Analyzing proprietary,private label,and non-brands in fresh produce purchases

Abstract

When purchasing packaged products within a supermarket, consumers choose between proprietary or private label brands. However, when purchasing fresh fruits and vegetables, non-branded produce is the dominant option—with proprietary and private label brands only recently becoming available. Previous fast-moving consumer goods (FMCG) research finds that proprietary and private label brands affect consumer loyalty—however, no research exists for fresh categories. This research is the first to determine the effect of emerging brands in fresh categories on consumer buying behavior. Our research examines consumers’ loyalty toward proprietary, private label, or non-branded fresh fruits and vegetables and the level of customer sharing between these options, using analytical approaches applicable to FMCG categories. The panel data contains nearly 46,000 households making over 8 million purchases in the United States during 2015. Results show that proprietary, private label, and now non-branded fresh produce have expected loyalty levels, for their size, and consumers share their purchases across the three options (i.e., consumers are not loyal to just one option). The study analyzes and interprets purchase data in fresh categories offering marketing academics and practitioners actionable advice for working with fresh produce purchase data.

Keywords

consumer behavior double jeopardy duplication of purchase fresh fruit and vegetable categories non-branded

Introduction

For several decades, marketing researchers became accustomed to working with consumer purchase data in fast moving consumer goods (FMCGs), developing approaches to guide their analyses (Graham et al., 2017). When documenting brand loyalty relative to size, use double jeopardy (DJ) (Ehrenberg et al., 1990). When outlining brands sharing customers, use duplication of purchase (DoP) (Goodhardt, 1966). These useful patterns measure loyalty toward, and competition between FMCG brands, product variants (Singh et al., 2008), and prices (Dawes et al., 2017). The recent expansion of these analytical tools includes the context of private label brands (Dawes & Nenycz-Thiel, 2013), an unusual condition given the lower branding investments. However, other products are receiving even less marketing and branding investment—commodities. To the best of our knowledge, few studies cover these products, giving academics and practitioner’s limited guidance on what to expect in data analysis.

The lack of non-branded products examination is surprising, given their traditional domination in fresh fruit and vegetable categories—a market in the United States worth US$992 billion, accounting for 5.5% of Gross Domestic Product (GDP), and employing 21.6 million people (USDA, 2018). A recent trend is emerging with the development of proprietary and private label brands. Fresh proprietary brands are focusing on the grower/manufacturer (e.g., Sunny Farms, Zespri, or Dole) or production region (e.g., New Zealand gold or Tasmanian apples). For consumers, proprietary fresh brands, like FMCG brands, are decision-making short-cuts, allowing consumers to identify products that satisfy their needs. Similarly, fresh proprietary brand advertisements could encourage first-time trial (Bogomolova et al., 2019), with recognition influencing consumer repeat purchase rates, and increasing loyalty to a proprietary brand, over non-branded produce.

Fresh category private label brands are another recent trend. Retailers who are already providing good-quality fresh produce at a reasonable price (Hutcheson & Moutinho, 1998) are now offering private label alternatives. These private label brands serve the same function as the proprietary brands, allowing consumers to identify and form loyalty quickly. Despite the unique nature of the fresh fruit and vegetable category, its significant economic contribution to all members of the supply chain, and the novel developments in its branding structure, no contextual research exists. Our research fills this gap by guiding marketing academics and practitioners toward analytical approaches and models for measuring buying behavior of proprietary, private label, and non-branded products in fresh categories.

The article begins by exploring proprietary, private label, and non-branded produce before outlining patterns of consumer buying behavior. Specifically, the relationship between how many people and how often they choose a brand. It continues by examining how brands share consumers within a category. This study contributes original empirical research into how consumers purchase non-branded products within the context of fresh categories using observational data. The article outlines several key concepts to marketing research practitioners including how to approach consumer purchase data in fresh categories, which analytical approaches to use, and what benchmarks they should expect.

Background

Development of brands in fresh categories

Decades of research demonstrates that within FMCG categories, brands affect consumer choices. Brands distinguish products (Laaksonen, 1994), signal consistent quality (Keller & Lehmann, 2006; Zeithaml, 1988), are easier to find (Shocker et al., 1991), and therefore encourage repeat purchases (Kotler, 1967). Brands are built by understanding and serving customers’ wants and needs. The concept of customer-based brand equity (CBBE) holds all of the knowledge that consumers have about a brand (Keller, 2001). CBBE builds on five components: perceived quality, brand awareness, brand associations, other proprietary brand assets, and brand loyalty (Aaker, 1991; Keller, 1993; Zeithaml, 1988). This study focuses on brand loyalty (Aaker, 1991), specifically, modeling the behavioral loyalty toward individual brands (DJ) or groups of brands (DoP) (Bryła, 2018; Ehrenberg & Goodhardt, 2002; Sharp et al., 2012; Winchester et al., 2015).

Brands aid consumer’s decision-making. Proprietary brands have existed in grocery stores for decades, and are often available throughout different retailers and store formats. For example, within the United States, Pepsi Cola is available in various retailers (e.g., Walmart and Safeway) and different store formats (e.g., hypermarkets, supermarkets, convenience stores) throughout the country (Kumar & Steenkamp, 2007). These proprietary brands enhance repeat purchase rates and brand loyalty within FMCGs (Ehrenberg et al., 2004; Goodhardt et al., 1984). Now, fresh category growers are starting to develop and promote proprietary (e.g., Chiquita bananas) or private label (e.g., Walmart stickered bananas) branded produce.

Private label brands first appeared in the United States in the early 1970s, initially accounting for 1% market share (Kumar & Steenkamp, 2007). Often, the price was lower, the packaging was plain, and they received less attention or advertising support (Kumar & Steenkamp, 2007; McEnally & Hawes, 1984). In the following decade, private label’s market share grew to 10% (Hoch & Banerji, 1993). After which, retailers, realizing that private labels with improved quality, increased promotion, and long-term strategies, enabled higher profit margins, and thus, increased production, which resulted in ~18% of category sales (Nielsen, 2018). A review of the literature finds that private labels benefit retailers in seven key ways—increasing category profits, heightening gross margins, increasing bargaining power with manufactures, decreasing the risk of carrying proprietary brands on promotion, differentiating competing chain stores, boosting store loyalty, and attracting price-sensitive and deal-prone consumers (Hyman et al., 2010).

Overall, proprietary and private label brands simplify consumers’ choice. Private labels share memory cues (Nenycz-Thiel et al., 2009), which are offset by how consumers see each store (e.g., premium retailers as having premium private labels and discount retailers as having discount private labels; Nenycz-Thiel & Romaniuk, 2012). Studies comparing proprietary brands with private labels examined the competition between them, finding that they retrieve the same attributes (e.g., costs a little less, or gives great clean for less) in the evoked set and brands compete with those associated with the same memory cue (Nenycz-Thiel et al., 2009). Prior knowledge of nonusers is higher for proprietary brands than private labels (Nenycz-Thiel & Romaniuk, 2014).

DJ—a behavioral loyalty pattern

DJ, the commonly repeating pattern closely described by simple mathematical models with underlying theoretical assumptions, was first described by Sociologist William McPhee (1963). For almost 60 years, the pattern consistently shows that small brands have fewer buyers, who purchase slightly less frequently, and bigger brands have more buyers who buy slightly more frequently (Ehrenberg & Goodhardt, 2002; Sharp et al., 2012). The difference between big brands and small brands is the percentage of category buyers purchasing in the period, and not their loyalty (often measured as purchase frequency, or share of category) (Graham et al., 2017; Greenacre et al., 2015; Uncles & Kwok, 2009). The pattern occurs in FMCGs including instant coffee (Greenacre et al., 2015), toothpaste, carbonated soft drinks, yogurt, instant noodle, soy sauce (Uncles & Kwok, 2009), and many other categories such as washing detergent (Scriven et al., 2017). Earlier research analyzing fresh category loyalty was not at a brand level, and the behavioral data were not observational (i.e., claimed) (Anesbury, Greenacre, et al., 2018).

