Assessing the Australian abalone market: a best-worst approach on 46 attributes

Australian consumers’ appetite for fishery and aquacultural products has grown consistently (ABARES, 2023a), increasing from 21 kg of fish and seafood consumed per person annually in 1997 to 26 kg in 2017 (Our World in Data, 2021). This growth would contribute positively to the environment, considering that fisheries and aquaculture emit a comparatively low carbon emission, as low as a tenth of the livestock industry (Hilborn et al., 2018). Understanding consumers’ preferences would fuel the growth of fishery and aquacultural products. This is especially relevant in Australia, considering that the consumption of fishery and aquacultural products remains low, ranking as the second least consumed meat (Tuynman et al., 2023). As such, there has been a push for a better understanding of the Australian market’s consumption patterns and motivations when purchasing fishery and aquacultural products (Christenson et al., 2017).

There has been a notable increase in research looking at Australian consumers’ overall perception toward seafood (e.g. McManus et al., 2007), drivers and barriers of seafood consumption (e.g. Birch et al., 2012) and acceptance of seafood (e.g. Danenberg et al., 2012). Much research primarily focuses on more everyday consumption and finfish categories, such as barramundi, salmon and tuna (e.g. Lawley et al., 2012; Grieger et al., 2012; Rahmawaty et al., 2013). There exists a need to explore the mollusc aquaculture market as it is also a promising alternative source of protein when it comes to environmental impact. According to Hilborn et al. (2018), mollusc aquaculture outperforms other aquaculture products (e.g. shrimp, catfish, tilapia and salmon) on energy usage, greenhouse gas production and acidification.

One high-value mollusc product that is often overlooked is abalone, which contributes greatly (approximately AUD 150 million annually) to the Australian economy (ABARES, 2023b). There is a large untapped market domestically in Australia since Australian consumers are mostly indifferent toward abalone (Tuynman et al., 2024). However, it remains unclear what product attributes are inherently essential to consumer decision-making concerning abalone. Investigating the salient product attributes could be of value for the Australian fishery and aquaculture industry since a stronger presence in the domestic market would contribute to their resilience and sustainable growth. This, will in turn, directly propel the growth of fishery and aquaculture consumption in Australia, and thus, potentially alleviate the environmental impact of the food supply chain.

Product attributes, in the form of cues, have been widely recognised as essential to consumers’ choice and experience with food products (e.g. Bernués et al., 2003; Bredahl, 2004; Grunert, 1996; Nocella et al., 2010; Van Loo et al., 2011). Our narrative review of prior studies on seafood (i.e. salmon, tuna and lobsters) and review of seafood products available in major Australian supermarkets (e.g. Aldi, Coles and Woolworths) indicated that there could be more than 80 product cues (e.g. flavour, brand, price and origin) that may affect consumers’ decision-making process (refer to Appendix A). Determining the relative importance of various product cues has been pivotal to understanding consumers’ decision-making (Louviere et al., 2000). To date, only a small number of studies have looked at the salience of abalone product cues (e.g. size, taste, texture, colour and price), but these studies are often based on the perspective of producers and distributors (Gao et al., 2002; Hwang et al., 1997; Oakes and Ponte, 1996).

Against this backdrop, the current study adopts cue utilisation theory (e.g. Cox, 1967; Jacoby et al., 1971; Olson and Jacoby, 1972) as the theoretical underpinning, and seeks to acquire Australian consumers’ preferences (i.e. rating) of 15 intrinsic and 31 extrinsic cues. Using the best–worst scaling method (BWS), a form of discrete choice experiment, the study will mimic an actual choice process and estimate the product cues’ relative importance (i.e. utility score). We then use the cues’ utility scores as the basis to conduct a two-stage clustering analysis involving hierarchical and k-means clustering techniques. This enables an accurate delineation of different consumer typologies in the Australian luxury seafood market based on behavioural data rather than simply relying on socio-demographic characteristics. Findings from the research will offer unprecedented insights into the fishery and aquaculture industry’s strategic planning and marketing, contributing to the optimisation of products and marketing communication and positively influencing consumers’ purchase intention toward seafood products.

