Myths and Realities of Retail Shopper Behaviour towards ‘Sustainable’ Brands

Abstract

Sustainable marketing aims to encourage consumer behaviour that will improve the environmental and social outcomes of consumption. Despite decades of effort, however, manufacturers and retailers often see disappointing shopper responses to their sustainable marketing efforts. This paper argues that this is because many sustainable marketing efforts are hampered by false assumptions about how buyers behave in retail settings. The purpose of this paper is to take two commonly accepted sustainable marketing retail beliefs—that ‘sustainable’ brand buyers are a different type of retail shopper and that they are more loyal to these brands than shoppers of non-sustainable brands—and draw upon two established marketing empirical generalisations, the Law of Brand User Profiles and the Law of Double Jeopardy, both built over decades of research, to show that these beliefs are, in fact, myths. We use 22 sets of continuous data spanning five categories in the UK to illustrate this. Mean Absolute Deviations were used to compare the profile of sustainable brand users against non-sustainable brand users. The Dirichlet model of buyer behaviour was applied to the data to examine loyalty to sustainable brands. The results show sustainable brands are just like all other retail brands in their performance. This is a positive finding as it means they can utilise ‘regular’ brand growth knowledge to increase their market share. Overall, the paper illustrates the process and benefits of moving to a view of sustainable marketing that has stronger scientific underpinnings and that leads to more realistic shopper response expectations for retailers and manufacturers.

1. Introduction

Rapid economic development creates an environment in which marketing flourishes to serve the immediate demands of consumers and societies [1]. These demands have such power that manufacturers may be driven to serve markets in ways that could destroy the long-term prospects of their own industries. Today, many environmental and social challenges, from waste, pollution, loss of biodiversity, and climate change to human rights and fair worker conditions, are at least in part the result of consumption needs and manufacturer efforts to meet them. The challenge remains for industries and economies to move to a more environmentally and socially sustainable path over the long term [2,3,4]. Marketing, as the core discipline of consumption, is uniquely positioned to help with this transition [3,5,6], and there is recognition among managers that sustainability is both a competitive imperative [7] and an important strategic goal for C-Suite to address [8,9,10].

The shift to a longer-term perspective on the marketing exchange process requires changes to both production and consumption systems. Many organisations have already adapted their production processes for improved usage of materials, waste, and energy. Because these changes deliver clear cost savings, management can justify them on both economic and sustainability grounds. However, the economic payoffs of other sustainability initiatives are less clear. For example, on-pack certifications and endorsements, which aim to influence consumer preference, typically incur additional licensing costs for organisations, which often lack full knowledge of their impact on resultant shopper demand. For optimal change, businesses need to understand both the behaviour-change drivers of their consumers and how their own efforts will be responded to [11].

The implementation of sustainability activities, with their purported payoffs, assumes that shopper behaviour will be influenced by the perception of added value, thus giving brands a competitive advantage [1,2,5,9,12,13]. Accordingly, sustainable marketing can be thought of as driving preference for brands that profitably meet the needs of today’s shoppers without compromising future generations’ ability to meet their own needs. This definition provides a basis for evaluating marketing strategies that drive brand preference and focuses the discussion of sustainable marketing on the brand rather than at the product category level. This shifts the discussion away from a public policy context (e.g., whether demand for a product category should be legislatively controlled) and towards competitive consumer goods marketing, which is the focus of this paper.

Despite the apparently worthwhile aim of sustainable marketing, the existing literature offers little solid information about the expected return on sustainability initiatives aimed at building brand preference [14]. This is because the field of marketing science is still relatively young, especially in the sustainability literature. The resulting lack of underlying empirical evidence behind much of the sustainable marketing literature has led to the perpetuation of sustainable marketing myths that have resulted in poor marketing practices and a disappointing consumer response when applied within the retail context. The progress of sustainable marketing will be hampered until it takes account of the findings from marketing science. This paper, thus, contributes to the marketing and sustainability literature by placing retail sustainable marketing efforts within the contexts of known patterns of consumer behaviour and competitive brand choice. Such an empirically based foundation offers a way of improving manufacturer and retailer sustainability initiatives that can both achieve scale in adoption and improve return on effort.

1.1. A Marketing Science View of Shopper Behaviour

Science has a long history of transforming the fields it touches, including marketing. Scientific laws are established by documenting empirical patterns that occur in many sets of data, along with any boundary conditions that identify where the patterns do not hold. The use of scientific knowledge to guide marketing practice and research is relatively recent [15], and the literature is still full of theories that provide seemingly obvious and intuitive explanations which are not empirically borne out. Nevertheless, over the last four decades, the adoption of a scientific approach to marketing has resulted in great advances being made, including the identification of consumer behaviour patterns that hold across multiple markets over time and across countries [16,17]. These scientific marketing laws, such as Double Jeopardy and Duplication of Purchase, explain much about repeat-purchase brand-buying behaviour and the patterns we see in markets at an aggregate level, as well as their linkage to the size of market brands. They provide a picture of how buyers choose between alternative options and serve as a framework for guiding marketing efforts and evaluating responses, including those in retail shopper behaviours around sustainability.

