The impact of wine label’s sensory descriptions in the consumers’ purchasing decision and consumer experience

i THE IMPACT OF WINE LABEL’S SENSORY DESCRIPTIONS IN THE CONSUMERS’ PURCHASING DECISION AND CONSUMER EXPERIENCE Ana Carolina Lavos Pelicano Dissertation report presented as partial requirement for obtaining the Master’s degree in Statistics and Information Management ii NOVA Information Management School Instituto Superior de Estatística e Gestão de Informação Universidade Nova de Lisboa THE IMPACT OF WINE LABEL’S SENSORY DESCRIPTIONS IN THE CONSUMERS’ PURCHASING DECISION AND CONSUMER EXPERIENCE by Ana Carolina Lavos Pelicano Dissertation report presented as partial requirement for obtaining the Master’s degree in Information Management, with a specialization in Marketing Intelligence Advisor: Diana Orghian November 2021 iii ABSTRACT The wine market is saturated and with increased diversity in styles, brands and prices, the wine purchase decision is difficult, and the consumer might get lost. When consumers are shopping for wine, they face a bewildering array of products bearing a wide range of information and since the consumers cannot usually taste the wine before purchase, they must make their decision based on the available information on the label of the bottle or past decisions. This investigation aims to demonstrate how the tasting note sometimes available on the label, will influence the wine drinking experience as well as the impact that the labels have on the evaluation of quality and desire of the wine. These characteristics determine the consumer's purchase intention based on the design of the label and its respective sensory descriptions. Understanding the consumers and what motivates them is the key to any marketing strategy and wine producers need to know how consumer involvement impacts their decision to purchase wine. To carry out this investigation, a questionnaire was developed. Two hundred responses were considered in a survey with two different scenarios, the two versions were randomly assigned among the participants. The objective was to understand how the sensory description (tasting note) influenced the answers. The study reveals that when consumers know a priori the flavors present in a certain bottle of wine, it changes the way people appreciate it, it makes the experience more exciting for consumers. From the analysis of the results, it is possible to conclude that the presence of detailed sensory descriptions on the labels will allow people to more easily choose the wine they want to consume. Consumers want to be guided in the discovery of the wine flavors to facilitate future purchase decisions. iv KEYWORDS Wine; Label; Sensory experience; Sensory description; Tasting note v RESUMO O mercado dos vinhos está saturado e, com cada vez maior diversidade de estilos, marcas e preços, a decisão de compra do vinho torna-se difícil e o consumidor pode sentir-se perdido. Quando os consumidores compram vinho deparam-se com uma gama desconcertante de produtos e com uma ampla gama de informações e, uma vez que os consumidores geralmente não podem provar o vinho antes de comprar, devem tomar uma decisão com base nas informações disponíveis no rótulo da garrafa. Esta investigação pretende demonstrar como a nota de prova irá influenciar a experiência de consumo do vinho, bem como o impacto que os rótulos têm na avaliação de qualidade e desejo do vinho. Essas características determinam a intenção de compra do consumidor com base no design do rótulo e as suas respetivas descrições sensoriais. Entender os consumidores e o que os motiva é a chave para qualquer estratégia de marketing e os produtores de vinho precisam de saber como o envolvimento do consumidor afeta sua decisão de compra do vinho. Para realizar esta investigação, foi elaborado um questionário. Duzentas respostas foram consideradas numa pesquisa com dois cenários diferentes, as duas versões foram distribuídas aleatoriamente entre os participantes. O objetivo foi compreender como a descrição sensorial (nota de prova) influenciou as respostas. O estudo revela que quando o consumidor conhece a priori os sabores presentes numa determinada garrafa de vinho, que isso muda a forma como as pessoas a apreciam, tornando a experiência mais entusiasmante para o consumidor. A partir da análise dos resultados, é possível concluir que a presença de descrições sensoriais detalhadas nos rótulos permitirá que as pessoas escolham com mais facilidade o vinho que vi desejam consumir. Os consumidores querem ser orientados na descoberta dos sabores do vinho, o que facilitará futuras decisões de compra. PALAVRAS-CHAVE Vinho; Rótulo; Experiência sensorial; Descrição sensorial; Nota de prova vii INDEX 1. Introduction ........................................................................................................................ 1 2. Literature review ................................................................................................................ 4 2.1. Portuguese wine market ............................................................................................. 4 2.2. Wine consumer behavior ............................................................................................ 6 2.3. Labels ......................................................................................................................... 9 2.3.1. The impact of wine label’s design in the consumer purchase intension ................ 9 2.3.2. The influence of perceived quality on purchase intention ................................... 10 2.4. Sensations (taste, smell, aroma and flavor) .............................................................. 12 2.5. Sensory descriptions ................................................................................................. 14 2.6. Olfactory memory .................................................................................................... 16 2.7. Sensory expectations ................................................................................................ 18 3. Methodology .................................................................................................................... 21 3.1. Research Approach .................................................................................................. 21 3.2. Research Design ....................................................................................................... 22 3.2.1. Questionnaire ....................................................................................................... 22 3.3. Data Collection ......................................................................................................... 24 3.4. Data Treatment and Analysis Techniques ................................................................ 25 4. Results and discussion ...................................................................................................... 26 4.1. Descriptive Analyses ................................................................................................ 26 4.1.1. Demographic characterization.............................................................................. 26 viii 4.1.2. Wine consumption habits ..................................................................................... 27 4.1.3. Wine attributes ..................................................................................................... 28 4.1.4. Tasting note exercise ............................................................................................ 