Understanding what factors positively drive consumers’ perceived quality has been central to academic and industry research. Cue utilisation theory offers a foundation to answer this question. The theory suggests that consumers rely on product cues, such as price, brand names and colour, as bases to develop their impression of the product and guide their decisions (e.g. Cox, 1967; Jacoby et al., 1971; Olson and Jacoby, 1972). According to Olson and Jacoby (1972), each product cue fundamentally has its magnitude of impact and relevancy to consumer usage). Past research also suggests that product cues not only shape consumer preferences but also predict their consumption experience (e.g. Kallas et al., 2019). For instance, Ye et al. (2020) found that the glossiness of packaging can serve as a produce cue to signal the tastiness (i.e. glossiness heightens perceived tastiness).

Product cues can be classified into intrinsic and extrinsic. Intrinsic cues are the product’s properties that cannot be changed without altering the product, whereas extrinsic cues are product-related but not physically part of the product elements (e.g. Shirai, 2020). Intrinsically, Lawley et al. (2021) found that aroma, colour and texture were the key quality indicators for finfish, specifically barramundi, among Australian consumers. Similarly, Murray et al. (2017) also indicated that Canadian consumers often associate aroma, texture and appearance with eating quality. Extrinsically, many cues, including name, price, organic certification, farming practice and safety standards, are essential in shaping consumers’ perceptions of fishery and aquaculture products (e.g. Ankamah-Yeboah et al., 2016; Christian et al., 2013; Whitmarsh and Palmieri, 2008).

In a recent review, Krishna and Elder (2021) propose an overarching framework explaining the complex process of food choice and experience. Their framework suggests that intrinsic cues, such as the colour and odour of the food, serve as bottom-up cues and impact the perceptual experience of consuming the food directly. Meanwhile, extrinsic cues can function as top-down cues that reinforce the perceptual experience of consumption. Thus, by employing the relevant cues, firms can heighten consumers’ perception of the quality of their products and, in turn, positively influence preferences and choices. Furthermore, by not over-employing irrelevant cues, firms can avoid overloading consumers with information and risk losing their attention (e.g. Chen et al., 2010).

Research on product cues in the mollusc aquaculture context remains scarce and even more so for the Australian market. There was only a handful of studies looking at salient mollusc product cues. Nguyen et al. (2015) found that consumers prefer live products over chilled products for mussels and oysters. Meanwhile, Carlucci et al. (2017) found that quality certification labels are an effective cue for consumers in determining product quality. Distant results in crustacean aquaculture (i.e. lobsters) indicate that meat content, texture and size are the primary cues that significantly impact Chinese consumers’ intent to purchase lobsters (Wang et al., 2021).

To date, there are only a few studies (albeit outdated) on the relevant cues impacting consumers’ purchase decisions of abalone products, such as size and texture (e.g. Gao et al., 2002; Hwang et al., 1997). However, findings from these studies were mostly derived from the perspective of producers and distributors. Most studies investigate production methods and treatments to achieve optimal sensory properties and intrinsic cues of abalone. For instance, Sanchez-Brambila et al. (2002) looked at tenderisation treatments and their effect on texture and taste, while Dong et al. (2018) examined the effect of temperature-time treatments on texture. To our knowledge, research on the impact of abalone product cues on consumer’s decision-making process remains scarce.

It remains unclear what intrinsic and extrinsic cues Australian consumers use to evaluate abalone products and determine their preferences. Therefore, a comprehensive investigation of product cues’ saliency (both intrinsic and extrinsic) would provide significant insights for the producers to deliver optimal product profiles, strengthen competitive advantage and encourage the adoption of mollusc, more specifically abalone, as an alternative source of proteins.