DJ is a generalizable outcome of the NBD-Dirichlet model (hereafter Dirichlet) (Goodhardt et al., 1984; Sharp & Ehrenberg, 2000). Dirichlet provides theoretical DJ values of brand penetration and average purchase frequencies (Kearns, 2010), allowing for deviation identification (i.e., niche and Change-of-Pace [CoP]). When the difference between the observed behavior and theoretical behavior is greater than 10%, there is a deviation (e.g., if fewer consumers are buying a brand more often than theoretically expected, a niche classification). To be considered niche, the observed penetration is less than 10%, and the average purchase frequency is greater than 10% of that expected (Anesbury, Nguyen, & Bogomolova, 2018; Kahn et al., 1988; Scriven et al., 2017). Niche brands have a low market share because of their positioning to a small loyal customer base. Often small companies or a small division of large enterprises follow this strategy to avoid confrontation with the market leader (Kahn et al., 1988).

Alternatively, if consumers are buying a brand less frequently than expected, the brand is classified as a CoP. Therefore, the observed penetration is greater than 10%, and the average purchase frequency is less than 10% of that expected (Anesbury, Nguyen, & Bogomolova, 2018; Kahn et al., 1988; Scriven et al., 2017). There are three common causes of CoP deviations—seasonality, buying for others (Scriven et al., 2017), and variety seeking (Kahn et al., 1988). Seasonality restricts the period the brand is available, and therefore, an inability to repeat purchase. Seasonality is a larger factor in the fresh categories compared with highly processed FMCGs with long shelf-lives, as the availability of fresh produce is dependent on growing seasons and harvest times. Seasonality influences price, origin, availability, and quality of fresh fruit and vegetables (Rekhy & McConchie, 2014). Therefore, seasonality plays a role in the consumer choices and the resulting fit of the DJ pattern. Buying for others, say as a gift, means people buy brands, with no intention of buying again.

Similarly, restricted distributions or functional differences cause niche brands. Private labels are only available in specific chains (Scriven et al., 2017), restricting their penetration to shoppers of these chains. Another example is Hispanic television networks being watched by Hispanic speakers, who are limited to those few channels (Goodhardt & Ehrenberg, 1969). When observed brand performance differs from the Dirichlet theoretical, marketers often notice something interesting about consumers loyalty toward those brands (Ehrenberg et al., 2004).

Although traditional reasons for DJ deviations (e.g., seasonality) equally affect proprietary, private label, and non-branded products, this situation has additional reasons. First, proprietary brands receive far more marketing support, both instore and outside. The consistency, higher quality, larger advertising, and media budgets all increase consumer’s propensity to purchase these brands. Therefore, proprietary brands are more likely to exhibit niche-like behavior from consumers.

Second, the restricted distribution and lack of advertising (limiting the potential customer base), and the historical position of private label brands (i.e., lower quality, restricting trial, limiting potential customer base) (Romaniuk et al., 2014), are all potential causes for a small but loyal customer base. The lack of advertising reduces knowledge among current non-brand buyers (Nenycz-Thiel & Romaniuk, 2014), decreasing the probability of growing the brand’s penetration and maintaining this small but loyal customer base. Empirical evidence shows that private labels, compared with proprietary brands, grow more in purchase frequency than share of category requirements (Romaniuk et al., 2014), which is the most prominent DJ loyalty measure. Furthermore, if consumers see private label brands as entailing lower price and lower quality, and they maintain restricted advertising and distribution, then they may have small but loyal customer bases. However, consumers shop at multiple supermarkets (Wrigley & Dunn, 1984), and therefore when analyzing within supermarket buying behavior, the occurrence of consumer buying at multiple supermarkets results in private labels being much more likely to be CoP.

Finally, non-branded products may be CoP, as previous research shows that their greatest impact occurs for penetration (not loyalty) if consumers see them as commodities, where quality is of less importance (Harris & Strang, 1985). Therefore, the first research question is as follows:

RQ1. Are proprietary, private label, and non-branded fresh produce bought as expected, given the DJ pattern?

However, the 10% deviation proposed by Kahn et al. (1988) and widely adopted (Anesbury, Nguyen, & Bogomolova, 2018; Scriven et al., 2017) is arbitrary (why not 5%, or 20%, or 50%?), so we use a secondary approach. Driesener et al. (2017) propose using eight goodness-of-fit tests to determine if Dirichlet accurately models the brand buying behavior. The tests compare the observed and Dirichlet theoretical values for both penetration and average purchase frequency in four ways. Using the estimated and the observed metrics (a) the comparisons of the averages, (b) correlations, (c) mean absolute deviations (MADs), and (d) mean absolute percentage errors (MAPEs). Therefore, the second research question is as follows:

RQ2. Do categories containing proprietary, private label, and non-branded fresh produce still operate within the Dirichlet goodness-of-fit tests?

The DJ pattern determines the expected loyalty toward individual brands within the context of category-wide buying behavior. However, consumers purchase more than one brand within a category (Banelis et al., 2013). Therefore, the DoP method (Ehrenberg & Goodhardt, 1969) and its application (Anesbury, Greenacre, et al., 2018) examine multi-brand loyalty.

DoP pattern of cross-brand purchases

DoP is an empirical generalization stating that all brands share their customers with other brands within the category, in line with their size. Brands share more of their customers with larger brands and fewer customers with smaller brands (Sharp, 2010). DoP originated from research analyzing television channel viewer’s loyalty, finding other factors (i.e., program content) had little influence (Ehrenberg & Goodhardt, 1969; Goodhardt, 1966). DoP also applies to radio listening (Lees & Wright, 2013) and FMCGs, including cereals (Anesbury, Nguyen, & Bogomolova, 2018), organic foods (Bryła, 2018), and fair trade coffee and tea (Winchester et al., 2015).

This study determines how proprietary brands, private-label brands, and non-branded produce share customers. DoP can determine if brands share customers unexpectedly. Deviations occur if brands share more or fewer customers than expected. Proprietary brands should share a higher proportion of their customers with larger private label or non-branded competitors and fewer with small private label brands or non-branded produce. However, private labels share consumers excessively (but not exclusively) with each other, and less with proprietary brands (Prendergast & Marr, 1997; Stern, 1994), but examining the pattern for non-branded produce is novel. In the pharmaceutical industry, while there is minimal oversharing within proprietary brands and non-branded (generic) drugs, sharing occurs between them (Stern, 1994). Furthermore, Dawes & Nenycz-Thiel (2013) found that private labels compete against proprietary brands within the same store and private label brands in different stores. Higher competition among private labels of different retailers occurs when the categories have high purchase frequencies (which is the case for fresh categories). Prior research also tells us that private label penetrations are higher in lower consequence categories (i.e., minimal quality difference among brands), and is lower where previous category experience is necessary (Baltas, 2003; Batra & Sinha, 2000).

Private label shopper profiles may be diverse. While earlier studies find that private label buyers have lower incomes, are price conscious, and are older (Ailawadi et al., 2001; Prendergast & Marr, 1997), more recent research found that private label buyers have higher social status (Baltas, 2003), and they appeal to a diverse array of consumer segments (Cuneo et al., 2019). The conflicting evidence regarding consumer demographics and their tendency to prefer private labels brands leads to our third research question:

RQ3. Do proprietary, private label, and non-branded fresh produce share customers as expected by DoP law?