29 4.1.5. Wine label sensory descriptions ........................................................................... 35 4.1.6. Wine label design ................................................................................................. 37 4.2. Hypothesis Validation .............................................................................................. 38 4.2.1. MANOVA ............................................................................................................ 38 4.2.2. Independent Samples T-Test ................................................................................ 39 5. Conclusions ...................................................................................................................... 44 6. Limitations and recommendations for future works ........................................................ 47 7. Bibliography ..................................................................................................................... 49 8. Annexes ............................................................................................................................ 53 8.1. Annex A – Questionnaire ......................................................................................... 53 ix LIST OF FIGURES Figure 2.1 - Portugal Total Beverage Alcohol by Category (Source: adapted from The IWSR, 2020)...........................................................................................................................4 Figure 2.2 - Wines: Local vs Imported (Source: adapted from The IWSR, 2020)……...5 Figure 2.3 - E-Commerce Value and Share (Source: adapted from The IWSR, 2020)…5 Figure 2.4 - Still Wine By Color Table (Source: adapted from The IWSR, 2020)……...6 Figure 4.1.2.1 - Regularity of Wine Consumption……………………………………..27 Figure 4.1.2.2 - Wine Purchase and Choice Decision……………………………….....28 Figure 4.1.4.1 - Tasting Note Exercise Without Any Element Highlighted…………....30 Figure 4.1.4.2 - Tasting Note Exercise With Highlights……………………………......31 Figure 4.1.4.3 - Tasting Note Exercise Without Tasting Note………………...……......33 Figure 4.1.4.4 - Tasting Note Highlighted (Second version)……………………….…..34 6 The red wine is the most consumed type of wine, in 2019 represented 60,3% of the total consumption of still wine in Portugal. White wine is the second most consumed type of wine (35,8%), in 2019 increased by 0,2% compared to 2018. The consumption of rosé wine has little expression in Portugal, its market share was 3,4% in 2019 (Young,2020). Figure 2.4. Still Wine by Color table (source: adapted from The IWSR, 2020). In a study conducted by Ferreira et al (2019), it was concluded that in Portugal women frequently associate wine to the context of consumption while men frequently associate wine to convivial and sensorial pleasure. The region of origin and prior knowledge experience seem to be the two main reasons for men to choose a wine, while women seem to rely more on wine brand and previous experience. Front label information (region of origin, awards and region illustration) seems to be more important for women, while the back label descriptors (grape variety, world heritage site and wine history) are more relevant for men (Ferreira et al, 2019). 2.2. WINE CONSUMER BEHAVIOR Holbrook and Hirschman (1982) suggested that consumers purchase products not only to solve a problem or to use a product, but also to have an experience, stressing the hedonic dimension of consumer behavior. 7 Empirical research pinpoints the importance of the sensory dimension of consumer experiences as key in engaging and co-creating value with consumers (Gentile, Spiller, & Noci, 2007, cited by Santos, 2019). For Kotler (2002), understanding consumer behavior is not simple, because consumers can say one thing, but do another. The task of marketing is to understand what happens in the buyer's consciousness between the arrival of external stimuli and purchasing decisions. Involvement with the product, in this case with wine, is one of the important elements for a better understanding of behavior (Bruwer & Buller, 2013). For this it is necessary to realize the main issues and attributes at the base of wine purchase (Corduas et. al., 2013), bearing in mind that these are very diverse, and may or may not be intrinsic to the product (Cox, 2009). In the study by Solomon et. al. (2006), there is recognition of three phases, the prepurchase, the purchase and the post-purchase (e.g. expectations and post-purchase behavior), and this consumer buying behavior is influenced by cultural, social, personal and psychological factors (Galalae & Voicu, 2013). According to Forbes (2008), the rational evaluation of wine attributes will influence consumer decisions. The challenge for understanding these preferences is the large number of attributes that exist in wine. Some consumers have different risk profiles, leading some to make "safe" choices and others to be adventurous in their choice of wine (Lockshin, 2003). Wine criteria include: brand, producer, grape variety, blend of grape varieties, vintage, region of origin, price, label, bottle type, cork type, bottle size, wine color, wine style and alcohol level (Geraghty , & Torres, 2009). These can be intrinsic, being related to sensory characteristics as well as the experienced and expected quality, or extrinsic, being related to the region, year, label, among others. While producers focus on internal aspects related to sensory characteristics, consumers often make their deduction of wine quality based on external attributes (Corduas, et. al., 2013). 8 Ferreira et al (2019) observed gender differences regarding the type of wine consumed in Portugal: on average, red wine is preferred by men while women prefer white and sparkling wine (Pettigrew, 2003). Bruwer et al. (2011) found that women and men have different sensory preferences and wine consumption behavior. Regarding the rank order importance of extrinsic cues on wine choice, for men the most important factors are the back label information, the region of origin and the grape variety. The most relevant cues for women are the brand and the sensory profile (Ferreira, LourençoGomes, Pinto, & Silva, 2019). Sanches (2013) identified hypermarkets, supermarkets and restaurants as the places of election when buying wine, in that specific order. Figueiredo et al. (2003) agreed that the price is the main reason for purchasing wine in these places, mainly due to the country economics but also due to the location of these establishments (e.g. near residential areas or workplaces). About the Portuguese online wine buying behaviour, Santos and Ribeiro (2012) findings suggested that the typical consumer profile that uses this channel for wine purchasing is young, male and with high income. This research survey also revealed that the price, availability, wide selection, region of origin, brand, recommendation of the online shop and prior experience were the most considered factors of influence on their purchase decision. As for the legend that cheap wine causes more headaches and worse hangovers, Waterhouse said that there's no particular reason to think that expensive wine is better than cheap wine. It necessarily means there may be a simpler explanation, that is that it's possible that people tend to drink more wine when it's cheap (Panko, 2017). 