Much research on the salience of product cues relies on direct rating scales (e.g. Likert scale). However, such a method carries certain disadvantages, such as homogeneous discrimination between alternative attributes, extreme responses and low reliability (see Tavares et al., 2010). In contrast, BWS offers better reliability and validity (Flynn and Marley, 2014) by forcing consumers to select the most and least important items (e.g. attributes, features and profiles) from multiple sets of different items. These forced trade-offs between the items mimic a real-life purchase decision, thereby increasing the discrimination among the items’ importance (e.g. Cohen, 2003), avoiding extreme bias (e.g. Jaeger and Cardello, 2009) and generating more reliable annotations (Kiritchenko and Mohammad, 2017). A recent direct comparison between BWS and the Likert scale by Heo et al. (2022) supported the notion that BWS can achieve much better discrimination among attributes.

As such, BWS has been increasingly adopted to study attributes of raw seafood (e.g. Nguyen et al., 2015; Sajiki and Lu, 2022), seabream and seabass products (e.g. Cantillo et al., 2021). However, BWS is often limited in the number of items that could be tested in order to avoid overloading respondents’ cognition (see Tavares et al., 2010). For instance, a BWS experiment of four product cues could incur a total of six permutations of pairwise comparisons that respondents need to examine. Considering that there are more than 80 product cues that could be tested, the current study designs a three-phase testing protocol (see details in “Materials and methods” section). This study will be the first to examine up to 46 items (i.e. product cues) of abalone products using the BWS method. We offer a novel and rigorous protocol to handle a large number of choice factors, while being able to maintain a high level of discriminant validity in our results.

Above and beyond product cues, identifying different consumer typologies within a market enables a richer understanding of consumers’ consumption motives. Past studies, especially in the context of food consumption, often delineate consumer typologies on the basis of high-level socio-demographic characteristics (for a review, see Cleveland et al., 2011). Those bases could include product knowledge (Rortveit and Olsen, 2007), level of involvement (e.g. Verbeke et al., 2007), consumption habit (Koutsimanis et al., 2012; Verbeke et al., 2007) and food consumption motives (Honkanen and Frewer, 2009; Milošević et al., 2012). Despite its merits, such an approach does not warrant a consistent and fullest insight into different consumer groups and their unique motivations and expectations (Aaker, 1995; Onwezen et al., 2012; Wedel and Kamakura, 2000).

This limitation is also observed in the literature on fishery and aquaculture products. For instance, Birch and Lawley (2012) classified Australian consumers based on their consumption frequency. They found that irregular fish consumers were more likely to perceive functional, social and psychological risks (associated with fish consumption) compared to regular consumers. Similarly, Wang and Somogyi (2020) segmented luxury seafood consumers based on whether they seek food value (i.e. appetite, health and novelty) or symbolic value (i.e. status, network or lifestyle). Focusing only on socio-demographics and consumption habits limits the current understanding of the determinants of abalone consumption. This study, therefore, will determine different consumer typologies based on the importance that they attach to different intrinsic and extrinsic cues (i.e. utility scores based on BWS results).

Such an approach has been shown to be more reliable than relying on socio-demographic characteristics. It can provide steady, consistent and valuable insights into different consumer segments and what they seek to fulfil or their expectations (Aaker, 1995; Onwezen et al., 2012; Wedel and Kamakura, 2000). As such, our study seeks to utilise the cues’ importance to identify different consumer segments existing in Australia. By identifying consumer segments using cues’ utility scores (i.e. choice of product cues in the decision-making process), our study can objectively pinpoint each segment’s unique drivers for preferences and characteristics. Taken together, the current study seeks to answer the following question:

A total of 200 Australian abalone consumers were recruited via Qualtrics. A stratified sampling technique was employed to recruit the respondents based on the Australian Bureau of Statistics census data (ABS, 2022a, b) regarding the Australian population’s age, gender, income and prior consumption experience with abalone at a state-by-state level. However, for the study, all participants must live in Australia and have previously purchased abalone products at least once over the past 24 months. Our final sample was relatively balanced in terms of gender (54% female), income (66% respondents with income higher than ∼$50,000) and age (M = 39.52, SD = 13.2, range = 19 to 75). In total, 65.5% of our participants identified themselves as Australian, whereas 23.5% identified with Asian origins. Most of our participants were well-educated, with 78.5% holding at least a diploma or higher degree. The majority (i.e. 79.5%) of the respondents resided in metropolitan areas, had at least two people in their household (93%), and had the prime responsibility for purchasing groceries in their household (91.5%).