Data and method

Data description

To analyze the loyalty toward proprietary, private label, and non-branded products, and the competition between them, we examine Nielsen panel data from the United States in 2015 (Kilts Centre for Marketing, 2019). The data set contains 45,958 households making up 8,123,409 purchases from 10 different fresh categories (apples, oranges, carrots, lettuce, mushrooms, spinach, tomatoes, onions, potatoes, and strawberries) from the top three grocery retailers. As with the previous research examining the loyalty toward or perceptions of private label brands (Prendergast & Marr, 1997; Nenycz-Thiel & Romaniuk, 2014, 2016), this study analyzes the data of each of the three retailers separately. Including multiple retailers is an advantage over past research (Hansen et al., 2006) as it gives insights into buyer behavior patterns for private labels of different retailer’s. Furthermore, examining 10 categories for three retailers, using disaggregate and aggregate data, provides us with a multiple sets of data approach (Bound & Ehrenberg, 1989; Ehrenberg, 1966, 1990). The built-in replication determines if an observation is once-off or reoccurring in different contexts. In this study, the analysis includes multiple categories and retailers. It is also essential to analyze large food retailers to ensure that they possess private label brands.

This study analyzes all three retailers individually and all 10 categories at two different levels, disaggregated and aggregated. The study examines disaggregated proprietary brands within each category (Dawes & Nenycz-Thiel, 2013) and examines an aggregated proprietary brand. Due to the data structure, the creation of aggregated proprietary superbrands was necessary to directly compare buying behavior with the private label and non-branded products. By default, Nielsen aggregate private label and non-branded data, and therefore, the aggregation of proprietary brands was necessary to make an accurate comparison. The proprietary superbrand is an aggregation of all proprietary brands within one retailer, under one umbrella. For example, for Retailer 1, an aggregation of the sales of eight apple brands, including Trout, North Bay, and Chelan Fresh was made. The aggregation is a very important feature of the Dirichlet (Goodhardt et al., 1984) that allows the analysis of behavior and ensures correct comparison of results. As demonstrated in this study, the feature goes beyond the aggregation of small brands as “other” (Driesener et al., 2017; Ehrenberg et al., 2004; Graham et al., 2017; Scriven et al., 2017).

First, the three disaggregated proprietary brands and the aggregated private label and non-branded produce were analyzed. Second, the aggregated “superbrands” were analyzed. That is, within each category, all proprietary brands, private label brands, and non-branded fresh produces (with penetration and market share higher than 1%) were aggregated. Previous studies have analyzed the aggregation of smaller brands as a “superbrand,” displaying no negative affect on the model specifications (Goodhardt et al., 1984; Habel & Lockshin, 2013). The two levels of analysis will give insights into the general purchase behavior of proprietary, private label, and non-branded products.

The analysis being with DJ—determining if consumers are purchasing proprietary, private label, and non-branded fresh fruits and vegetables as expected. Adopting the 10% deviation of observed and theoretical Dirichlet values (Kahn et al., 1988) identifies brands buying more, or less frequently than expected for their size. Next, the analysis of variance (ANOVA) calculations of the disaggregated data, using the mean penetration and average purchase frequencies for three retailers, show any statistically significant differences between the proprietary, private label, and non-branded products. ANOVA calculations of the Dirichlet model of proprietary superbrands tests whether the mean values of several independent groups, which are defined by a categorical independent variable (the three retailers) differ.

Given the 10% deviation benchmark limitation (Kahn et al., 1988), the study uses the eight goodness-of-fit tests (Driesener et al., 2017) to determine if the fresh category buying is in a Dirichlet manner. The tests involve the comparison between the observed and Dirichlet theoretical values for both penetration and average purchase frequency in four ways—comparisons of the averages, correlations, MAD, and MAPE.

The analysis continues with DoP—determining if consumers are purchasing across proprietary, private label, and non-branded fresh fruits and vegetables as expected. That is, brands will share customers in line with their size. To determine if there is higher, lower, or expected consumer sharing, we employ the partition sharing index (PSI). PSI calculates the level of customer sharing between brands of relative size. A PSI of 1.0 shows the brands share customers as expected. A PSI of 0.25 states that brands share 75% fewer customers than expected. A PSI of 1.25 indicates that the two brands share 25% more customers than expected. Sjostrom et al. (2014) stated a PSI > 1.20 warrants further analysis for managerial conclusions. Furthermore, the interpartition PSI calculates the level of customer sharing between all proprietary brands with private label brands and between all proprietary brands with non-branded products. The same analysis occurs for aggregated data with proprietary superbrands.

Results

DJ

Figure 1 shows the observed and theoretical values of the disaggregated tomato data. The Dirichlet workbook, developed by Kearns (2010), calculates the theoretical values for penetration (x-axis) and the average purchase frequencies (y-axis).

Figure 1.

Double jeopardy and Dirichlet benchmark (United States, disaggregate tomato brands, Retailer 2, 2015).

We systematically calculate the DJ observed and theoretical values for all fresh categories and all retailers. Table 1 data show that 44% of disaggregate proprietary brands behave as expected, 18% behave as niche, and 37% as CoP. 67% of private label brands behave as expected, 3% behave as niche, and 40% as CoP. Furthermore, 60% of non-branded products behave as expected, 0% behave as niche, and 40% as CoP.

Table 1.

Overview of double jeopardy (DJ) results (United States, disaggregate and aggregate tomato brands, all retailers, 2015).

Produce categories	n (brands)	Expected (%)	Niche (%)	Change-of-pace (%)
Disaggregated
• Lettuce	14	58	16	26
• Mushrooms	12	56	19	25
• Apples	14	55	5	41
• Carrots	11	55	9	36
• Spinach	18	42	0	58
• Potatoes	9	41	18	41
• Tomatoes	23	41	32	27
• Oranges	15	38	14	48
• Strawberries	20	28	25	47
• Onions	10	27	45	27
• Proprietary brands	146	44	18	37
• Private label	30	67	3	30
• Non-branded	30	60	0	40
Aggregated
• Proprietary brands	30	83	3	13
• Private label	30	57	3	40
• Non-branded	30	60	0	40

When examining the data of the aggregated proprietary brands, similar results were found, with the majority of brands behaving as expected. Interestingly, the classification of brands is different for proprietary brands and private label brands, while non-branded produce follows the same behavior. When the proprietary brands are separate, 44% behave as expected, but are 83% when aggregated. Only the mushroom, spinach, and onion category do not behave as expected; all others behave as DJ predicts. When analyzing private labels, 67% of the separate proprietary brands behave as expected, while 57% when aggregated. The numbers for non-branded produce are identical for both data sets.

The ANOVA results in Table 2 show that proprietary brands and private label brand values differ significantly; while proprietary brands and non-branded, as well as non-branded and private labels, do not differ significantly. Therefore, proprietary brands and private labels have different purchase frequencies and penetrations, while proprietary brands and non-branded, as well as non-branded and private, labels show no difference in purchase frequencies and penetrations across all retailers.

Table 2.

ANOVA results of double jeopardy (DJ) and Dirichlet benchmarks (United States, aggregate tomato brands, all retailers, 2015).

		Proprietary brandsMean/SD	Non-brandedMean/SD	Private labelsMean/SD
Mean of all retailers	Penetration (%)*	27.3^a/12.8	11.9^ab/11.7	0.2^b/0.3
Mean of all retailers	Average purchase frequency*	2.9^a/0.8	1.8^ab/0.7	1.0^b/0.0

ANOVA: analysis of variance; SD: standard deviation.

Different letters a, b signify differences between the brands according to Turkey’s post hoc test.

significance level p = .05.

Employing Driesener et al. (2017) tests of fit, using the data of Figure 1—tomatoes disaggregate analysis for Retailer 2 was 8/8 (minimum 5/8). Table 3 shows 76% (46 of 60) analyses, and the categories have five or more tests passed. The majority of the categories, including non-branded products, operate in a Dirichlet world (Sharp et al., 2012).