9 2.3. LABELS 2.3.1. The impact of wine label’s design in the consumer purchase intension Wine and other consumable products have a unique constraint that sets them apart from other traditional non-consumable products, primarily because the quality of the product cannot be assessed until it has been consumed (Barber & Almanza, 2006). The result is that consumers face more complex buying decisions for wine than for many other consumable products (Barber, Almanza & Donovan, 2006, cited by Barber, Ismail & Dodd, 2007). Packaging is considered one of the most efficient marketing tools to reach the consumer, since it has a longer lifespan than an advertisement or promotion, and because it is seen by more people (Stahlberg & Maila, 2010). Package design and the design of labels on wine bottles have been shown to affect the hedonic rating of products (Gmuer, Siegrist, & Dohle, 2015; Westerman et al., 2013, cited by Apaolaza et al, 2017). The front label is the first line of communication to entice the consumer, so it is therefore extremely important that the characteristics appearing on the label meet certain basic standards yet are visually attractive to stand out from the large number of competing bottles available on the retail shelf (Barber & Almanza, 2006) because consumers with low involvement shop with their eyes (Chaney, 2000; Jennings & Wood, 1994; Olsen et al., 2003; Thomas & Pickering, 2003). Not all consumers are alike. Some know what they want and select a wine based on their experience, or the wine’s style, region, or producer. Others, however, do not know and scour a bottle’s labels to satisfy their information requirements (Thomas & Pickering, 2003). Combris et al (1997), cited by Thomas & Pickering (2003), indicates that consumers may decide wine based on objective characteristics (appearing on the label of the wine bottle). Alternatively, Kidd (1993) considers that shape and color can impact choice more, and that 10 the front labels should be reserved for making a positioning statement (Thomas & Pickering, 2003). Thomas and Pickering (2003), cited by Barber & Almanza (2006), determined that the grape variety, brand name, and price were the most important informational items consumers used to assess wine products before purchase. According to a study on the importance of labels, the back label has been identified as an underutilized area for providing information (Barber & Almanza, 2006). The most direct factors used by the manufacturer and retailer to influence expectations are: advertising, package, labeling, and information (Deliza & Macfie, 1996). The packaging also allows to communicate the personality of the brand and is a way to create a competitive advantage in relation to the competitors (Solomon, Marshall, & Stuart, 2012). Arguably, the array of information on wine labels identifies the special qualities of a bottles’ contents. Also, the information presented coveys a wine’s character and helps make a statement to potential buyers. Label information and the consumer’s evaluative criteria contributes to assessing and deciding upon competing alternatives (Thomas & Pickering, 2003). 2.3.2. The influence of perceived quality on purchase intention Many wine producers acknowledge that to a new generation of consumers, the package is as important as the product (Batt, 1998; Batt & Dean, 2000; Chaney, 2000; Ivinsky, 2000; Jennings & Wood, 1994; Olsen et al., 2003; and Thomas & Pickering, 2003, cited by Barber et al, 2007), so, the decision to buy the product is more influenced by the customer’s perception of quality than by quality itself (Birčiaková et al, 2016). Because the front label is the first line of communication, it is therefore extremely important that the characteristics appearing on the label are visually attractive and enticing (Barber, Ismail & Dodd, 2007). 11 Perceived quality is usually pinpointed as having a more active role in influencing consumers’ purchase intention (Richardson, Dick and Jain, 1994), which justifies its importance for the current study. To evaluate quality, consumers need to have at their disposal an array of cues capable of signing the excellence of the product, thus the process of quality perception may be interpreted as comprising two stages: before consumption, in which it is formed a quality expectation, and upon consumption, when quality performance is evaluated. Both formed on the bases of quality cues and attributes (Steenkamp & Trijp, 1996). By applying Ophuis and Trijp (1995) perspective it to the wines’ sector is commonly agreed that quality is not only related to the product itself (sensory properties, origin, vintage, ...), but also to the consumer as well (involvement level, country of origin, expectations, ...). It may be even stated that quality can be dependent on the context in which the product is consumed (Sáenz-Navajas, Ballester, Peyron and Valentin, 2014). Behind the fancy labels and bottles and newfangled chemical manipulation, the biggest factor influencing the price of wine is an old one: terroir, or the qualities a wine draws from the region where it was grown. Famous winemaking areas such as Bordeaux can still land prices 10 times higher than just as productive grape-growing land in other areas, says Waterhouse. Many of these winemakers grow varieties of grapes that produce less quantity but are considered by winemakers to be far higher quality (Panko, 2017). Wine producers must take into account some aspects that could negatively affect consumers' perception of quality, such as screw-top closures, non-glass containers, label logo and design, large formatted bottles or boxed wines, these examples have indicated a lesser quality product to consumers (Reidick, 2003, cited by Barber & Almanza, 2006). Understanding which product characteristics lead to a higher quality inference in the consumers’ mind is a powerful tool to design more suitable marketing strategies. In the words 12 of Lawless (1995), quality is bound to consumer perception, opinion and attitude, and it is in fact a critical factor involved in consumers’ decision process at the buying moment. 2.4. SENSATIONS (TASTE, SMELL, AROMA AND FLAVOR) Drinking wine is among the most complex multisensory experiences, where all of the senses tend to be activated. Hence, wine tasting experiences must also be studied from a multisensory perspective. In fact, not all of the sensations that the brain associates with the experience of wine originate from the tongue. Olfaction, for instance, is dominant before, and while wine is being tasted. Thus, for a more precise understanding of the way a consumer experiences wine, it is crucial to consider how the senses interact prior to, during, and even after the wine tasting experience itself (e.g., with the latter potentially affecting the purchase intention, and/or willingness to pay for a food/drink product) (Campo, Reinoso-Carvalho, & Rosato, 2021). It may not always be possible to activate the verbal label for the odor, even though its general class (e.g., fruity), its familiarity, and its hedonic aspects may be appreciated. (Stevenson, Case, & Mahmut, 2007). Stevenson et al (2007) consider that this argument is premised on the relationship between the sense experience produced by perceiving an object and the generation of its veridical label or name. For Treasury Wine Estates, the process of developing a wine for the mass-market starts in a kind of sensory insights lab. In these laboratories, focus groups of professional tasters blindsample a variety of wine products. The best ones are then sampled by average consumers to help winemakers get a sense of which sensory profiles would do best in stores and restaurants (Panko, 2017). The center of the aroma wheel consists of generic object groups (i.e. fruits, spices, flowers). Each object group is then subdivided towards the periphery of the wheel (i.e. 13 beneath fruit: tree fruit, dried