The research procedure comprised three main phases. The first phase involved the selection of abalone product cues for testing. This phase aimed to develop a list of relevant abalone product cues that were informed by prior literature, existing commercial products and industry experts. The second phase involves the development of BWS choice sets and procedures. Rstudio was used to generate choice sets for the BWS experiment. Finally, the third phase involved transferring the choice sets to Qualtrics for data collection and analysis. The following sections will outline our research procedure in detail.

Following Sakolwitayanon et al. (2018), we conducted a three-step exploratory study to develop a list of relevant abalone product cues.

Firstly, we used the Google Scholar database to identify relevant peer-reviewed articles that directly examine abalone-related attributes (i.e. product cues) in consumption contexts. However, only a few studies investigated abalone products’ cues. Thus, we expanded our search to other seafood and aquaculture products (e.g. seafood, lobster and salmon). Certain keywords were used to identify relevant articles. “Seafood”, “attribute” and “cue” must be mentioned in the abstract, whereas “consumption”, “quality”, “perception”, “purchase” and “experience” must be presented in the body. All abalone product cues and corresponding sources (i.e. peer-review articles) were listed in Appendix A.

Secondly, we reviewed products (e.g. abalone, lobster, salmon, etc.) that are being sold online and in Australian supermarkets (e.g. Woolworths, Coles, IGA and ALDI). From both reviews (i.e. literature and industry practice), we found a total of 89 abalone product cues (refer to Appendix A).

Thirdly, three researchers and two industry experts reviewed and narrowed the list down to a more manageable list of product cues to be tested. This process involved removing duplication, difficult-to-comprehend cues (e.g. shell pigmentation, metallic taste and min weight per piece), legally compulsory cues (e.g. HACCP cue), technical-oriented cues and non-significant cues (refer to Appendix A). We finally acquired a total of 46 cues that may generate 46, 69 or 138 choice sets in a balanced incomplete block design (BIDD) (e.g. Takeuchi, 1962). The 46-choice-sets design contained ten cues in each choice set, which is not appropriate for BWS, as four to six items per choice set is the most optimal design (e.g. Cohen, 2009). Meanwhile, the 69-choice-sets and 138-choice-sets designs would put a significant cognitive burden on consumers and negatively affect their response quality. Thus, we decided to examine 15 intrinsic cues and 31 extrinsic cues separately (refer to Table 1). All cues were presented in their original form from the literature and existing products.

The current study adopted a BIBD to develop the BWS experiment. A BIBD allows researchers to avoid over or under-representation of any cues (Lee et al., 2007, 2008; Massey et al., 2015), better control for “context effects” (Lee et al., 2007, 2008) or “demand effects” (Massey et al., 2015; Mori and Tsuge, 2017). That is because there is a fixed number of cues in each choice set, and all cues occur the same number of times.

Employing BIBD, two series of choice sets, with one examining 15 intrinsic cues and the other examining 31 extrinsic cues, were generated. Rstudio packages, including “support.BWS” (Aizaki and Fogarty, 2019) and “crossdes” (Sailer, 2004), were used to assist with the development of choice sets and questionnaires. The first series of choice sets (i.e. intrinsic cues) follows a 7 (options) by 15 (choice sets) by 3 (times of presentation) design. In this design, there were 15 choice sets with seven cues in each, and each cue appeared three times randomly across the whole series. The second series of choice sets (i.e. extrinsic cues) follows 6 options by 31 choice sets by 6 times of presentation design.

The choice sets were then transferred to the online survey programmed in Qualtrics. The online survey is comprised of four parts. The first part filtered participants based on age, gender, income, previous consumption and location to acquire a representative sample of the average Australian consumers of abalone products. The second and third parts contained two BWS experiments, one for 15 intrinsic and another for 31 extrinsic cues. Participants’ socio-demographics and consumption frequency were collected in the fourth part.