Table 3.

Dirichlet model fit (United States, disaggregate and aggregate tomato brands, all retailers, 2015).

Category	Aggregated			Disaggregated
Retailer	1	2	3	1	2	3
Apple	6	8	8	6	8	7
Oranges	6	8	8	6	8	5
Carrots	4	8	8	4	8	7
Lettuce	5	6	8	4	6	8
Mushrooms	4	4	7	4	4	7
Spinach	4	4	7	4	3	5
Tomatoes	6	8	8	4	8	7
Onions	7	8	8	7	7	7
Potatoes	8	8	7	7	8	8
Strawberries	8	4	7	7	3	8

DoP

DoP metricizes customer sharing between disaggregate proprietary brands, and aggregated private labels and non-branded tomatoes superbrands. Table 4 shows Sunset shares 59% of its customers with non-branded tomatoes, 30% with others, 29% with Ol’e vegetables, and 1% with private label brands.

Table 4.

Duplication of purchase table (United States, disaggregate tomato brands, Retailer 1, 2015).

Buyers of . . .	Pen (%)	Who also bought . . .
Buyers of . . .	Pen (%)	Non-branded	Other	Sunset	Ol’e Vegetables	Santa Sweets	Policella	Intergrow	Grecos	5 Reyes	Bonna Terra	Golden Sun	Fresh Point	Peak	Red Sun	Top Line	Flavor 1st Growers and Packers	Private label
Non-branded	25.4		29	16	11	8	7	5	6	6	5	4	3	3	4	2	2	1
Other	10.5	70		19	14	11	7	5	7	6	9	3	2	1	5	3	3	1
Sunset	6.8	59	30		29	4	3	5	5	4	1	4	3	2	2	1	6	1
Ol’e Vegetables	5.2	56	27	38		3	3	3	4	6	1	1	0	0	0	2	3	1
Santa Sweets	3.1	69	36	9	4		7	2	7	10	12	6	12	3	11	12	0	1
Policella	2.7	66	25	7	5	8		0	8	8	9	10	0	10	3	0	0	0
Intergrow	2.1	64	25	17	8	3	0		5	0	2	6	1	1	2	0	4	1
Grecos	2.0	78	37	15	9	10	11	6		9	14	5	0	12	9	0	0	0
5 Reyes	1.7	82	39	17	18	17	13	0	11		16	1	0	0	2	5	0	0
Bonna Terra	1.7	81	54	3	2	22	15	2	17	16		8	0	8	9	4	0	0
Golden Sun	1.7	56	20	18	3	12	16	8	6	1	8		25	18	16	0	2	0
Fresh Point	1.4	51	17	16	1	27	0	1	0	0	0	28		0	16	0	0	0
Peak	1.4	51	7	9	1	6	20	2	18	0	10	21	0		13	0	0	1
Red Sun	1.3	72	42	11	0	25	7	3	13	3	12	20	17	13		0	0	0
Top Line	1.3	48	25	5	7	29	1	0	0	7	6	0	0	0	1		0	1
Flavor 1st Growers and Packers	0.9	56	30	44	17	0	1	9	0	0	0	3	0	0	0	0		2
Private label	0.2	77	54	31	23	9	0	9	3	0	3	3	0	6	3	6	9
Average duplication		65	31	17	10	12	7	4	7	5	7	8	4	5	6	2	2	1

The 1.10 PSI in Table 5 shows non-branded fresh produce share 10% more customers with private label brands than expected. Proprietary tomato brands of Retailer 1 share 10% more of their customers than expected. Overall, they share 12% more of their customers with private labels, while they share 8% less with non-branded tomatoes. The results of Retailer 2, and Retailer 3 of the tomato DoP (Tables 8 and 10) and partition sharing indexes (Tables 9 and 11), are in Appendix 1. We systematically analyze all 10 fresh categories interpartition PSI of private label brands and non-branded produce with proprietary brands.

Table 5.

Partition sharing indexes (United States, disaggregate tomato brands, Retailer 1, 2015).

	Pen (%)	Non-branded	Private label	Other	Peak	Grecos	Golden Sun	Santa Sweets	Red Sun	Fresh Point	5 Reyes	Policella	Sunset	Top Line	Ol’e Vegetables	Intergrow	Flavor 1st Growers and Packers	Bonna Terra
Non-branded	25.4		1.1	1.0	0.7	1.1	0.8	1.0	1.0	0.7	1.2	0.9	0.8	0.7	0.8	0.9	0.8	1.2
Private label	0.2	1.1		1.9	1.5	0.5	0.6	1.0	0.8	0.0	0.0	0.0	1.7	1.6	1.6	1.5	3.5	0.6
Other	10.5	1.0	1.9		0.2	1.3	0.7	1.3	1.4	0.6	1.4	0.9	1.0	0.9	0.9	0.9	1.0	1.9
Peak	1.4	0.7	1.5	0.2		3.2	4.7	0.7	3.5	0.0	0.0	2.7	0.5	0.0	0.1	0.3	0.0	2.0
Grecos	2.0	1.1	0.5	1.3	3.2		1.2	1.2	2.4	0.0	1.9	1.5	0.8	0.0	0.6	1.0	0.0	3.1
Golden Sun	1.7	0.8	0.6	0.7	4.7	1.2		1.4	4.4	6.3	0.3	2.2	0.9	0.0	0.2	1.3	0.7	1.8
Santa Sweets	3.1	1.0	1.0	1.3	0.7	1.2	1.4		2.9	3.2	2.0	0.9	0.5	3.4	0.3	0.4	0.0	2.6
Red Sun	1.3	1.0	0.8	1.4	3.5	2.4	4.4	2.9		4.4	0.6	0.9	0.6	0.1	0.0	0.6	0.0	2.5
Fresh Point	1.4	0.7	0.0	0.6	0.0	0.0	6.3	3.2	4.4		0.1	0.0	0.9	0.0	0.1	0.2	0.0	0.0
5 Reyes	1.7	1.2	0.0	1.4	0.0	1.9	0.3	2.0	0.6	0.1		1.7	0.9	1.4	1.3	0.1	0.0	3.3
Policella	2.7	0.9	0.0	0.9	2.7	1.5	2.2	0.9	0.9	0.0	1.7		0.4	0.1	0.4	0.0	0.1	2.0
Sunset	6.8	0.8	1.7	1.0	0.5	0.8	0.9	0.5	0.6	0.9	0.9	0.4		0.3	2.0	0.9	2.4	0.1
Top Line	1.3	0.7	1.6	0.9	0.0	0.0	0.0	3.4	0.1	0.0	1.4	0.1	0.3		0.5	0.0	0.0	1.2
Ol’e Vegetables	5.2	0.8	1.6	0.9	0.1	0.6	0.2	0.3	0.0	0.1	1.3	0.4	2.0	0.5		0.6	1.2	0.1
Intergrow	2.1	0.9	1.5	0.9	0.3	1.0	1.3	0.4	0.6	0.2	0.1	0.0	0.9	0.0	0.6		1.5	0.4
Flavor 1st Growers and Packers	0.9	0.8	3.5	1.0	0.0	0.0	0.7	0.0	0.0	0.0	0.0	0.1	2.4	0.0	1.2	1.5		0.0
Bonna Terra	1.7	1.2	0.6	1.9	2.0	3.1	1.8	2.6	2.5	0.0	3.3	2.0	0.1	1.2	0.1	0.4	0.0

Managerially significant deviations (0.81.2) are highlighted gray.