fruit, tropical fruit). The sub-groups are further subdivided into several specific attributes belonging to them (i.e. beneath tree fruit: apple, cherry and so on). When training odor perceptions with the ‘aroma wheel’, known references in the wine are used (i.e. lemon and vanilla) (Herdenstam et al, 2009). Food and drink experiences are multisensory in nature. Considering flavor, these involves sensations beyond the taste buds. Flavor perception usually integrates taste (the sensations gathered by the taste buds—sweet, salt, bitter, sour, and umami) with olfaction and possibly with elements of touch (Campo, Reinoso-Carvalho, & Rosato, 2021). The sensory characteristics of wine, such as color, aroma, and mouthfeel, are closely linked to wine quality, which is hard to define because of its complexity and multidimensional aspects (Yang & Lee, 2020). When thinking on the role of the five basic senses in the experience of wine, hearing has perhaps the least obvious role. In the recent decades, researchers have been proving that what we hear can significantly affect the tasting experience of foods and drinks, including wine (Areni & Kim, 1993). A couple of pioneer studies have been conducted focusing on how the music that one listens while shopping can affect our purchasing behavior for wine. Such shopping experience was further enhanced by French and German music, being played on alternate days. Interestingly, when French music was playing, French wines outsold the German ones, and vice-versa. The influence of type of music on the experience of wine was also assessed by, but this time with a focus on flavor perception (Areni & Kim, 1993 cited by Campo, Reinoso-Carvalho, & Rosato, 2021). Wine is renowned and valued for the variety and complexity of sensations that it elicits, including visual (hue, intensity and clarity), olfactory (orthoand retro-nasal aroma), taste (sourness, sweetness, bitterness and saltiness), and those associated with the chemesthesis and somatosensory systems. When activated by wine, the latter two systems contribute to 14 what we experience as “tactile sensations” or mouthfeel, with polyphenolic compounds (tannins) recognized as one of the main elicitors in red wine (Pickering, 2008). Ventura and Worobey (2013) suggest that flavor sensations originate principally from the integration of the smell and taste, where smell becomes essential for the brain while developing food preferences, and while understanding flavor. Indeed, olfactory stimuli, being processed via ortho, or retronasal, olfaction, are particularly important for the brain while processing a wine’s flavor (Ventura & Worobey, 2013). Therefore, when thinking about the different ways in which to approach the different multisensory information that may be part of an experience with wine, the question of whether such sensations may elicit specific emotions, should be at stake. The emotions add value to the whole concept behind the corresponding wine experience (Campo et al, 2021). 2.5. SENSORY DESCRIPTIONS The language of tasting notes is not always precise, since it is used to express sensations that can be very subjective. Wine tasters use descriptors to express what they see, smell and taste in a wine, but these common terms are often used even by wine experts with what has been termed as a blithe lack of precision (López Arroyo & Roberts, 2020). Wine descriptors are common terms that can be used to describe a particular wine and the sensations it evokes; however, although their purpose is to clarify sensations, they often end up creating more of a problem as the descriptors are themselves not always clear (López Arroyo & Roberts, 2020). 15 It may not always be possible to activate the verbal label for the odor, even though its general class (e.g., fruity), its familiarity, and its hedonic aspects may be appreciated (Stevenson, Case, & Mahmut, 2007). Thorngate concludes that wine is a complex product, both in its chemical composition and in respect to content of sensory attributes generating a spectrum of different odor, taste and tactile perceptions (Herdenstam et al, 2009). The wine-description task demonstrated a degree of skill, in terms of specificity and quantity of wine-relevant descriptors, as a function of expertise (Zucco, Carassai, Baroni, & Stevenson, 2011). According to Paradis (2010) and Dubois (2017), the role of words and expressions in the language of wine is, among others, to activate kinaesthetic experiences that play a central role in the descriptions of appearance, smell, taste and mouthfeel in wine description (López Arroyo & Roberts, 2020). A likely reason why verbal context is so influential in olfaction is that odors are invisible, poorly localized in space (Engen, 1982), and inherently difficult to verbally identify (Cain, 1979; Lawless & Engen, 1977), and, as such, the drive to seek meaning from external context is especially high (Engen, 1987, 1991; Herz, 2001; Herz & von Clef, 2001). Names can influence odor pleasantness, familiarity, and intensity ratings (Ayabe-Kanamura et al., 1998; Distel et al., 1999; Distel & Hudson, 2001). For this dissertation, I considered a very interesting study of Herz (2003), whose aim was to determine the degree to which ratings of odor hedonics are influenced by verbal information in comparison to pure sensory experience and whether the effects of verbal context are different if they are self-generated or explicitly provided. He also aimed to discover how sensory valence and verbal labels would interact to influence hedonic evaluations and assessed how odor hedonic perception elicited by the stimulus alone (without 22 • H5: People enjoy wine more if they have access to sensory information • H6: Consumers prefer to discover the aromas and flavors present in the wine on their own rather than being guided in this discovery • H7: Consumers by identifying the flavors they like the most it the wine, in the future will be easier to choose which one to buy Regarding the research method, quantitative research was adopted, as the purpose was to reach a high number of respondents. This method aims to quantify the data and usually apply some form of measurement and statistical analysis (Malhotra et al., 2017). The same author considers that with the growth in the use of e-mail and the internet as a means of communication, online methods have become the most viable and popular means of conducting surveys. 3.2. RESEARCH DESIGN 3.2.1. Questionnaire This quantitative research included an online survey carried out through the Qualtrics platform, as it allows for an easy structuring, execution, sharing and analysis of the research. Some of the advantages of using online questionnaires are the time and cost of the survey, as the questionnaire is reached more quickly, receives more responses in less time, does not use paper, reaches a large number of respondents and is geographically distant (Vasconcellos Guedes and Guedes, 2007). The universe corresponds to “people who like wine”, in order to confirm this requirement, a verification question was asked at the beginning of the questionnaire. It is made up of men and women residing in Portugal, over 18 years of age. The sample size is 200 individuals. 