Three steps of data analysis were conducted to determine the cues’ importance. Firstly, we tested data integrity to ensure there were no design and data processing errors. Data integrity was checked using two tests. Firstly, the maximum score of one cue chosen as best or worst must be in the range of -r × N to r × N (i.e. r refers to the number of times each cue appears across the choice sets). Secondly, the “sum of best” of all cues must be equal to the “sum of worst” of all cues and equal b × N.

After ensuring the data integrity, we calculated each cue’s aggregated BWS score, ratioscale, and relative importance. The aggregated BWS score determined the ranking of each cue’s importance. The aggregated BWS score was calculated using the following formula: TB – TW (TB and TW refer to the number of times one cue was chosen as best and worst, respectively).

The ratioscale was then calculated using the following formula √(TB/TW). Ratioscale indicates the probability of a cue being selected as most important. To avoid dividing by 0 (i.e. a cue might have never been selected as the Least Important by a participant), researchers added 0.5 to the TW with a value of 0 (Cohen, 2009). The cue with the highest ratioscale was benchmarked as 100 to compare the relative importance of other cues (Cohen, 2009). The relative importance generated insight into the probabilistic nature of each cue’s importance (i.e. how important was cue A compared to the others) on a scale of 0–100.

Hierarchical and k-means cluster analyses were conducted twice to identify unique consumer segments based on the importance of intrinsic and extrinsic cues separately. Hierarchical cluster analysis was conducted first using Ward’s method (Punj and Stewart, 1983) and squared Euclidean distances (Knezevic et al., 2019) to generate a dendrogram, which was then used to identify the optimal number of clusters (i.e. segments). The optimal number of segments was identified by creating a cut-off point on the dendrogram, where there was a relatively large jump in the distance (Azabagaoglu and Gaytancioglu, 2009; García-Solano et al., 2015; Bodor et al., 2021; Tullis and Albert, 2013). The solutions (i.e. number of segments) were then validated in k-means cluster analyses. We conducted MANOVA to confirm whether there were significant differences among the segments based on Wilk’s Lambda (Hair et al., 2010). Post hoc Tukey was used to identify the product cues that defined each segment.

From the aggregated score, we identified that “ocean-fresh flavour”, “texture” and “umami” were the most important intrinsic cues, with “ocean-fresh flavour” being 13 and 50% more important than “texture” and “umami”, respectively (refer to Table 2). Meanwhile, “with shell or without shell”, “colour of the foot” or “shape of the meat” were the least important cues. Specifically, they were at least five times less important than “ocean-fresh flavour”.

“Rich in nutrients”, “country of origin” and “responsible and sustainable farming” were the three most important cues based on aggregated scores (refer to Table 3). There was also a relatively small difference among the three cues regarding their relative importance. Meanwhile, the least important cues were “cooking and serving tips”, “food awards won” and “Halal approved”, which were at least 3.7 times less important than “rich in nutrients”.

Based on the dendrogram generated by hierarchical cluster analysis, we determined a five-cluster solution as the most acceptable solution. This solution was then validated using k-means cluster analysis. The five-cluster solution took the least number of iterations required to reach convergence, indicating its appropriateness. MANOVA was conducted and generated Wilk’s lambda value (0.025) with an F-ratio (18.604), giving a p-value of less than 0.001. This confirmed the significant heterogeneity among the clusters (i.e. segments) concerning abalone product intrinsic cues.

The five unique segments were named “appearance lover” (N = 55), “sweet and juicy eater” (N = 37), “conventional seafood buyers” (N = 24), “ocean-fresh flavour advocate” (N = 57) and “size matters buyer” (N = 27). Referring to Table 4, the “ocean-fresh flavour advocate” preferentially attended to “ocean-fresh flavour” and “umami” cues significantly higher than other segments. Meanwhile, the “appearance lover” emphasised the importance of appearance cues (i.e. colour of the foot, lips, shape of the meat, etc.) significantly higher than other segments. The “sweet and juicy eaters” were characterised by “sweetness” and “juiciness” cues, whereas the “size matters buyer” paid attention to “net weight” and “size and weight per piece”. The smallest segment was the “conventional seafood buyers”, who emphasised the “colour of the meat” and “aroma” compared to others.