Notwithstanding the above, marketing’s scientific laws remain relatively unknown. Despite their relevance to marketers across various contexts, including sustainability, few use them as a guide to understanding the market and predicting buyer behaviour, as underlying principles for an organisation’s marketing activity, or for evaluating consumer responses to that activity [15]. It is, perhaps, surprising that these laws have not received wider attention and adoption. But, there are several good reasons for this. These include the more time-consuming nature of the work, given the number of data sets typically involved in establishing the laws. Also, securing suitable data sets provides an obstacle, as does the fact that the work is not seen as ‘attractive’ due to its emphasis on replication. Analysis of data becomes scientific when the resulting patterns are shown to extend across many different conditions. This can be achieved only by applying previous knowledge rather than applying significance tests or multivariate techniques to analyse the results of an individual inquiry in isolation, as is the common practice [18]. Additionally, the prevailing myths sound intuitively appealing when presented with the authority of successful practitioners and academics [19]. They also influence the research approaches for monitoring market response, and once research methods are adopted, research practice and theory become mutually supportive and, thus, embedded.

So now we find ourselves with several beliefs and practices in current marketing literature that continue to misguide marketers who do not possess knowledge of the discipline’s scientific laws. Examples include the beliefs that it is desirable to obtain 100% behavioural loyalty from buyers, that segmentation and targeting will increase the effectiveness of communication efforts, and that increased customer loyalty (i.e., getting existing customers to buy more of your brand) is the path to brand growth [16]. Such beliefs limit a brand’s opportunity to achieve growth and create frustration for marketers when the desired market response is not obtained.

Here, we familiarize the reader with the two established marketing laws we draw upon in this paper to examine retail shopper behaviour towards brands that position themselves based on their sustainability credentials.

1.2. The Law of Brand User Profiles

This generalisation was first documented over two decades ago and has been repeatedly empirically verified since then across four geographical regions, 25 years, and more than 50 product categories [20,21,22,23,24]. It explains that buyers of competing brands seldom differ, even if the competing products are seen as different by some consumers. In other words, each brand’s customer base looks the same in terms of demographics, attitudes, and media habits. This is not surprising when you consider the repertoire nature of brand-buying: if people are buying from multiple brands in a category, then it logically follows that the buyer profiles of the brands in that category will be much the same. The positive implications of this law are that each brand has a broad target rather than being restricted to one type of buyer or segment in a market: the only requirement for a person to be a potential buyer of a brand is that the person be a category buyer.

1.3. Law of Double Jeopardy

The Double Jeopardy Law states that brands within the same category vary little in terms of how frequently they are purchased. The big variation between brands in a category comes from the number of buyers they have or their penetration. This means that the big difference between big and small market share brands is not the size of their customer base but the rate at which they are bought. Smaller brands suffer in two ways: they have fewer customers, and those customers buy the brand (slightly) less often. This is the Law of Double Jeopardy. This pattern occurs because if a brand is small, it is highly likely that heavy category buyers will buy it occasionally but split their loyalty by purchasing other, large share brands in that category at a higher rate. Light buyers will not know about the small brand at all and will not buy it. This means loyalty is predictable based on a brand’s market share size [25,26,27,28]. The myth of certain brands, especially those with a smaller market share, having highly loyal consumers is one of marketing’s most common misperceptions. Brands with very high buyer loyalty for their market penetration simply do not exist [16]. The ‘loyalty’ a brand will achieve can, therefore, be predicted from the size of its customer base, and this is what is described by the Double Jeopardy Law. It is the norm for buyers in a category to buy across a range of brands rather than being sole-brand loyal. Such polygamous loyalty was noted as far back as the 1950s [29]. Everyone has a different repertoire of brands they use within a category, and the likelihood of them choosing any one brand within their repertoire is probabilistic and relatively stable over time. This allows the prediction of a range of metrics, including penetration, purchase frequency, and the incidence of 100% loyal buyers. This Law of Double Jeopardy is captured in Goodhardt, Ehrenberg, and Chatfield’s (1984) [30] Dirichlet model that has been shown to predict the main patterns of choice behaviour well in many frequently bought categories such as retail grocery [31]. The Dirichlet describes how consumers buy and how brands compete, including how many category shoppers buy at all, how often they buy, and how they share their purchases across competing brands in the category [32].

1.4. Sustantiable Marketing Retail Shopper Behaviour Myths

This paper examines two common beliefs about shopper behaviour and sustainable brands in the grocery retail setting. It discusses each within the context of the empirically based scientific evidence of buyer behaviour that shows them to be more myth than reality. In doing so, it builds a more solid foundation on which organisations can base their sustainable marketing efforts and a more realistic framework for setting expectations of retail shopper responses to these efforts.