23 The questionnaire was shared among family and friends on Facebook, Instagram and various WhatsApp groups and was available for 2 weeks, thus obtaining 261 responses, however only the responses of individuals who responded to the survey in full (progress 100%) were considered. Initially, a pre-test was carried out, which was made available to 30 people, in order to test the questionnaire's operability and obtain some feedback regarding the formulation of the questions and answer options. After some corrections so that no questions raised any doubts among the respondents, the questionnaire was then launched. At the beginning of the questionnaire, a brief introduction about the purpose of the study was presented and the respondent was asked whether if would accept to participate in the study on a completely voluntary basis, so the questionnaire only continued if the answer was positive. Then, was asked whether individual likes wine or not. If the respondent did not like wine, the survey ended there. On the other hand, if the respondent liked wine, would be asked about the frequency of consumption, the preferred time of day to consume this type of drink, and also the preferred places of consumption and purchase. It was also important for this study to understand if people, when they buy wine, know exactly what wine to buy or if, when choosing, they are based on criteria that value or ask someone for an opinion. The respondents had to select the attributes they most value and the most important sources of information when buying wine for themselves, friends and even family members. There was a moment of the survey that had two versions, which were randomly drawn among the participants, randomized controlled trial method was used to see if note makes a difference. Respondents observed an image with a bottle of white wine and its label, but they did not have access to exactly the same information contained on the label. Some answered questions 24 related to the wine based on a tasting note, and others did not have the tasting note to guide them in their sensory experience, for the same questions. The objective was to understand how the tasting note influences their answers and how it differs, according to sensory information. There was a second moment in which the respondents of the first version, who initially had access to a tasting note, answered again to some questions, with that same tasting note highlighted. For those who did not have a tasting note initially, they obtained a highlighted tasting note at this stage. The objective was to verify if this highlight would have an impact on people's attention and, consequently, on the perspective created about the wine. Then, all participants answered the same questions regarding sensory experiences that appear on wine labels and label design. The questionnaire had several types of questions, including closed multiple-choice questions with nominal and ordinal scales, classification questions, matrix-type questions and the Likert scale. Finally, sociodemographic questions were asked. 3.3. DATA COLLECTION The primary data of this dissertation were obtained through an online questionnaire, and according to Nayak and Narayan (2019) the survey is one of the most common primary data collection methods. Secondary data were collected through the entire bibliography used in the literature review, highlighting as main resources the scientific articles present in the databases B-On, Research Gate, ProQuest, Google Scholar and statistical data from The IWSR Drinks Market Analysis, among other materials considered relevant throughout the investigation. 25 3.4. DATA TREATMENT AND ANALYSIS TECHNIQUES Through the Qualtrics platform, the questionnaire data were transferred to Microsoft Excel and JASP statistical analysis programs in order to process the data. After compiling all responses, an analysis was performed in order to evaluate and eliminate invalid responses. Thus, the questionnaire obtained 261 responses, but only 200 were used. Considering that the study universe is based on wine consumers, 61 of these responses were not counted because they were non-wine lovers and/or people who responded very incompletely, so they were not analyzed. 26 4. RESULTS AND DISCUSSION The purpose of this chapter is to describe, analyze and interpret the data collected with the aim of later drawing conclusions. 4.1. DESCRIPTIVE ANALYSES 4.1.1. Demographic characterization Demographic characterization data are shown in table 4.1.1. The age group that registered the highest number of responses was between 21 and 30 years old (38.5%), on the contrary, to a lesser extent, there was the age group with less than 20 years old (1%) and the group between 31 and 40 years old (5.5%). Most respondents are female, representing 55% of the sample, with 43% male. In this analysis, the sample comprises 200 individuals. Table 4.1.1 Characterization of the demographic variables of the sample, by absolute frequency and percentage (N=200). Demographic Variable Values Frequency Percentage Age Group Until 20 21 to 30 31 to 40 41 to 50 51 to 60 More than 60 2 77 11 22 42 46 1% 38,5% 5,5% 11% 21% 23% Gender Female Male Other Prefer not to answer 110 86 1 3 55% 43% 0,5% 1,5% 27 4.1.2. Wine consumption habits Regarding wine consumption habits, most respondents assumed they are wine lovers (58%), the remaining 42% only drink wine socially. The majority of the respondents drink wine with a lot of regularity, which is a good indicator for this study because means that the responses are more reliable once they are active wine consumers (Figure 4.1.2.1). Figure 4.1.2.1 Regularity of wine consumption. When the respondents want to buy wine, normally they decide at the time of purchase based on attributes they value (48%), the other 52% are divided between knowing exactly what to choose (26%) and ask someone who understands more than them (26%). The answers obtained regarding the attributes are in accordance with Forbes (2008), that consider that the rational evaluation of wine attributes will influence consumer decisions, and with the authors Combris et al (1997) that consider that consumers may decide wine based on objective characteristics (appearing on the label of the wine bottle). 1% 11% 13.50% 11.50% 24.50% 29% 9.50% Less than once a year Anually More than once a year Monthly More than once a month Weekly More than once a week Everyday 28 Figure 4.1.2.2. Wine purchase and choice decision. 4.1.3. Wine attributes The respondents had to select up to 4 attributes they value more when they purchase wine. The top 3 were the type of wine (25,48%), the price (20,64%) and the region of the wine. The wine label information was the less valued (5,26%). The results obtained differ from the authors Thomas and Pickering (2003), which consider that the most important informational items consumers value more are the grape variety, brand name, and price. Table 4.1.3 Wine Attributes Attributes Frequency Percentage Type of wine 184 25,48% Region 131 18,14% Brand 90 12,47% Price 149 20,64% Wine label design 69 9,56% Wine label information 38 5,26% Awards 61 8,45% 26% 48% 26% Usually when I want to buy wine, I ask someone who understands more than me Usually when I want to buy wine I decide at the time of purchase based on attributes I value Usually when I want to buy wine I know exactly what to choose 29 4.1.4. Tasting note experiment This survey had two versions, which were randomly drawn among the participants. Respondents did not have access to exactly the same information contained on the label. Some answered the questions based on a tasting note, and others did not have a tasting note to guide them in their sensory experience, in a first phase. The objective was to understand how the tasting note influenced their answers. This was a between-subject manipulation, where the independent variable was the presence of note highlight. • In the first version (N=97), there were two distinct phases. o First, people were asked about the tasting note, it was present but without any element being highlighted: Figure 4.1.4.1. Tasting note exercise without any element highlighted. Before responding to the next questions, people were asked about “How much do you like white wine?”, on a scale of 0 to 5, the average was 3.8. 