. Oswald, I. W. H.; Paryani, T. R.; Sosa, M. E.; Ojeda, M. A.; Altenbernd, M. R.; Grandy, J. J.; Shafer, N. S.; Ngo, K.; Peat, J. R. I.; Melshenker, B. G.; Skelly, I.; Koby, K. A.; Page, M. F. Z.; Martin, T. J. Minor, Nonterpenoid Volatile Compounds Drive the Aroma Differences of Exotic Cannabis. ACS Omega 2023, 8 (42), 39203–39216. https://doi.org/10.1021/acsomega.3c04496.
6. Oswald, I. W. H.; Ojeda, M. A.; Pobanz, R. J.; Koby, K. A.; Buchanan, A. J.; Del Rosso, J.; Guzman, M. A.; Martin, T. J. Identification of a New Family of Prenylated Volatile Sulfur Compounds in Cannabis Revealed by Comprehensive Two-Dimensional Gas Chromatography. ACS Omega 2021, 6 (47), 31667–31676. https://doi.org/10.1021/acsomega.1c04196.
7. Rettberg, N.; Biendl, M.; Garbe, L.-A. Hop Aroma and Hoppy Beer Flavor: Chemical Backgrounds and Analytical Tools—A Review. J. Am. Soc. Brew. Chem. 2018, 76 (1), 1–20. https://doi.org/10.1080/03610470.2017.1402574.
8. Developing a Modern Cannabis Aroma Lexicon. https://www.cambiumanalytica.com/resources/publications/cambium-analytica-hypermetrics-pt-1 (accessed 2023-06-05).
9. Terpenes. Trichome Analytical. https://trichomeanalytical.com/terpenes/ (accessed 2024-05-31).
10. Mudge, E. M.; Brown, P. N.; Murch, S. J. The Terroir of Cannabis: Terpene Metabolomics as a Tool to Understand Cannabis Sativa Selections. Planta Med. 2019, 85 (9/10), 781–796. https://doi.org/10.1055/a-0915-2550.
11. Abrams, J. S.; Ellyson, W. J. T.; Gomez, V. J.; Talleyrand, J. L. Aroma in Cannabis: A Foundation for Chemotype Classification. In Recent Advances in the Science of Cannabis; CRC Press, 2021.
12. Kneubühl, M.; André, A.; Chetschik, I. Characterisation of the Key-Aroma Compounds among the Volatile Constituents in Different Hemp Strains (Cannabis Sativa L.); Zenodo, 2021. https://doi.org/10.5281/zenodo.5513767.
13. Moreno-Chamba, B.; Salazar-Bermeo, J.; Hosseinian, F.; Martin-Bermudo, F.; Aguado, M.; De la Torre, R.; Martínez-Madrid, M. C.; Valero, M.; Martí, N.; Saura, D. Aromatic and Cannabinoid Profiles of Cannabis Inflorescences and Seed Oils: A comprehensive approach for variety characterization. Industrial Crops and Products, Volume 210, 2024,118143, ISSN 0926-6690, https://doi.org/10.1016/j.indcrop.2024.118143.