Table 6 shows on average in the mushroom category, proprietary brands share 12% more customers than expected; sharing 32% less with private labels, but 380% more with non-branded mushrooms. On average, the PSI of proprietary brands of all retailers is 1.25, which means they are sharing 25% more customers than expected. They share 12% fewer customers with private label and 49% more customers with non-branded produce than expected. Overall, the results show that proprietary brands share more customers than expected; sharing less with private label brands and sharing more customers than expected with non-branded produce.

Table 6.

Partition sharing index (United States, disaggregate tomato brands, all retailers, 2015).

Category average	Partition sharing index (PSI)	Interpartition (PSI)
Category average	Proprietary brands	Private label	Non-branded
Apples	1.41	1.00	0.94
Oranges	1.20	1.23	1.00
Carrots	1.61	0.88	0.99
Lettuce	1.29	0.75	1.64
Mushrooms	1.12	0.68	4.80
Spinach	0.68	1.13	0.96
Tomatoes	1.15	0.81	0.99
Onions	1.09	0.87	1.11
Potatoes	1.40	0.86	1.01
Strawberries	1.13	0.31	1.15
Average Retailer 1	1.17	0.96	1.85
Average Retailer 2	1.26	0.83	1.59
Average Retailer 3	1.34	0.85	1.03
Average of all Retailers	1.25	0.88	1.49

In addition, the results of the interpartition (PSI) of the aggregated data set show that across the 10 categories, for the three retailers, proprietary brands share 2% fewer customers with private label brands, and 17% more customers with non-branded produce, than expected. There is a slight deviation with Retailer 1, where proprietary and private label brands share 15% more customers; potentially due to Retailer 1 highly advertising their private label brands, and thus, consumers are more aware of them than of Retailers 2 and 3. Another reason might be that Retailers 2 and 3 have fewer private label brands in the fresh categories, and therefore, the availability of non-branded produce is higher.

Discussion and implications

Our research enriches the analytical tool kit of marketing researchers by showcasing the application of two known FMCGs consumer behavior models (i.e., DJ, Ehrenberg et al., 1990, and DoP, Goodhardt, 1966) to a novel context. The fresh category economic significance for all supply-chain members (USDA, 2018) urges the development of benchmarks to guide the marketing research industry, particularly with the recent development of proprietary and private label brands.

Our research illustrates the application of consumer behavioral metrices to measure consumer loyalty toward proprietary brands, private label brands, and non-branded produce, and the levels of customer sharing between them. Table 7 shows that all three brands, at an aggregate level, behave as expected—given established FMCGs category benchmarks. The inclusion of non-branded produce in the analysis (complementing the branded and private labels) does not interfere with Dirichlet modeling, and customer sharing is as expected. At a disaggregated level, the patterns are mostly as expected, with a higher percentage of DJ deviations, slightly fewer categories fitting the Dirichlet benchmarks, and some slight oversharing of customers.

Table 7.

Overview of research questions and results.

Research questions	Aggregated brands	Disaggregated brands
RQ1: Are proprietary, private label, and non-branded fresh produce bought as expected, given the double jeopardy pattern?	Yes—only 3% of proprietary brands behave as niche and 40% of private label and non-branded behave as CoP.	Mostly—18% of proprietary brands behave as niche, and 30% of private label and 40% of non-branded behave as CoP.
RQ2: Do categories containing proprietary, private label, and non-branded fresh produce still operate within the Dirichlet goodness-of-fit tests?	Yes—80% of the categories had an acceptable level of fit.	Mostly—73% of the categories had an acceptable level of fit.
RQ3: Do proprietary, private label, and non-branded fresh produce share customers as expected by duplication of purchase law?	Yes—proprietary brands share just 2% fewer customers with private label brands and 17% more customers with non-branded produce.	Mostly—proprietary brands share 25% more customers with themselves, 12% fewer customers with private label, but 49% more with non-branded produce.

For decades, marketers have known that within a store, consumers purchase proprietary and private label brands—they now know they similarly purchase non-branded produce (commodity). Overall, consumers are purchasing non-branded products in the same way they purchase proprietary and private label brands of a similar size. Therefore, non-branded products within a retailer are bought by consumers in the same manner as either proprietary or private label brands in a different retailer. There are some slight differences, in that non-branded products are bought by slightly more consumers, who repeat buy them less frequently than expected. The suspicion is non-branded products do not share the benefits of lower quality and price (Kumar & Steenkamp, 2007; McEnally & Hawes, 1984), or memory structures of private labels (Nenycz-Thiel et al., 2009; Nenycz-Thiel & Romaniuk, 2012). Similarly, non-branded products do not have the distinguishability or consistent quality of proprietary brands (Keller & Lehmann, 2006; Laaksonen, 1994), which increase the probability of repeat purchase (Ehrenberg et al., 2004; Goodhardt et al., 1984).

Overall, with disaggregate proprietary brands, the results matched earlier FMCG studies (i.e., slightly higher incidence of CoP, rather than niche) (Scriven et al., 2017). In contrast, the aggregated results differ from earlier studies as they show fewer incidences of both niche and CoP. The lower likelihood of fresh brands being CoP might be due to modern cultivation in a controlled environment (e.g., heated greenhouse and year-round availability).

Where there were deviations, it is due to functionality at a category level. That is, many people purchase from the category, but not often (do not desire the taste often), where the category is only required occasionally (only used in select recipes). There are also instances at a brand level, where the private label and non-branded fresh produce are less likely to be niche. This occurs when it is harder to communicate lower quality and lower price to consumers. One reason for this could be the lack of packaging. In an age of consumer backlash against plastic packaging (Klaiman et al., 2016), retailers may opt to distinguish themselves by offering a unique variety, or to communicate sensory cues such as taste or texture (Imm et al., 2012; Swahn et al., 2012). The apple category contains the botanical variety of communication (e.g., Jonagold, Elstar, Granny Smith), while the tomato category sometimes communicates taste and usability. CoP results are similar for the aggregated and disaggregated data.

When buying from the category on multiple occasions, consumers are purchasing across proprietary, private label, and non-branded products. Consumers are not limiting their purchases to a single category, having purchased from the cheapest option or the most consistent quality option. Instead, these results indicate that consumers contain all three types of brands within their repertoire. Consumers are purchasing the more expensive, highly consistent quality proprietary brands and the non-branded products that are cheap and vary in quality. In the instance where proprietary brands share most customers and fewer with private label or non-branded products, the retailer may have more proprietary brands in their assortment. Furthermore, the customer does not always have a choice between all three brand options. Some produce might only be available in one brand version.

Theoretical implications

The research has several theoretical contributions. This research expands the known conditions of the NBD-Dirichlet model—more specifically, the DJ and DoP patterns (Ehrenberg et al., 1990; Goodhardt, 1966), into a new context of frequently bought commodity products—fresh fruits and vegetables. Of particular novelty is the examination of a condition of historical dominance of non-branded alternatives within the category, and the recent emergence of proprietary and private label brands.

Our research builds on and reinforces the earlier research of Anesbury, Greenacre, et al. (2018), who analyze fresh category loyalty with claimed behavioral data. The current research analyses observed behavioral data, at a brand level, once again finding the same FMCG consumer behavior patterns. The new evidence within this study demonstrates that even a category that is severely affected by seasonality, price, origin, availability, and quality of fresh fruit and vegetables (Rekhy & McConchie, 2014) still operates in a Dirichlet world (Sharp et al., 2012).

One questionable aspect of previous Dirichlet model deviations is the seemingly arbitrary 10% deviation benchmark (Anesbury, Nguyen, & Bogomolova, 2018; Kahn et al., 1988; Scriven et al., 2017). Our research, while using this approach, provides additional support for the work of Driesener et al. (2017) who proposed using eight goodness-of-fit measures to provide a more robust determination of Dirichlet modeling.