30 The majority of the respondents considers that need more information about the flavor (43,3% against 36,1% that do not agree, 20,6% do not agree nor disagree), despite the general opinion that they would like the wine based on the information available on the label. Table 4.1.4.1 Questions of the first version of the survey, present tasting note without highlights Totally disagree (1) Disagre e (2) Slightly disagree (3) Do not agree nor disagree (4) Agree (5) Slightly agree (6) Totally agree (7) Mean SD From the description, I think I would like this wine 4 6 6 16 21 37 7 4,89 1,52 From the description, I think I would buy this wine 6 12 4 16 26 30 3 4,51 1,63 The label description is enough to form an idea about the taste of this wine 0 14 8 19 21 27 8 4,65 1,53 I need more information about the flavor 9 18 15 20 18 14 3 3,76 1,66 o There was a second moment when the tasting note was highlighted. The objective was to verify if this highlight would have an impact on the attention of individuals and, consequently, on the perspective created about the wine (Figure 4.1.4.2). 31 Figure 4.1.4.2. Tasting note exercise with highlights. As we can see in the table below, the highlight of the tasting note had a positive impact, people pay more attention and value it more if it is clearly visible. It is information that many people are unaware of, which is sometimes present in the back wine label. In this second moment, with the tasting note highlighted, 87,6% of the individuals agree that the tasting note on the label allows them to form a more precise idea about the wine and 77,3% consider this information important to make a decision about purchasing the product, 69,5% of the participants consider that when they identify the aromas that most like, it will help them in the future for choosing wine (137 in 197 answers). 38 When asked about if they’ve already bought wine that didn’t know because liked the label, 47% said yes. Table 4.1.6 Wine label design questions Totally disagree (1) Disagre e (2) Slightly disagree (3) Do not agree nor disagree (4) Agree (5) Slightly agree (6) Totally agree (7) Mean SD The label design influences my choice when buying the wine 9 48 0 26 59 39 19 4,36 1,79 When I don't know which wine to buy, I prefer to buy wines that have beautiful labels 26 67 0 30 26 31 20 3,68 2,02 Labels with a simple and careful look convey the feeling of wine quality 6 22 0 41 51 57 23 4,86 1,55 I've already bought wine that I didn't know because I liked the label 28 57 0 21 22 42 30 3,99 2,17 In accordance with the authors Birčiaková et al (2016), the decision to buy the wine is influenced by the customer’s perception of quality. 4.2. HYPOTHESIS VALIDATION The results obtained for the validation of the research hypotheses of this study are presented below. 4.2.1. MANOVA 39 Table 4.2.1 MANOVA: Pillai Test Cases df Approx. F Trace Pillai Num df Den df p (Intercept) 1 1605.461 0.987 9 190.000 < .001 Condition 1 5.208 0.198 9 190.000 < .001 Residuals 198 We can see from the Manova table 4.2.1 that the Condition has a statistically significant effect on the way consumer experiences wine at the p<.001level. The one-way ANOVA revealed that there was a statistically significant difference between the two groups (F(1,1) = [5.208], p = <.001). 4.2.2. Independent Samples T-Test In order to determine whether there is a statistical evidence that the associated population means are significantly different, independent samples t-test were made. 40 Table 4.2.2 Independent Samples T-Test t df p From the description, I think I would like this wine 0.633 198 0.527 From the description, I think I would buy this wine 1.110 198 0.268 The label description is enough to form an idea about the taste of this wine 2.917 198 0.004 I need more information about the flavor -4.203 198 < .001 The tasting note on the label allows me to form a more precise idea about this wine -0.989 198 0.324 I consider this information important to make a decision about this wine -2.334 198 0.021 How would this information influence the appreciation/tasting of this wine? -0.706 198 0.481 I prefer to discover the flavors of wine by myself rather than being guided in this discovery 3.791 198 < .001 By identifying the aromas I like, it will help me in the future when choosing wines -1.859 198 0.065 ᵃ Note. Student's t-test. ᵃ Levene's test is significant (p < .05), suggesting a violation of the equal variance assumption Table 4.2.3 Independent Samples T-Test Group Descriptives Group N Mean SD From the description, I think I would like this wine 1 97 4.89 1.52 2 103 4.75 1.58 From the description, I think I would buy this wine 1 97 4.51 1.63 2 103 4.24 1.71 The label description is enough to form an idea about the taste of this wine 1 97 4.65 1.53 2 103 3.98 1.70 I need more information about the flavor 1 97 3.76 1.66 2 103 4.71 1.53 The tasting note on the label allows me to form a more precise idea about this wine 1 97 5.53 1.60 2 103 5.73 1.29 I consider this information important to make a decision about this wine 1 97 5.19 1.45 2 103 5.64 1.31 How would this information influence the appreciation/tasting of this wine? 1 97 5.18 1.16 2 103 5.30 1.35 I prefer to discover the flavors of wine by myself rather than being guided in this discovery 1 97 3.96 1.77 2 103 3.03 1.70 By identifying the aromas I like, it will help me in the future when choosing wines 1 97 5.39 1.41 2 103 5.74 1.22 41 Note: Group 1 represents the respondents of the first version, group 2 represents de second version. H1: After reading the label, consumers think they will like the wine Regarding how much consumers thought they would like the wine after reading the sensory description, there was no significant effect, t(198) = 0.633, p = 0.527, despite the Tasting Note Condition group of consumers consider it slightly more useful (M=4.89, SD=1.52) than the one with Without Tasting Note Condition (M=4.748, SD=1.582). Thus, H1 was not supported by the data. H2: The label description generated purchase intention for the wine Regarding the purchase intension for the wine after reading the label description, there was no significant effect between the two groups, t(198) = 1.110, p = 0.268, despite the second group of consumers consider it slightly more useful (M=4.243, SD=1.712) than the first one (M=4.505, SD=1.627). H3: Sensory description helps to form an idea about the taste of wine To test hypothesis 3, an average has been done between questions 3 and 4 due to their similarities. The 103 participants from the Without Tasting Note Condition (M=4.345, SD=1.614), compared to the 97 participants from the Tasting Note Condition (M=4.206, SD=1.593), demonstrated that sensory description significant helps them better to form an idea about the taste of the wine t(198) = -0.643, p= 0.003. H4: The tasting note is important to make a decision about which wine to buy To test hypothesis 4, an average has been done between questions 5 and 6 due to their similarities. 