Following the same analysis procedure for the intrinsic cues, a six-cluster solution was identified as the most appropriate. MANOVA analysis generated a Wilk’s lambda value (0.012) with an F-ratio (7.68), giving a p-value of less than 0.001, confirming the significant heterogeneity among the segments. We named the segments from cluster 1 to cluster 6 as environmentalist, health-conscious, utilitarian, first-in-first-out, naturalist and regio-centric buyers, respectively. Referring to Table 5, the “environmentalist buyers” emphasised the importance of “traceability information” and “harvest method”. Meanwhile, the “health-conscious buyers” emphasised “rich in nutrients”, “nutrition information” and “health star ratings”. We had the “regio-centric buyers”, who only paid attention to abalone products’ “regionality” cue. In contrast, the “naturalist buyers” preferentially paid attention to “no artificial additives”, “antibiotics free”, “no preservatives” and “no GMO”. Results of other segments’ characteristics can be reviewed from Table 5.

To describe and identify significant differences among the segments’ socio-demographic characteristics, a Crosstab comparison with Chi-square and Phi and Cramer’s V enabled was used. Chi-square tests (with Yates Continuity Correction) indicated no significant differences between the five segments regarding their age, ethnicity, living location, employment status, abalone purchase frequency and grocery purchasing responsibility (i.e. who is responsible for buying food). The results indicated that “appearance lovers” and “size matters” were the only two segments with a female-dominated population. Additionally, the “appearance lovers” also possessed a significantly higher number of people with a high-income level (i.e. higher than $90,000) and higher education levels than other segments (refer to Table 6).

No significant differences were found between the six extrinsic cue preference segments concerning age, gender, income, employment status and grocery purchasing responsibility. We found that the “first-in-first-out” and “regio-centric” segments had a significantly higher proportion of people living outside of metropolitan city areas. Meanwhile, the “utilitarian” and “naturalist” had a significantly higher number of Asians in their population compared to other segments. Purchasing frequency of abalone was higher in the “health-conscious” and “regio-centric” segments. Compared to other segments, the “first-in-first-out” had a higher proportion of people with a lower level of education (i.e. secondary school and below). Details regarding each segment’s socio-demographic characteristics can be found in Table 7.

To our knowledge, this is the first study to explore the importance of an exhaustive list of product cues (15 intrinsic and 31 extrinsic cues) in the fishery and aquaculture literature. Our study, therefore, offers timely and meaningful implications by generating a wide range of insights via more granular data with superior discrimination concerning the importance of cues and heterogeneity of the consumer market for abalone products in Australia.

Replicating findings from past studies in the fishery and aquaculture literature, the study demonstrates that food-related decisions are innately driven by sensory experience (e.g. flavour and texture) regardless of product categories (e.g. Freitas et al., 2020; Nurliza et al., 2021; Wang et al., 2018). Interestingly, we find that Australian consumers prefer to be “told” explicitly about the product quality (i.e. ocean fresh flavour, umami and texture). Our results show that cues such as “ocean-fresh flavour”, “umami” and “texture” were 2.21, 1.94 and 1.46 times more important than “colour of the meat”, respectively. This challenges the existing belief that conventional intrinsic cues (e.g. colour, shape and size) are reliable predictors of consumers’ choice of seafood, such as lobster (Wang et al., 2021), tilapia (Darko et al., 2016) and barramundi (Lawley et al., 2021).

The low importance of conventional intrinsic cues (i.e. colour, shape and size) could be indirectly attributed to Australian consumers’ lack of objective knowledge to evaluate fishery and aquacultural products (e.g. Guo and Meng, 2008). In this research, Australian consumers, on average, purchased abalone products less than once per month; thus, they were not familiar with abalone products and may find it challenging to evaluate abalone products using conventional intrinsic cues (Birch and Lawley, 2014). This finding implies that point-of-purchase communication (e.g. packaging and salesperson) should explicitly communicate the consumption experience (e.g. experience the umami) to positively influence consumers’ impressions of fishery products, especially when they are not familiar with the products.