Finally, our research adds additional support for the Dirichlet model’s ability to go beyond an aggregated “all other” brand (Goodhardt et al., 1984). That is, the aggregation of all remaining small brands—(Driesener et al., 2017; Ehrenberg et al., 2004; Graham et al., 2017; Scriven et al., 2017), to the creation of multiple distinct grouping of brands sharing a similar characteristic (i.e., superbrands—Goodhardt et al., 1984; Habel & Lockshin, 2013).

Practical implications

The research helps marketers operating in the fruit and vegetable markets to better understand how consumers purchase proprietary, private label, and non-branded products. It shows that known consumer buying behavior patterns observed in the FMCG literature for decades (Ehrenberg et al., 1990; Goodhardt, 1966) are also evident in this novel context. This applicability means that fresh marketers can increase consumption by increasing penetration, rather than loyalty (Romaniuk et al., 2014). Remarkably, this is true regardless of the brand being proprietary, private label, or non-branded.

Those who manage proprietary, or private label, brands still have a clear advantage over non-branded produce. They can build CBBE (Aaker, 1991; Keller, 1993; Zeithaml, 1988) and mental availability (Romaniuk, 2013, 2016). Furthermore, proprietary brands have a distinct advantage over private label brands. Their ability to be bought from multiple retailers (Kumar & Steenkamp, 2007) means they can increase their physical availability (Romaniuk & Sharp, 2016), beyond anything a private label could ever hope to achieve (while keeping in mind physical availability can also be improved within store via relevance and prominence).

Given that consumers of proprietary brands purchase private label and non-branded produce, and vice versa, retailers must stock a variety of brands. The clear display of polygamous loyalty, manifesting as repertoire buying behavior (Banelis et al., 2013; Trinh et al., 2017), is evident even in a category where brands are only recently emerging. For growers, the expectation is that consumers will not solely purchase only that brand, or love that brand (Batra et al., 2012), and setting such goals is unrealistic and counterproductive to growth.

Other studies show that consumers are willing to pay price premiums for proprietary brands, but not for non-branded products (Sethuraman, 2003; Steenkamp et al., 2010). Consumers, in turn, benefit from the typical advantages of branded products. They benefit from consistent quality (Zeithaml, 1988), ease of finding the fresh produce within stores (Shocker et al., 1991), and, therefore, their ability to repeat purchase (i.e., be loyal customers) (Kotler, 1967). From a retailer’s perspective, creating their own private label brands allows them to control their in-store space further, give their products a better positioning, and add merchandizing, which may increase sales and profitability (Romaniuk et al., 2014).

For those working in marketing research and analytical roles, they can confidently include non-branded products within their NBD-Dirichlet analyses. They can be sure to know that consumers will have similar levels of loyalty toward non-branded products as they do for proprietary or private label brands of the same size, and that consumers will purchase predictably. These insights allow marketing researchers to expand their analytical tool kit when they work with fresh categories data. They now have benchmarks for what to expect when they apply the DJ and DoP analyses to proprietary, private label, and commodity-type products of fruits and vegetables.

Furthermore, the introduction of prepacked fresh offers (as proprietary brands), such as perfectly sized fruit and vegetables to fit kids’ lunch boxes, meet consumers’ need for convenience—an increasingly prevalent trend. For retailers, this allows the expansion of the fresh offering, from nonbranded (loose), to proprietary or private label (prepacked) options. Given that fresh produce generally has higher profit margins than other categories in the supermarket, expanding the fresh offer will contribute to the retailer’s overall bottom line. In addition, including misshapen fresh produce at a lower price point could meet consumers’ emerging demand for strategies to reduce food waste, while also offering customers with different paying abilities more choices. The result could also expand the overall fresh sales, as cheaper options or more convenient prepacked offers could bring to the market new consumers who would not have bought fresh categories before. For example, those who could not afford it (for cheaper products), or those who were living in smaller apartments and did not want to deal with raw produce (i.e., whole cabbage, zucchini, carrot), but would happily buy a prewashed, precut pack for a stir-fry.

Limitation and future research

Data used within this article was solely from the United States. Extending the findings to other countries will generalize the results and account for any cultural differences. The study analyzes brands, but did not account for different variants, including organic production, price-premium, or sustainably sourced produce. Future research should replicate this study to these contexts—previously done for variants (Singh et al., 2008) and categories (Dunn et al., 2019).

Future research should investigate the relationship between price loyalty and brand loyalty. Specifically, if consumers are loyal toward the proprietary, private label, or non-branded products, or if they are loyal toward the price tiers (i.e., always the lowest price within the category). Customers may set cut-off prices in their mind and only purchase brands below the line. This research analyzes one private label type. Retailers are introducing different private label qualities, and future research should analyze the different quality levels of private labels.

Moreover, future research should examine if and how consumers perceive proprietary, private-label, and non-branded items. Previous FMCG research examined trust, value for money, and quality aspects (Nenycz-Thiel & Romaniuk, 2016). Furthermore, extending the research of Nenycz-Thiel and Romaniuk (2014) for FMCG may allow for the examination of patterns in brand image data for private label and proprietary brands—including non-branded.

Finally, while the 10% deviation of observed from theoretical values are considered the Dirichlet benchmark (Anesbury, Nguyen, & Bogomolova, 2018; Kahn et al., 1988; Scriven et al., 2017), the percentage is arbitrary. Further research needs to establish an empirically grounded deviation threshold or an equation for calculating the threshold, possibly using the category penetration or purchase rate.

Footnotes

Appendix 1

Table 11.

Partition sharing indexes (United States, disaggregate tomato brands, Retailer 3, 2015).