42 From Tasting Note Condition (M=5.356; SD=1.524) to Without Tasting Note Condition (M=5.685; SD=1.299) there was no significant effect on the fact that the tasting note allows consumers to form a more precise idea about the wine, t(198) = -1.662, p= 0.173, despite of in the second condition the 103 participants consider that the tasting note will be more important to make a decision about which wine to buy than the first 97 participants. H5: People enjoy wine more if they have access to sensory information From Tasting Note Condition (M= 5.175, SD=1.155) to Without Tasting Note Condition (M= 5.301, SD= 1.349) there was no significant effect on the fact that people enjoy more the wine if they have access to sensory information, t(198) = -0.706, p= 0.481, H5 was not supported. H6: Consumers prefer to being guided in the discovery of aromas and flavors present in the wine than discover it by themselves The 103 participants from the Without Tasting Note Condition (M=3.029, SD=1.695), compared to the 97 participants from the Tasting Note Condition (M=3.959, SD=1.773), demonstrated to be more interested in being guided in the discovery of flavors through the tasting note than discovering it alone t(198) = 3.791, p= .001. H7: Consumers by identifying the flavors they like the most in the wine, in the future will be easier to choose which one to buy There was marginal effect on the impact that identifying the flavors people like the most has on making it easier to choose wine in the future between the two groups, t(198) = -1.859, p = 0.065a, with the Without Tasting Note Condition group of consumers considering it more 43 useful (M=5.738, SD=1.220) than the first one with Tasting Note Condition (M=5.392, SD=1.411) 44 5. CONCLUSIONS Purchase intention was proven to be highly affected by individuals’ wine desire, their quality perceptions and attention levels to the bottle of wine, in the literature. One of the objectives of this research was to understand the attributes that consumers most value in a wine at the time of purchase. The results of this study indicate that the type of wine, price, region, and brand are the attributes most valued by individuals, at the end of the list there are awards, label design and label information. Generally, when people buy wine, they don't know exactly which wine they want to buy and end up deciding at the time of purchase based on attributes they value, so it is possible to conclude that people are generally open to new suggestions, and that due to the extensive offer, the wine consumer likes to discover and buy different wines, is predisposed to be influenced by certain characteristics and is aware that they exist, so winemakers can do a lot of different things. The wine product packaging consists of a series of interconnected components, such as the shape and color of the bottle, the type of cap and the presentation of the label, consisting of the design, logo and information provided. The perceived quality of these characteristics has a great influence on the purchase intention of consumers. To be able to assess the quality of the wine, consumers need to have at their disposal a set of clues that guide them in that direction. Thus, the quality perception process can be composed of three stages: before consumption, in which an expectation of quality is formed, in consumption, when quality performance is evaluated, and finally, the pleasure obtained through access to a different sensory experience, by feeling the presence of the flavors and aromas mentioned on the label. 45 Understanding which product features lead to higher quality inference in consumers' minds is a powerful tool for designing better marketing strategies. Quality is linked to consumer attitude, being in fact a critical factor involved in the purchase decision process. It is possible to conclude that the use of the right words is very important, must be used a vocabulary that clearly describes the sensations present in that wine, as well as having a complete label, with all the necessary information, and a careful label that conveys the feeling of quality. By creating two different versions of the survey, it was possible to obtain some conclusions about how the tasting note influences the wine consumption experience, since the respondents did not have access to exactly the same information contained on the label. At the beginning, the first group of participants answered sensory questions based on a label with a tasting note, and in the second group, there was no tasting note to guide them in these questions. The mean in each question was higher in the version with the tasting note presented, between 1 and 7, the opinion was more favorable and in accordance to the questions of the first version. The highlight of the tasting note had a positive impact, people pay more attention and value it more if it is clearly visible, it was concluded that many people did not pay attention to the back label because they were not aware of the existence of the tasting notes present on the label. It’s possible to conclude that the presence of detailed sensory descriptions on the labels will allow people to more easily choose the wine they want to consume, the Hypothesis 3 – Sensory description helps to form an idea about the taste of wine – was supported. It’s also possible to conclude that sensory descriptions will allow people to enjoy it in a different way, people feel excited about the fact that they can identify a lot of different aromas 46 and flavors in the wine. Consumers prefer being guided in the discovery of the wine flavors and aromas than discover by themselves - hypothesis 6 was supported. In the beginning of the survey, the sensory description was the less valued attribute, but after demonstrated the advantages of the back label throughout the questionnaire, the participants admitted that are more interested in reading the descriptions of sensory experiences that appear on wine labels in the future. With this dissertation it is expected to help wine marketeers to increase the sales by improving brand’s image and the quality perception of consumers through the wine label’s design and label’s information. Understanding how men and women looking for information on a wine bottle can help marketers communicate with specific market segments and the knowledge of the importance given to wine label information would offer wine makers a value-added impression of specific information elements and its value to wine consumers This paper provides some insights to design marketing campaigns regarding product customization at the level of label information and design. 47 LIMITATIONS AND RECOMMENDATIONS FOR FUTURE WORKS As many other investigations, this study entails certain limitations that need to be addressed. Unfortunately, the data collection for this dissertation was not what was initially planned, which would be through an experience at NOVA Marketing Analytics Lab. The idea was to have a face-to-face wine tasting in which participants would have access to a label and a different tasting note, in order to analyze the influence that the sensory description has on the consumption experience. Due to the covid-19 Pandemic it was not possible, the Lab was closed for security reasons and the best alternative was the online survey, using two different versions. This research is limited by the sample used. A large part of the respondents belongs to the age group between 21 and 30 years old, which may bias the results, as this is not the age group that most consumes wine in Portugal, the conclusions are biased towards “novices” in wine, considering the low level of specialization of the sample sector. Men are also the gender that consumes the most wine, however more responses were obtained from women to this survey. In this sense, it would be interesting to replicate the experiment with a more diverse group of participants and assess whether