1.5. Belief One: Brands That Claim to Be Sustainable Appeal to a Specific Type of Retail Shopper

It is commonly asserted in the sustainable marketing literature that consumer demand for offerings with improved environmental and social outcomes is the driving force behind manufacturers’ efforts to improve the sustainability of their marketing offerings, organisational strategies, and structures [33,34,35]. The belief that being ‘sustainable’ is a key brand choice criterion for a significant group of consumers has fuelled both an influx of new products and efforts to identify and grow the ‘sustainable’ consumer in the hope that targeting them will result in new and profitable business [36,37,38].

Multiple studies attempt to identify either a socio-demographic [36,39,40,41,42], attitudinal and psychographic [43,44,45,46,47], or behavioural [48,49,50,51] profile of the sustainable consumer, only to yield weak relationships and conflicting results [50,52,53,54]. Yet, the existing research on purchasing brands positioned on their sustainability provides little consensus about the identity and nature of these ‘green’ consumers, except that they are elusive to the marketers who pursue them [38,55,56].

The widely acknowledged lack of consistency between the majority of consumers’ expressed concerns about sustainability issues and their actual willingness to reflect those concerns in their consumption choices is referred to in the literature as the Attitude–Behaviour Gap [37,54,57,58]. There have been multiple attempts to explain this gap, such as an inadequate or incomplete conceptualization of the influences acting on the consumer [42,54], a lack of understanding of the barriers that get in the way of choosing sustainable brands [59], or the context of the purchase and the level of compromise required to choose a sustainable brand [60]. This paper takes an alternative position and empirically examines whether sustainable brands are indeed just ‘normal’ brands with a buyer profile that matches that of other brands in their categories. It draws upon the Law of Brand User Profiles to do so. Given the Law of Brand User Profiles, we would expect to find that a sustainable brand would indeed attract a few customers for whom the sustainability feature was a driving criterion for purchase but that the overall majority of its customers would be ‘normal’ category buyers, who are occasionally partial to sustainable product features. Examination of the customer base of any brand with a functional variant or feature reveals exactly this pattern across contexts that range from credit cards to coffee [23]. This would explain why the sustainable brand shopper segment has proven so elusive: we would see that many buyers who are not primarily driven by sustainability claims still purchase these sustainable brands rather than the brands being bought solely by a small, loyal niche of buyers who focus all their category needs on them.

1.6. Belief Two: Retail Shoppers Are More Loyal to Brands That Claim to Be Sustainable

It is claimed that retail brands that position themselves based on ‘sustainability’ not only appeal to a specific segment of consumers but also attract higher levels of loyalty than their non-sustainable competitors [34,61,62,63,64,65]. The sustainability benefits are assumed to insulate consumers from developing a preference for other competing brands. Furthermore, there are research claims that buyers of sustainable brands are willing to pay a price premium, which is another indicator of higher loyalty [34,35,44,66,67,68]. Sustainable brands are also claimed to be niche [63,65,69], which, by definition, would mean that they attract a higher-than-expected level of loyalty for their brand size [70]. However, little research has been conducted on actual purchases of sustainable brands to determine whether they indeed attract higher behavioural (rather than attitudinal) loyalty and whether they have niche status.

Bearse et al. [68] conducted one of the few studies that assessed the actual purchase of sustainable brands in relation to loyalty, finding that these brands’ buyers are ‘loyal’. However, their assessment of ‘loyalty’ was based on these buyers purchasing more than expected and shopping more often than non-buyers of sustainable brands. Additionally, they only looked at the sustainable brand buyers’ stated likelihood of purchasing the sustainable brand again rather than using actual purchase behaviour. Most other studies that have assessed sustainable brand loyalty have used claimed repeat-purchase behaviour [71] or attitudinal measures [44,72,73] to do so; however, these cannot be reliably linked to actual purchase behaviour due to the Attitude–Behaviour Gap [74,75].

Using the Law of Double Jeopardy, marketers can predict their brand’s level of loyalty rather than supposing that loyalty is a function of their brand being better or different or the result of some special marketing action or brand feature. The fact that such behavioural loyalty is so predictable makes it very easy for marketers to set realistic goals for a brand’s performance and to understand that the focus for a brand should be on building availability, both in the mind of the consumer and on the shelf. A brand with a wide distribution that is familiar to consumers will have higher penetration and associated loyalty than a small, unknown brand, no matter how sustainable. Additionally, a small brand that wishes to grow will have to attract new buyers rather than plan for existing customers to buy it more often.

This paper applies the Law of Double Jeopardy to categories containing sustainable brand offerings to see whether customers do indeed form a grouping of more ‘loyal’ customers who concentrate their purchasing amongst sustainable brands or whether the brands follow the penetration, purchase frequency, and incidence of 100% loyal buyer patterns we would normally expect to see.