	Pen (%)	Non-branded	Private label	Desert glory NatureSweet	Desert Glory NatureSweet Cherub	Sunbursts	NatureSweet	Other	Village Farms	Chubby Cheeks	Victory Garden	Andrew & Williamson Fresh Produce-Nbl	Flavor-Pic Tomato Co.	Red Sun	Santa Sweets	Del Cabo	Sunset Y.E.L.O	Sunset	Green Giant	Sunset Zima	Del Campo
Non-branded	4.8		0.4	1.4	0.7	0.8	0.6	0.8	0.7	0.9	0.8	0.8	0.4	1.1	0.9	0.9	1.0	0.9	0.3	1.2	1.6
Private label	0.5	0.4		0.5	1.5	1.7	0.3	0.6	0.6	1.0	0.0	0.5	0.6	0.3	0.6	0.2	0.0	0.7	0.0	0.0	0.4
Desert Glory NatureSweet	0.5	1.4	0.5		0.8	4.3	0.4	1.5	1.7	0.5	1.5	2.1	0.7	1.2	1.6	2.2	2.9	1.4	0.5	2.1	3.0
Desert Glory NatureSweet Cherub	0.7	0.7	1.5	0.8		5.7	4.0	0.6	0.9	0.0	1.3	2.1	0.0	0.9	0.7	0.7	0.0	0.5	0.7	0.0	1.0
Sunbursts	0.1	0.8	1.7	4.3	5.7		9.4	1.0	1.1	0.0	0.0	0.0	0.0	1.2	2.5	0.9	0.0	0.3	0.0	0.0	1.7
NatureSweet	0.3	0.6	0.3	0.4	4.0	9.4		0.6	0.5	0.0	0.7	0.7	0.0	1.0	1.0	0.4	0.7	0.3	0.8	1.2	0.0
Other	3.8	0.8	0.6	1.5	0.6	1.0	0.6		1.3	1.2	0.9	0.6	0.8	1.0	0.7	0.9	1.8	0.9	0.5	1.6	1.1
Village Farms	0.4	0.7	0.6	1.7	0.9	1.1	0.5	1.3		3.2	0.0	0.6	0.0	1.8	0.2	2.2	4.4	2.3	0.6	5.4	0.6
Chubby Cheeks	0.2	0.9	1.0	0.5	0.0	0.0	0.0	1.2	3.2		0.0	0.0	0.0	0.0	1.6	0.0	0.9	0.8	0.0	0.0	1.9
Victory Garden	0.3	0.8	0.0	1.5	1.3	0.0	0.7	0.9	0.0	0.0		1.9	0.0	0.6	1.0	2.4	0.8	0.4	0.9	0.7	0.9
Andrew & Williamson Fresh Produce-Nbl	0.3	0.8	0.5	2.1	2.1	0.0	0.7	0.6	0.6	0.0	1.9		0.0	1.0	0.9	0.5	0.9	0.5	0.0	0.8	0.9
Flavor-Pic Tomato Co.	0.2	0.4	0.6	0.7	0.0	0.0	0.0	0.8	0.0	0.0	0.0	0.0		0.0	0.2	0.0	0.0	0.0	0.0	0.0	1.2
Red Sun	0.8	1.1	0.3	1.2	0.9	1.2	1.0	1.0	1.8	0.0	0.6	1.0	0.0		0.5	1.8	2.3	1.4	0.3	1.3	1.2
Santa Sweets	1.5	0.9	0.6	1.6	0.7	2.5	1.0	0.7	0.2	1.6	1.0	0.9	0.2	0.5		0.2	0.6	0.7	0.7	0.6	1.3
Del Cabo	0.5	0.9	0.2	2.2	0.7	0.9	0.4	0.9	2.2	0.0	2.4	0.5	0.0	1.8	0.2		3.1	2.0	0.0	4.0	1.0
Sunset Y.E.L.O	0.3	1.0	0.0	2.9	0.0	0.0	0.7	1.8	4.4	0.9	0.8	0.9	0.0	2.3	0.6	3.1		3.8	0.0	14.1	1.7
Sunset	1.7	0.9	0.7	1.4	0.5	0.3	0.3	0.9	2.3	0.8	0.4	0.5	0.0	1.4	0.7	2.0	3.8		0.0	4.4	1.0
Green Giant	0.3	0.3	0.0	0.5	0.7	0.0	0.8	0.5	0.6	0.0	0.9	0.0	0.0	0.3	0.7	0.0	0.0	0.0		0.0	0.9
Sunset Zima	0.3	1.2	0.0	2.1	0.0	0.0	1.2	1.6	5.4	0.0	0.7	0.8	0.0	1.3	0.6	4.0	14.1	4.4	0.0		0.7
Del Campo	0.3	1.6	0.4	3.0	1.0	1.7	0.0	1.1	0.6	1.9	0.9	0.9	1.2	1.2	1.3	1.0	1.7	1.0	0.9	0.7

Managerially significant deviations (0.81.2) are highlighted gray.

Declaration of conflicting interests

Researcher(s) own analyses calculated (or derived) based in part on data from The Nielsen Company (US), LLC, and marketing databases provided through the Nielsen Datasets at the Kilts Centre for Marketing Data Center at The University of Chicago Booth School of Business. The conclusions drawn from the Nielsen data are those of the researcher(s) and do not reflect the views of Nielsen. Nielsen is not responsible for, had no role in, and was not involved in analyzing and preparing the results reported herein.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Zachary William Anesbury

References

Aaker

D. A.

(1991). Managing brand equity: Capitalizing on the value of a brand name. The Free Press.

Ailawadi

K. L.

Neslin

S. A.

Gedenk

(2001). Pursuing the value-conscious consumer: Store brands versus national brand promotions. Journal of Marketing, 65, 71–89.

Anesbury

Greenacre

Wilson

Huang

(2018). Patterns of Fruit and Vegetable Buying Behaviour in the United States and India. International Journal of Market Research, 60(1), 14–31.

Anesbury

Nguyen

Bogomolova

(2018). Getting a “Sweet” deal: Does healthfulness of an SKU influence consumer loyalty? European Journal of Marketing, 52(9–10), 1802–1826.

Baltas

(2003). A combined segmentation and demand model for store brands. European Journal of Marketing, 37(10), 1499–1513.

Banelis

Riebe

Rungie

(2013). Empirical evidence of repertoire size. Australasian Marketing Journal, 21(1), 59–65.

Batra

Ahuvia

Bagozzi

(2012). Brand love. Journal of Marketing, 76(2), 1–16.

Batra

Sinha

(2000). Consumer-level factors moderating the success of private label brands. Journal of Retailing, 76(2), 175–191.

Bogomolova

Anesbury

Kapulski

Lockshin

Bogomolov

(2019). Exploring the incidence and antecedents of buying an FMCG Brand and UPC for the first time. Journal of Retailing & Consumer Services, 46, 121–129.

10.

Bound

J. A.

Ehrenberg

(1989). Significant sameness. Journal of the Royal Statistical Society, 152, 241–247.

11.

Bryła

(2018). Organic food online shopping in Poland. British Food Journal, 120, 1015–1027.

12.

Cuneo

Milberg

S. J.

del Carmen Alarcon-del-Amo

Lopez-Belbeze

(2019). Private label and manufacturer brand choice in a new competitive reality: Strategic directions and the future of brands. European Management Journal, 37(1), 117–128.

13.

Dawes

Bond

Hartnett

Sharp

(2017). Does double jeopardy apply using average spend per buyer as the loyalty metric? Australasian Marketing Journal, 25, 261–268.

14.

Dawes

Nenycz-Thiel

(2013). Analyzing the intensity of private label competition across retailers. Journal of Business Research, 66(1), 60–66.

15.

Driesener

Rungie

C. E.

Banelis

(2017). If the model fits, use it: Methods and benchmarks for evaluating NBD-Dirichlet goodness-of-fit. Australasian Marketing Journal, 25, 288–293.

16.

Dunn

Nenycz-Thiel

Dawes

(2019, December). Double jeopardy in product category purchasing. In ANZMAC (pp. 1–4). Victoria University of Wellington.

17.

Ehrenberg

(1966). Regularities of behaviour. Applied Statistics, 15, 13–27.

18.

Ehrenberg

(1990). A hope for the future of statistics: MSOD. The American Statistician, 44(3), 195–196.

19.

Ehrenberg

Goodhardt

(1969). Practical applications of the duplication of viewing law. Journal of the Market Research Society, 11(1), 6–24.

20.

Ehrenberg

Goodhardt

(2002). Double jeopardy revisited, again. Marketing Insights, Marketing Research, 14(1), 40–42.

21.

Ehrenberg

Goodhardt

Barwise

T. P.

(1990). Double jeopardy revisited. Journal of Marketing, 54(3), 82–91.

22.

Ehrenberg

Uncles

M. D.

Goodhardt

G. G.

(2004). Understanding brand performance measures: Using Dirichlet benchmarks. Journal of Business Research, 57(12), 1307–1325.

23.

Goodhardt

G. J.

(1966). Constant in duplicated television viewing. Nature, 212(5070), Article 1616.

24.

Goodhardt

G. J.

Ehrenberg

A. S. C.

(1969). Duplication of television viewing between and within channels. Journal of Marketing Research, 6(2), 169–178.

25.

Goodhardt

G. J.

Ehrenberg

A. S. C

Chatfield

(1984). The Dirichlet: A comprehensive model of buying behaviour. Journal of the Royal Statistical Society, 147(5), 621–655.

26.

Graham

Bennett

Franke

Henfrey

C. L.

Nagy-Hamada

(2017). Double jeopardy–50 years on. Reviving a forgotten tool that still predicts brand loyalty. Australasian Marketing Journal (AMJ), 25(4), 278–287.

27.

Greenacre

Tanusondjaja

Dunn

Page

(2015). Using choice experiments to find double jeopardy patterns. International Journal of Market Research, 57(5), 743–758.

28.