the results would still be supported or even if they could have more relevance in terms of explained variance. Another limitation is the sample size, which is small (N = 200). In a future research, it would be advantageous to carry out a less extensive survey, as there were many people who left the survey incomplete, which meant that only 200 responses could be used (whose progress rates were 100%). Throughout the investigation, there was always a constant concern to ensure that no unwanted clues could affect consumers' emotions and choices. To the best of the investigator's ability, most factors were controlled, with labels shown for the same wine, and a 54 Q6Em que locais prefere beber vinho? (Pode selecionar mais que uma opção) o Em casa o Nos restaurantes o Nas esplanadas o Em bares o Eventos/ ocasiões especiais Q7Por favor, ordene por ordem de preferência os locais onde compra vinho (arrastando as opções) Restaurantes Garrafeiras Online Supermercados Q8Por favor, selecione os atributos que considera mais importantes quando compra vinho (pode escolher até 4 opções): Quando compro vinho para mim o Tipo de vinho (branco, tinto, etc) o Região de origem o Marca o Preço o Informação do rótulo o Design do rótulo o Prémios e distinções Quando compro vinho para beber com amigos o Tipo de vinho (branco, tinto, etc) o Região de origem o Marca o Preço o Informação do rótulo o Design do rótulo o Prémios e distinções Quando compro vinho para beber com familiares o Tipo de vinho (branco, tinto, etc) o Região de origem o Marca o Preço o Informação do rótulo o Design do rótulo o Prémios e distinções Q9Por favor, selecione as fontes de informação que considera mais importantes quando compra vinho (pode escolher até 3 opções): 55 Quando compro vinho para mim o Anúncios publicitários o Informação no rótulo o Conselhos de staff ou produtores o Conselhos de amigos/familiares o Notícias e artigos o Só compro vinhos que conheço Quando compro vinho para beber com amigos o Anúncios publicitários o Informação no rótulo o Conselhos de staff ou produtores o Conselhos de amigos/familiares o Notícias e artigos o Só compro vinhos que conheço Quando compro vinho para beber com familiares o Anúncios publicitários o Informação no rótulo o Conselhos de staff ou produtores o Conselhos de amigos/familiares o Notícias e artigos o Só compro vinhos que conheço Q10Por favor, avalie numa escala de 0 a 5 o quanto gota de vinho branco o 1 o 2 o 3 o 4 o 5 Q11Por favor, observe o seguinte vinho branco e leita atentamente o seu rótulo: (V1 observed this image) (V2 observed this image) 56 Responda às questões tendo em conta a informação disponível no rótulo: Totally disagree Disagree Slightly disagree Do not agree nor disagree Agree Slightly agree Totally agree From the description, I think I would like this wine o o o o o o o From the description, I think I would buy this wine o o o o o o o The label description is enough to form an idea about the taste of this wine o o o o o o o I need more information about the flavor o o o o o o o Q12Por favor, observe o rótulo do mesmo vinho branco anterior, mas com nota de prova destacada: (V1 and V2 observed the same image) Responda às próximas questões tendo em conta a notas de prova evidenciada pelo retângulo amarelo: Totally disagree Disagree Slightly disagree Do not agree nor disagree Agree Slightly agree Totally agree The tasting note on the label allows me to form a more precise idea about this wine o o o o o o o I consider this information important to make a decision about this wine o o o o o o o How would this information influence the appreciation/tasting of this wine? (from very negatively to very positively) o o o o o o o I prefer to discover the flavors of wine for myself rather than being guided in this discovery o o o o o o o By identifying the aromas I like, it will help me in the future when choosing wines. o o o o o o o 57 Q13Responda, numa escala de 1 (discordo totalmente) a 7 (concordo totalmente), qual o seu grau de concordância com cada uma das afirmações seguintes sobre experiências sensoriais associadas ao vinho (ex:sabor, textura, cor, etc): Q13.1Interessa-me ler as descrições sobre as experiências sensoriais que aparecem nos rótulos dos vinhos o 1-Discordo totalmente o 2-Discordo o 3-Discordo ligeiramente o 4-Não concordo nem discordo o 5-Concordo ligeiramente o 6-Concordo o 7-Concordo totalmente Q13.2Nunca dei importância à descrição sensorial dos rótulos por não saber que existiam o 1-Discordo totalmente o 2-Discordo o 3-Discordo ligeiramente o 4-Não concordo nem discordo o 5-Concordo ligeiramente o 6-Concordo o 7-Concordo totalmente Q13.3Leio sempre as descrições das experiências sensoriais nos rótulos dos vinhos que compro, quando têm essa informação o 1-Discordo totalmente o 2-Discordo o 3-Discordo ligeiramente o 4-Não concordo nem discordo o 5-Concordo ligeiramente o 6-Concordo o 7-Concordo totalmente Q13.4Prefiro comprar vinhos que tenham informações sobre as experiências sensoriais na embalagem o 1-Discordo totalmente o 2-Discordo o 3-Discordo ligeiramente o 4-Não concordo nem discordo o 5-Concordo ligeiramente o 6-Concordo o 7-Concordo totalmente Q13.5As descrições das experiências sensoriais dos rótulos influenciam a minha escolha no momento de comprar o vinho o 1-Discordo totalmente 58 o 2-Discordo o 3-Discordo ligeiramente o 4-Não concordo nem discordo o 5-Concordo ligeiramente o 6-Concordo o 7-Concordo totalmente Q13.6Já comprei um vinho com base na descrição de sabores e aromas presente no rótulo o 1-Discordo totalmente o 2-Discordo o 3-Discordo ligeiramente o 4-Não concordo nem discordo o 5-Concordo ligeiramente o 6-Concordo o 7-Concordo totalmente Q13.7Se souber à priori os sabores presentes no vinho, isso influenciará a forma como o irei apreciar o 1-Discordo totalmente o 2-Discordo o 3-Discordo ligeiramente o 4-Não concordo nem discordo o 5-Concordo ligeiramente o 6-Concordo o 7-Concordo totalmente Q13.8As descrições sensoriais do rótulo acabam por influenciar a experiência de consumo do vinho o 1-Discordo totalmente o 2-Discordo o 3-Discordo ligeiramente o 4-Não concordo nem discordo o 5-Concordo ligeiramente o 6-Concordo o 7-Concordo totalmente Q13.9Sinto-me capaz de identificar os sabores e aromas que gosto num vinho o 1-Discordo totalmente o 2-Discordo o 3-Discordo ligeiramente o 4-Não concordo nem discordo o 5-Concordo ligeiramente o 6-Concordo o 7-Concordo totalmente 59 Q14Responda, numa escala de 1 (discordo totalmente) a 7 (concordo totalmente), qual o seu grau de concordância com cada uma das afirmações seguintes sobre o design dos rótulos: Q14.1O design do rótulo influencia a minha escolha no momento de comprar o vinho o 1-Discordo totalmente o 2-Discordo o 3-Discordo ligeiramente o 4-Não concordo nem discordo o 5-Concordo ligeiramente o 6-Concordo o 7-Concordo totalmente Q14.2Quando não sei qual vinho comprar, prefiro comprar vinhos que tenham rótulos bonitos o 1-Discordo totalmente o 2-Discordo o 3-Discordo ligeiramente o 4-Não concordo nem discordo o 5-Concordo ligeiramente o 6-Concordo o 7-Concordo totalmente Q14.3Rótulos com um aspeto simples e cuidado transmitem a sensação de qualidade do vinho o 1-Discordo totalmente o 2-Discordo o 3-Discordo ligeiramente o 4-Não concordo nem discordo o 5-Concordo ligeiramente o 6-Concordo o 7-Concordo totalmente Q14.4Já comprei vinho que não conhecia por gostar do rótulo o 1-Discordo totalmente o 2-Discordo o 3-Discordo ligeiramente o 4-Não concordo nem discordo o 5-Concordo ligeiramente o 6-Concordo o 7-Concordo totalmente Q15Género o Feminino o Masculino o Outro o Prefiro não responder 60 Q16Idade Page | i