2. Materials and Methods

2.1. Data and Overall Approach

Social desirability surrounds sustainable behaviour measures, as everyone wants to be seen to be doing the ‘right thing’. This means that claimed behaviour measures are less reliable than studying actual behaviours [76,77,78] and that we need to examine actual purchase behaviour to gain a more accurate understanding of consumer response—one free from social influences [79]. In this paper, therefore, we move away from self-reporting measures to examine actual purchase behaviour relating to sustainable brands [14].

We draw upon 22 sets of continuous panel data spanning five categories (shampoo, tea bags, instant coffee, fabric softener, and bathroom tissue) in the UK market to empirically examine the two retail sustainability beliefs. The data covers between three and five separate 52-week periods of continuous panel data for each product category. This is the empirical generalisation approach, which uses many sets of data to establish that results hold over many conditions [18,80]. The data are super-panel data that provide details of actual purchases, thus delivering the desired move away from claimed behaviours and attitudes.

Each product category selected contained at least three sustainable brands and three non-sustainable brands, which totalled a minimum of 1% market share, to ensure reliable comparisons between the brand groupings. Up to 30 brands within each category were included in the analysis. These brands comprised the top market share brands in the category, plus a selection of sustainable brands. As sustainable brands were generally smaller-share brands and, thus, were not always within the top 30 market-share brands of the category, at least six of the top sustainable market-share brands from each category were included in the analysis to ensure that enough sustainable brands were available for comparison. The exception to this pattern was the fabric softener category, in which two of the largest market share brands were sustainable brands. If market share affects results, this will be seen in differing findings for fabric softeners.

As consumers might be loyal to the sustainability attribute rather than to individual sustainable brands, which can have poor shelf availability [81,82,83], the brands were grouped by whether they were positioned based on sustainability or not. This left us with two ‘super brands’ to analyse—those positioned based on sustainability and those not. Instead of averaging the results for each brand, super-brands allow for one overall result that can be compared to another result; the super-brand method has, therefore, been utilised by academics such as Goodhardt, Ehrenberg, and Chatfield, who identified that one of the benefits of the Dirichlet model is that super brands can be created without other specifications in the Dirichlet model being affected [30]. Thus, this approach of combining product line variants (SKUs) continues to be used [84,85].

Our analysis treats all the sustainable SKUs as one brand and all the non-sustainable SKUs as another. This removes the impact of individual brands from the analysis. Sometimes, individual brands attract unusual brand-buying behaviour due to variables created by the brand rather than by whether the SKU is sustainable or not; therefore, super-brand analysis can help to determine whether sustainable brand-buying behaviour can be seen without these variables impacting the results [30]. It can also help determine whether consumers are more loyal to sustainable brands as a whole. As sustainable brands are generally small brands, they may not be widely available, and thus, it may be hard for consumers to be loyal to individual brands; the super-brand analysis helps to identify whether consumers are purchasing any sustainable brand and are, thus, loyal to sustainability as an attribute rather than to the brands themselves.

2.2. Identification of Sustainable Brands

An expert panel of 26 marketing scientists was used to identify the sustainable brands for the study. An online survey then asked them to define ‘sustainable’ brands in general. For specific categories, we elicited unprompted explanations of what constitutes a ‘sustainable’ brand and whether they would consider a product ‘sustainable’ within a specified category if it were to assert a claim from a given list. A 70% agreement threshold was used to classify a brand as ‘sustainable’ based on an evaluation of its packaging.

The results were aggregated and weighted depending on the level of expertise of the marketing scientist. Individual SKUs within each potential category were then assigned a score based on how the brand’s marketing communications and packaging matched the sustainable brand descriptions. Categories that did not match the minimum criteria were excluded from the study. The sustainable and non-sustainable SKUs were then aggregated at brand level. If a single brand had some sustainable and some non-sustainable SKUs, these were separated into two versions of the brand to ensure the differences were maintained in analyses.