Habel

Lockshin

(2013). Realizing the value of extensive replication: A theoretically robust portrayal of double jeopardy. Journal of Business Research, 66(9), 1448–1456.

29.

Hansen

Singh

Chintagunta

(2006). Understanding store-brand purchase behavior across categories. Marketing Science, 25(1), 75–90.

30.

Harris

B. F.

Strang

R. A.

(1985). Marketing strategies in the age of generics. Journal of Marketing, 49(4), 70–81.

31.

Hoch

S. J.

Banerji

(1993). When do private labels succeed? Sloan Management Review, 34(4), 57–68.

32.

Hutcheson

G. D.

Moutinho

(1998). Measuring preferred store satisfaction using consumer choice criteria as a mediating factor. Journal of Marketing Management, 14(7), 705–720.

33.

Hyman

M. R.

Kopf

D. A.

Lee

(2010). Review of literature–future research suggestions: Private label brands: Benefits, success factors and future research. Journal of Brand Management, 17(5), 368–389.

34.

Imm

B.-Y.

Lee

J. H.

Lee

S. H.

(2012). Effects of sensory labels on taste acceptance of commercial food products. Food Quality and Preference, 25(2), 135–139.

35.

Kahn

B. E.

Kalwani

M. U.

Morrison

D. G.

(1988). Niching versus change-of-pace brands: Using purchase frequencies and penetration rates to infer brand positionings. Journal of Marketing Research, 25(4), 384–390.

36.

Kearns

(2010). Dirichlet.Xls. Massey University.

37.

Keller

K. L.

(1993). Conceptualizing, measuring, and managing customer-based brand equity. Journal of Marketing, 57(1), 1–22.

38.

Keller

K. L.

(2001). Building customer-based brand equity. Marketing Management, 10(2), 14–19.

39.

Keller

K. L.

Lehmann

D. R.

(2006). Brands and branding: Research findings and future priorities. Marketing Science, 25(6), 740–759.

40.

Kilts Centre for Marketing. (2019). Nielsen Datasets. The University of Chicago Booth School of Business.

41.

Klaiman

Ortega

D. L.

Garnache

(2016). Consumer preferences and demand for packaging material and recyclability. Resources, Conservation and Recycling, 115, 1–8.

42.

Kotler

(1967). Marketing management: Analysis, planning, and control. Prentice-Hall.

43.

Kumar

Steenkamp

J.-B. E. M.

(2007). Private label strategy: How to meet the store brand challenge. Harvard Business Press.

44.

Laaksonen

(1994). Own brands in food retailing across Europe. Oxford Institute of Retail Management.

45.

Lees

Wright

(2013). Does the duplication of viewing law apply to radio listening? European Journal of Marketing, 47(3–4), 674–685.

46.

McEnally

Hawes

(1984). The market for generic brand grocery products a review and extension. Journal of Marketing, 48, 75–83.

47.

McPhee

W. N.

(1963). Formal theories of mass behavior. Free Press.

48.

Nenycz-Thiel

Romaniuk

(2012). Value-for-money perceptions of supermarket and private labels. Australasian Marketing Journal, 20(2), 171–177.

49.

Nenycz-Thiel

Romaniuk

(2014). The real difference between consumers’ perceptions of private labels and national brands. Journal of Consumer Behaviour, 13(4), 262–269.

50.

Nenycz-Thiel

Romaniuk

(2016). Understanding premium private labels: A consumer categorisation approach. Journal of Retailing and Consumer Services, 29, 22–30.

51.

Nenycz-Thiel

Sharp

Dawes

Romaniuk

(2009). Competition for memory retrieval between private label and national brands. Journal of Business Research, 63(1), 1142–1147.

52.

Nielsen. (2018). Private-label brands are hungry for more of the global food pie. https://www.nielsen.com/us/en/insights/article/2018/private-label-brands-are-hungry-for-more-of-the-global-food-pie/

53.

Prendergast

G. P.

Marr

N. E.

(1997). Generic products: Who buys them and how do they perform relative to each other? European Journal of Marketing, 31(2), 94–109.

54.

Rekhy

McConchie

(2014). Promoting consumption of fruit and vegetables for better health. Have campaigns delivered on the goals? Appetite, 79, 113–123.

55.

Romaniuk

(2013). Modeling mental market share. Journal of Business Research, 66(2), 188–195.

56.

Romaniuk

(2016). Building mental availability. In Romaniuk

Sharp

(Eds.), How brands grow: Part 2 (pp. 62–86). Oxford University Press.

57.

Romaniuk

Dawes

Nenycz-Thiel

(2014). Generalizations regarding the growth and decline of manufacturer and store brands. Journal of Retailing and Consumer Services, 21(5), 725–734.

58.

Romaniuk

Sharp

(2016). How brands grow: Part 2. Oxford University Press.

59.

Scriven

Bound

Graham

(2017). Making sense of common Dirichlet deviations. Australasian Marketing Journal (AMJ), 25(4), 294–308.

60.

Sethuraman

(2003). Measuring national brands’ equity over store brands. Review of Marketing Science, 1(2), 1–25.

61.

Sharp

(2010). How brands grow. Oxford University Press.

62.

Sharp

Ehrenberg

(2000). Managerial uses of descriptive marketing models—The case of the Dirichlet. In ANZMAC (pp. 296–301). Griffith University.

63.

Sharp

Wright

Dawes

Driesener

Meyer-Waarden

Stocchi

Stern

(2012). It’s a Dirichlet world: Modeling individuals’ loyalties reveals how brands compete, grow, and decline. Journal of Advertising Research, 52(2), 203–213.

64.

Shocker

A. D.

Ben-Akiva

Boccara

Nedungadi

(1991). Consideration set influences on consumer decision-making and choice: Issues, models, and suggestions. Marketing Letters, 2(3), 181–197.

65.

Singh

Ehrenberg

Goodhardt

(2008). Measuring customer loyalty to product variants. International Journal of Market Research, 50(4), 513–532.

66.

Sjostrom

Corsi

A. M.

Driesener

Chrysochou

(2014). Are food brands that carry light claims different? Journal of Brand Management, 21(4), 325–341.

67.

Steenkamp

J.-B. E. M.

van Heerede

H. J.

Geyskens

(2010). What makes consumers willing to pay a price premium for national brands over private labels? Journal of Marketing Research, 47(6), 1011–1024.

68.

Stern

(1994). Prescriptions for branded and generic pharmaceuticals. Journal of Brand Management, 2(3), 177–183.

69.

Swahn

Mossberg

Öström

Å.

Gustafsson

I. B.

(2012). Sensory description labels for food affect consumer product choice. European Journal of Marketing, 46(11–12), 1628–1646.

70.

Trinh

Anesbury

Driesener

(2017). Has behavioural loyalty to online supermarkets declined? Australasian Marketing Journal, 25, 1–24.

71.

Uncles

M. D.

Kwok

(2009). Patterns of store patronage in urban China. Journal of Business Research, 62(1), 68–81.

72.

USDA. (2018). Ag and food sectors and the economy. https://www.ers.usda.gov/data-products/ag-and-food-statistics-charting-the-essentials/ag-and-food-sectors-and-the-economy.aspx

73.

Winchester

Arding

Nenycz-Thiel

(2015). An exploration of consumer attitudes and purchasing patterns in fair trade coffee and tea. Journal of Food Products Marketing, 21(5), 552–567.

74.

Wrigley

Dunn

(1984). Stochastic panel-data models of urban shopping behaviour: 2. multistore purchasing patterns and the Dirichlet model. Environment and Planning A, 16, 759–778.

75.

Zeithaml

V. A.

(1988). Consumer perceptions of price, quality, and value: A means-end model and synthesis of evidence. The Journal of Marketing, 52(3), 2–22.