The study also extends cue utilisation theory by showing that not all cues are evaluated in the same manner. Consumers preferentially attend to simple and easy-to-interpret cues. Our results show that when two cues communicate the same attribute (e.g. naturalness), the easy-to-interpret cue may have a stronger predictive value toward the judgment process than a difficult-to-interpret cue. For instance, “no artificial additives”, “no antibiotics” and “no preservatives” were 1.44, 1.13 and 1.01 times more important than “organically grown”, respectively. It is possible that consumers tend to reject cues that lack vividness and comprehensibility (e.g. Nisbett and Ross, 1980; Westbrook and Fornell, 1979). This is supported by Steenkamp (1989) who suggested that food consumers often prefer simple cues with detailed descriptions to infer product quality immediately.

The current study is also the first to illustrate the heterogeneity of Australian consumers in the abalone market and contributes significantly to understanding luxury fishery and aquaculture market heterogeneity. Using cues’ utility scores to delineate consumer typologies, we identified five distinct consumer segments based on the importance of intrinsic cues of abalone: appearance lover, sweet and juicy eater, conventional seafood buyers, ocean-fresh flavour advocate and size matters buyer. Six unique consumer segments were also identified based on extrinsic cues of abalone: environmentalist, health-conscious, utilitarian, first-in-first-out, naturalist and regio-centric buyers. These segments all possess unique demands and motivations when purchasing mollusc products (i.e. abalone), while showing no significant differences regarding their socio-demographic characteristics. This study illustrates the clear advantages of segmenting consumers based on the importance of product cues. We show that conducting cluster analyses based on product cues’ utility scores allows researchers to generate more robust insights into consumers and their attention to specific product features (e.g. Szymkowiak et al., 2020; Gosine and McSweeney, 2019; De Pelsmaeker et al., 2017).

From the methodological point of view, this study is the first to apply BWS to examine a large number of items (up to 46 cues). The application of BWS allows us to elicit consumers’ preferences through an array of trade-offs made by the consumers in multiple hypothetical situations (e.g. Louviere et al., 2000). Since food-related choices are often made by considering the spectrum of attributes, examining the various trade-offs made by consumers allows researchers to fully understand the antecedents of preferences and choices. By designing and validating a three-phase testing protocol, our study lays the groundwork for future research into food-related behaviours that encounter a large number of choice factors.

The current study demonstrates the significance of eating quality in Australian consumers’ judgment of abalone products, offering support to the Australian abalone industry’s current strategic focus. We found that consumers highly valued cues such as “ocean-fresh flavour”, “umami” and “texture”. Interestingly, Australian consumers did not use conventional search cues like colours, size and shape to guide their evaluation. This contradicts the industry’s current belief and practice, as they often rely on the use of the meat colour, lips and shape of the meat to communicate their products’ quality, which is especially relevant for export markets. The current research implies that explicit communication of eating quality (e.g. experience the umami) on packaging may lead to more positive expectations of the eating quality and heighten the buying intention of Australian consumers than simply relying on intrinsic cues of the products.

In our study, consumers also use “quality grading” to guide their evaluation. This further cements the notion that Australian consumers do not fully trust their subjective assessment of abalone via conventional intrinsic cues. This proposes an opportunity to develop an industry-wide grading system for abalone to differentiate consistent and high-quality Australian products. The proven success of industry-wide quality grading systems in other industries like beef or lamb (D’Souza et al., 2017; Lyford et al., 2010) indicates the potential benefits for the Australian abalone industry to adopt a similar approach.

In line with prior research, the current study illustrates the emerging importance of health benefits, sustainability and naturalness in food-related decision-making (Ankamah-Yeboah et al., 2016; Birch et al., 2012; Pérez-Ramírez et al., 2015). Our study shows that health benefits, sustainability and naturalness are relatively similar in their importance to Australian consumers. This suggests that Australian consumers are not willing to make a trade-off between these attributes, indicating that the Australian abalone industry should adopt healthiness, sustainability and naturalness as the future strategic pillars.