2.3. Method and Law for Examining Belief One

The belief that brands that claim to be sustainable appeal to a specific type of retail shopper is empirically examined by drawing on the Law of Brand User Profiles. The analysis considers the characteristics of buyers within each socio-demographic group and determines whether buyers with those characteristics are more likely to purchase a brand making sustainability claims or not. We adopt an altered version of the original Hammond, Ehrenberg, and Goodhardt [20] Mean Absolute Deviation (MAD) analysis of brand user profiles [20], which was created by Nelson-Field to ensure that the inclusion of small market share brands, a category which often includes sustainable brands, does not skew the results [86]. Nelson-Field’s (2009) concerns related to the ‘tails’ of the data, in that unusually large or unusually small deviations can greatly affect the MAD for a given brand or category: in the original MAD methodology, such an event could make it appear that a whole category deviated from the benchmark due to one unusually large deviation. The second concern Nelson-Field (2009) raised relates to continual averaging: category and brand MADs are the averages of averages, which can mask results by giving a false impression of relativity among brands. This is of concern because median deviations may vary substantially between brands, and reporting only the averages would ignore this. To solve these issues, the adapted approach reports the total number of deviations within the analysed groups instead of reporting MADs or comparing the maximum and minimum percentages. The analysis is conducted by comparing the percentages of each socio-demographic group that purchased each brand and calculating deviations to the category purchase rate for each group. The number of significant deviations (>6% above or below the category profile) for each group is then calculated to determine the proportion of sustainable brands and non-sustainable brands that are above or below the expected level (i.e., the category purchase rate). Any deviations below this level from the category average were considered not to be of any practical significance—that is, not warranting any additional marketing actions. This is an accepted approach for MAD analysis, where traditional statistical tests are not able to be applied [23,87].

The proportions of positive and negative deviations are then compared across the two super brands. If a higher proportion of positive deviations and a lower proportion of negative deviations is more evident for sustainable brands than for non-sustainable brands, it can be concluded that, for the relevant socio-demographic group, the deviations reflect that the user base of sustainable brands is more than expected, and hence, skewed on that demographic.

The super-brand approach has the added advantage that it can determine whether a specific socio-demographic group is more likely to purchase sustainable brands overall or not, as if a socio-demographic group purchases more sustainable brands overall but not any one individual brand, an individual brand analysis would not pick this up.

2.4. Method and Law for Examining Belief Two

To examine whether retail shoppers are more loyal to brands that claim to be sustainable, we draw upon the Law of Double Jeopardy to first examine the relationship between the size of the brand’s customer base (penetration) and its purchase frequency. Additionally, we examine the incidence of 100% loyal customers, calculated by dividing the number of category buyers who purchased only the given brand and no other brand within the given time frame by the number of category buyers who simply purchased the brand.

In most of the categories examined, there was a large difference in the market share size of the sustainable and non-sustainable brands, with sustainable brands generally having a much smaller market share. Due to Double Jeopardy, these brands would naturally attract less loyalty than the bigger non-sustainable brands [16]. The market share size of the brands, therefore, needed to be controlled for in this analysis. The Dirichlet model was used to analyse the data because it removes the impact of market share and is designed to identify behavioural patterns within markets [88]. The Dirichlet model calculates theoretical values for each brand based on the observed penetration and average purchase frequencies of both brands and category. These theoretical values can be compared to the actual values for each brand to determine whether the metric levels are ‘as expected’ for the given market share of each brand. In calculating the theoretical values, the Dirichlet model removes the effect of Double Jeopardy, hence controlling for size effects.

The loyalty metrics calculated by the Dirichlet model used in this study were penetration, average purchase frequency, and 100% loyalty. If the deviation between the observed and theoretical values for both penetration and average purchase frequency of a brand is above 0.5% or below −0.5%, it is considered significant. The deviation for 100% loyalty is considered significant if it is above 2% or below −2%. In Dirichlet analysis, it is, however, the general patterns across a number of categories and brands that we look at to establish main effects rather than individual significance so the persistence of deviations may be revealed across many datasets, resulting in a generalised finding [89].

3. Results

3.1. Findings Belief One: Brands That Claim to Be Sustainable Appeal to a Specific Type of Retail Shopper

To illustrate how significant patterns are identified within each dataset, the social class results from the instant coffee category are shown below.

Table 1. Illustration of deviations, one year of instant coffee buyer profiles (source: Kantar Superpanel).

Brands	Deviations	% Significant Deviations
		Class AB	Class C1	Class C2	Class D	Class E
Sustainable	Positive	6	0	6	6	6
	Negative	19	25	0	0	0
Non-sustainable	Positive	69	23	0	8	8
	Negative	8	0	69	31	38

illustrates the pattern observed in the results from one year out of the five examined for this category.

Table 1 shows that the social class of instant coffee buyers appears to have a managerially significant impact on sustainable brand purchases, with social class AB being much more likely to purchase sustainable instant coffee brands (69% positive deviations from the category average, 8% negative) and social class C1 being slightly more likely to purchase sustainable instant coffee brands (23% positive deviations, no negative). Conversely, social class C2 was much less likely to purchase sustainable instant coffee brands (no positive, 69% negative), and social classes D and E were slightly less likely to purchase sustainable brands (8% positive, 31% negative and 8% positive, 38% negative). Higher social classes, thus, appeared to be more likely to purchase sustainable brands in the instant coffee category in this year's period.