Our results show that consumers tend to reject information lacking vividness and comprehensibility. The evidence suggests that effective communication requires the Australian abalone producers to use concrete and easy-to-interpret language. For instance, brands could explicitly explain their production methods, beyond certificates and basic information, to elicit positive consumer sentiments and heighten purchase intention. This is particularly relevant for the food industry, considering their credibility is being challenged (e.g. Pelly et al., 2020), prompting the significance of transparent communication to capture consumer trust.

One contradiction observed in our study is that Australian consumers indicate that the abalone species is one of the least important quality indicators. This is interesting since the literature often highlights species as a strong determinant of seafood selection (Alfnes et al., 2018). Abalone species is also a cue that many Australian producers use as a quality signal and competitive advantage. Our study shows that this strategy might not be as effective as they expect, since Australian consumers are less familiar with abalone products (Alfnes et al., 2018). Therefore, there could be value in developing educational campaigns to equip Australian consumers with a better understanding of the differences between abalone species and to build stronger marketing associations with specific abalone species grown in Australia.

From our segmentation analyses, we also identify the highly valuable consumer segments that are large in size and offer the potential for a premium return on investment. We find that 70% of the Australian market is comprised of hedonic-oriented consumers (i.e. healthiness, naturalness and sustainability). They also prefer abalone products packed with and marketed to deliver ocean-fresh and umami flavour. By anchoring on consistent eating quality, healthiness, naturalness and demonstrable sustainability, Australian abalone producers could strengthen their positioning and image in the domestic market against international competitors. Furthermore, hedonism-oriented consumers are also less likely to be influenced by price, allowing the producers to command a higher price without affecting their sales. These granular insights enable the industry to optimise targeting strategies.

A deep dive into the market heterogeneity also shows that sustainability is not necessarily a make-or-break issue for most Australian consumers. Although “responsible and sustainable farming practice” was rated as the third most crucial attribute overall, the segmentation analysis reveals that there is only one group (i.e. environmentalist buyers) that highly emphasises environmental and sustainability issues. Furthermore, this group only accounts for 18.5% of the market size, signalling that sustainability in food supply chain still has a long way to become the top-of-mind cues for the mass market. This finding coincides with many prior studies across different food contexts and countries (e.g. Yue et al., 2024). The evidence emphasises a need for increasing attention and exposure to sustainability issues associated with the food supply chain across various media platforms. This should raise individuals’ awareness, and in turn, their purchase intention toward sustainable food choices (Diprose et al., 2018; Lee et al., 2019).

The current study is not without its limitations. Firstly, the sample size employed in this study was relatively small; thus, a larger sample size could be employed in the future to increase the reliability and representativeness of the results. Secondly, the study did not compare intrinsic and extrinsic cues directly due to the large number of items or product cues being tested. Future studies could extend on our results and directly compare a smaller number of both intrinsic and extrinsic cues. Another solution is to apply a nested BIDD (Deppe et al., 2001), in which researchers could examine a large number of items without the constraint that each respondent must view every attribute. Finally, BWS does not estimate the optimal level of each cue (i.e. high versus low price), which can provide meaningful insights to design the optimal product profile. Therefore, other methods (e.g. conjoint experiment and BWS cases 1 and 2) could be considered for future studies to determine the optimal level of cues and their economic values, allowing the identification of a high-value combination of cues to aid a firm’s decision-making and priority.

Chien Duong: conceived and designed the study, collected data, performed analysis and wrote the paper. Billy Sung: conceived and designed the study, and wrote the paper. Sean Lee: conceived and designed the study, and wrote the paper. Julia Easton: conceived and designed the study, and wrote the paper.

We also thank Patrick Sheehan (Jade Tiger Abalone) for their in-kind contribution and valuable industry-related input; we thank Professor Mark Gibberd (Curtin University) for their input regarding the study design.

The supplementary material for this article can be found online.