For the full analysis, five socio-demographic variables were examined: age (under 18, 18–24, 25–34, 35–44, 45–54, 55–64, 65+); gender (male, female); social class (AB, C1, C2, D, E); number of household members (1, 2, 3, 4, 5+); number of children in the house (0, 1, 2, 3+); life-cycle stage (pre-family, young family, middle family, family 10+ years, older dependent kids, empty nesters, retired); and location (London, Midlands, North East, Yorkshire, Lancashire, South, Scotland, East England, Wales, and West, South West).

Table 2. Overall socio-demographic sustainable brand user profiles (source: Kantar Superpanel).

Brands	Instant Coffee	Tea	Shampoo	Fabric Softener	Bathroom Tissue
Gender	- *	-	√	-	-
Age	√	√	√	√	√
Household size	√	√	√	√	√
Social class	Higher more likely to buy sustainable	Higher more likely to buy sustainable	√	√	√
Life-cycle stage	√	√	√	√	√
Location	√	√	√	√	√

* A ‘-’ indicates that the listed socio-demographic group was not analysed in the specified category, while a ‘√’ indicates that no significant deviations were identified. More detailed results for each category are shown in Appendix A.

summarizes the data for these five categories across multiple years of data and the six socio-demographic variables. It shows that the deviations presented in Table 1 are indeed relatively rare and that for the 26 category variables examined, only two show significant, consistent deviations for sustainable brands. These are both for social class, with sustainable brands showing a higher social class profile for their buyer base than non-sustainable brands.

Overall, sustainable brand buyers generally do not have a specific socio-demographic profile that differs from that of all category buyers except in terms of being slightly more likely to be of a higher social class. Future research needs to be conducted to identify the key reasons for this.

These results help to explain why the socio-demographic profiles of sustainability consumers have been elusive in the past, as sustainable brands generally appeal to all consumers and only slightly more to consumers from higher social classes in two of the five categories examined. The findings strengthen the empirical generalisation that consumer socio-demographic profiles have, for the most part, little impact on brand choice and that each brand’s buyers look pretty much just like buyers of the category.

3.2. Findings Belief Two: Retail Shoppers Are More Loyal to Brands That Claim to Be Sustainable

Table 3. Loyalty deviations for sustainable and non-sustainable brands (source: Kantar Superpanel).

Category	Year	Non-Sustainable Brands						Sustainable Brands
		Pene			APF			Pene			APF
		O	T	Dev	O	T	Dev	O	T	Dev	O	T	Dev
Instant coffee	1	5.3	5.2	0.1	2.9	3.2	−0.3	0.5	0.4	0.1	2	3	−1
	2	5.3	5.2	0.1	3.0	3.2	−0.3	0.5	0.4	0.1	2.2	3.0	−0.8
	3	5.3	5.3	0.0	3.0	3.1	−0.1	0.4	0.4	0.1	2.1	2.9	−0.8
	4	5.2	5.2	0.0	2.9	3.1	−0.2	0.3	0.2	0.1	2.1	2.9	−0.8
	5	6.1	6.0	0.1	2.9	3.2	−0.2	0.4	0.2	0.1	1.9	2.9	−1.0
Tea bags	1	5.1	5.1	−0.2	3.2	3.0	0.3	0.6	0.5	0.1	2.3	2.8	−0.5
	2	5.1	5.2	−0.1	3.1	2.9	0.2	0.5	0.4	0.1	2.2	2.7	−0.5
	3	4.9	5.0	−0.1	2.9	2.8	0.1	0.5	0.4	0.1	2.1	2.6	−0.5
	4	5.4	5.5	−0.1	2.9	2.1	0.2	0.5	0.4	0.1	2.2	2.6	−0.4
	5	5.4	4.5	0.9	2.9	3.1	−0.2	0.5	1.7	−1.2	2.2	3.0	−0.8
Fabric softener	1	4.9	4.9	0.0	2.8	2.9	−0.1	3.6	3.6	0.0	2.8	2.9	−0.1
	2	4.7	4.7	0.0	2.7	2.8	−0.1	3.5	3.6	−0.1	2.2	2.7	−0.5
	3	4.4	4.3	−0.1	2.5	2.4	0.1	4.3	4.3	0.0	2.1	2.5	−0.3
	4	4.2	4.3	−0.1	2.3	2.3	0.0	4.3	4.3	0.0	2.1	2.4	−0.2
	5	4.5	4.5	0.0	2.3	2.4	−0.1	5.3	5.4	−0.1	2.3	2.4	−0.1
Bathroom tissue	1	7.9	7.8	0.1	3.5	3.1	0.4	2.6	2.0	0.6	2.9	2.8	0.1
	2	9.0	9.5	−0.5	3.2	3.1	0.1	2.9	2.3	0.6	2.3	2.8	−0.5
	3	11.6	12.3	−0.7	3.4	3.2	0.2	2.6	1.9	0.8	1.8	2.9	−1.1
Shampoo	1	2.6	2.6	0	1.7	1.7	0	0.5	0.5	0	1.4	1.6	−0.3
	2	2.6	2.6	0	1.7	1.7	0	0.4	0.4	0	1.3	1.6	−0.3
	3	2.5	2.5	0	1.7	1.7	0	0.3	0.3	0	1.2	1.6	−0.4
	4	2.8	2.8	0	1.6	1.7	−0.1	0.5	0.5	0	1.4	1.6	−0.2

shows the penetration and average purchase frequency patterns seen for sustainable and non-sustainable brands. A brand is considered to attract higher-than-expected loyalty if its penetration is below expected and its average purchase frequency and 100% loyalty are above expected [88,90]. It is considered a niche brand if its penetration is significantly below expectations and its average purchase frequency is significantly above expectations.

As expected, the non-sustainable super brand has a higher penetration than the sustainable super brand. Across the years and within each category, we can see quite high stability in terms of the sustainable and non-sustainable brands’ penetration in the category. For purchase frequency, we see the expected Double Jeopardy pattern of the high-penetration non-sustainable super brand having slightly greater purchase frequency than the sustainable super brand but far less variation compared to differences in penetration.

When we compare the observed and predicted penetration and the average purchase frequency for both sustainable and non-sustainable brands, we see very few differences. In other words, both super brands show the buying patterns we would expect to see given their market shares. This provides empirical support for the Double Jeopardy Law holding for sustainable brands.

Table 4. Sole loyalty metric deviations (source: Kantar Superpanel).

Category	Year	Non-Sustainable Brands			Sustainable Brands
Instant coffee	1	13	15	−2	9	13	−4
	2	13	14	−1	9	12	−3
	3	13	14	0	9	12	−3
	4	14	15	0	10	13	−3
	5	15	14	1	12	13	−1
Tea bags	1	15	15	0	9	13	−5
	2	13	14	−1	8	13	−5
	3	14	15	−1	7	14	−6
	4	16	14	2	9	13	−4
	5	16	14	2	9	13	−4
Fabric softener	1	13	14	−1	8	13	−5
	2	12	13	−1	8	13	−5
	3	12	13	−1	23	12	11
	4	12	14	−2	15	14	1
	5	14	15	−1	16	15	1
Bathroom tissue	1	10	9	1	4	8	−1
	2	10	10	0	8	8	0
	3	9	8	1	3	7	4
Shampoo	1	20	20	0	15	19	−4
	2	19	20	−1	18	18	0
	3	20	21	−1	16	19	−3
	4	22	22	0	17	20	−3
Average		14	15	1	11	13	−2

examines the incidence of 100% loyal buyers, again using the super-brand approach. If sustainable brands were attracting higher levels of loyalty than ‘normal’, this would show positive deviations between observed and expected results at a higher incidence than is seen for non-sustainable brands.

In general, Table 4 shows that sole loyalty is around the same level for both the sustainable and the non-sustainable brands. It is also low, at an average of just 14% annually for non-sustainable brands and 11% for sustainable. In other words, 89% of buyers of sustainable brands were also buying other brands in the same time period. The average expected and observed levels of solely loyal buyers for both sustainable and non-sustainable brands are close to what was expected given their market share, with just a 1% excess for non-sustainable brands and a −2% deficit for sustainable brands observed on average across the 22 sets of data. This approach uses many sets of data to demonstrate Significant Sameness [18,91]. The results refute the findings from Peattie and Crane (2005) [61], Bonini and Oppenheim (2008) [62], and Beattie et al. (2009) [92], which claim that the attribute of being sustainable attracts high loyalty. It provides further empirical support for the Double Jeopardy Law holding in the context of sustainable brands.

That we see pretty much as-expected levels of loyalty for the sustainable super brand indicates that buyers do not purchase only sustainable brands within a category. It suggests that the buyers of sustainable brands simply purchase them as they would any other brand—in line with their market share. Sustainable brands are, therefore, a part of most consumers’ repertoires instead of the whole repertoire for a small group of consumers.

4. Discussion

Many organisations are leading the way in producing retail brands that provide positive social and environmental impacts relative to those of their competitors. To communicate these efforts, organisations need to understand how shoppers make purchasing choices and how these are affected by sustainability initiatives. This paper identifies two key beliefs that dominate retail sustainable marketing literature and industry practice. These beliefs are empirically shown to be myths that limit the marketing community’s ability to create mainstream solutions that will drive preference for sustainable brands. Drawing upon established marketing science findings, this paper presents a more realistic view of the shopper landscape in which sustainable brands must compete, thereby helping marketers and managers understand the boundaries of what is achievable so that they can set appropriate targets for sustainability initiatives.

The overall results support the findings of Tanner and Wölfing [49] and Kinnear, Taylor, and Ahmed [45], who concluded that socio-demographic profiles are a poor predictor of sustainable brand purchases. The results also partially explain why the sustainability consumer has been so empirically elusive, as such brands generally appeal to all consumers. The findings also support the body of prior work showing that competing brands generally appeal to same-category buyers, whether they are positioned differently or not [20,23,93]. These results strengthen the empirical generalisation that consumer socio-demographic profiles have minimal impact on brand choice. The implication is that sustainable brands appeal to most of the market. While many consumers are interested in sustainable attributes, this concern is relative to others they hold for their purchases and is often smaller. Because consumers have more important needs than just sustainability, their ability to display a preference for sustainable brands is limited by trade-offs made when considering their purchases. Buyers primarily seek functional attributes when making brand choices.

Organisations with sustainable consumer goods brands should, therefore, target all category buyers rather than feel that they are restricted to appealing to just a specific socio-demographic profile. These results present a positive message for sustainable marketing efforts. Potential buyers are the entire market, so efforts do not need to be restricted to just a few consumers who place sustainability above all else in their brand choices.

The finding that sustainable brands are ‘normal’ and fit the Dirichlet model (see Ehrenberg et al., 2004 [88]) for expected penetration, purchase, and loyalty metrics is also a positive finding. The implication is that sustainable brands should grow in the same way as other brands: through improved mental and physical availability [16]. This is the opposite approach to the loyalty strategy that many sustainable brand managers have taken in the past [94,95]. These results are, therefore, relevant and useful to sustainable brand managers who want to understand how to successfully grow their brands.

5. Conclusions, Limitations, and Future Research

Many organisations are leading the way in producing brands that provide positive social and environmental impacts relative to their competitors. To communicate these efforts, organisations need to understand correctly how consumers make purchasing choices and how these are affected by sustainability initiatives. This paper identifies two key myths that dominate sustainable marketing literature and industry practice. Such beliefs limit the ability of the marketing community to create mainstream solutions that will drive preference for sustainable brands. Drawing upon established marketing science findings, this paper presents a more realistic view of the landscape in which sustainable brands must compete, thereby helping marketers and retailers understand the boundaries of what is achievable so that appropriate targets can be set for sustainability initiatives. In this way, a preference for more sustainable brands can be developed, further accelerating industry commitments to improving society and the environment. It is a step towards overcoming the Attitude–Behaviour Gap quandary that has long affected sustainable marketing. Such a scientifically grounded approach is needed to identify the long-term issues in sustainable marketing that deserve focus, to help set realistic goals, and to determine the range of strategies that can be pursued given what is known about consumer behaviour.

Sustainable marketing seeks to alter an entire exchange system, and this is no small task. It requires significant innovation and, ultimately, financial capital. The current challenge for sustainability is to move beyond the niche to the mainstream, to scale and be sustained across companies, industries, and whole markets. This paper establishes a position from which to continue making this change.

The research findings suggest that sustainable brands are bought in much the same way other consumer brands are. The brand user profile findings suggest that there is no specific segment within the markets examined buying or not buying sustainable brands. Therefore, sustainable brand marketing efforts must be targeted at all category buyers, with no requirement to restrict or tailor efforts to specific consumer segments. Our results on consumer loyalty conform to the Double Jeopardy pattern, where the biggest difference between high market share and lower market share brands is the number of people who buy them, not how often they are bought. The implications for manufacturers are that sustainable brands, which are typically characterised by low market share, should primarily focus on increasing the number of people who buy them as opposed to attempting to increase the frequency of purchases amongst their existing customer base.

Overall, this research has successfully extended the consumer buying behaviour findings to the new context of sustainable brands. For sustainable brands to grow, their penetration levels must be increased. This is the primary challenge and will, in turn, lead to slightly increased consumer loyalty for sustainable brands as this is achieved. This is the key strategy to promoting and mainstreaming sustainable brands and addressing their current low level of purchase.

In addition to providing a new empirical framework to examine the marketing of sustainable brands, this research also contributes to the academic knowledge of the boundary conditions for the Dirichlet model and the ubiquitousness of buyer behaviour it describes [88]. However, there is still a need for the continued growth of additional, robust, empirically based knowledge in this field. This paper was constrained to just five grocery categories, two myths, and one country in its empirical examination. With the growing global presence of sustainable brands, more research is needed to extend existing knowledge across countries, categories, and time. Additionally, the noted deviation of sustainable brands along the social class variable is worthy of further investigation. The authors encourage other researchers to further identify and challenge existing myths and practices through such an evidence-based approach to knowledge development. Future research can also complement this research by determining the effect of physical availability on the buying behaviour of sustainable brands, as both physical and mental availability have been found to shape the observed patterns of brand buying behaviour.

Future research should also investigate the marketing evaluation metrics that will be most useful in measuring the effect of sustainable marketing initiatives on a business. Such metrics are required if managers are to have realistic expectations about what their sustainability efforts and investments can deliver and if the efforts can be made transparent to stakeholders at board-level reporting.