One of the major trends in branding is for firms to have fewer, stronger “power” or mega brands (Hill et al. 2005). Such brands often offer products in multiple diverse categories. Taking the lead from brands such as Virgin, Yamaha, and Samsung, which have used their brands across disparate product categories for years, numerous marketers have transformed customers’ conceptions of their brands through a series of category extensions. For example, Swiss Army successfully expanded from knives to watches and luggage; Zippo went beyond lighters to introduce watches, fragrances, and knives; and Samsonite went from luggage to high-end men’s shoes, furniture, and stationery. In the supermarket, Lipton is associated with both tea and soup; Pepperidge Farm is associated with cookies, crackers, and bread; and, in the U.K., Heinz is associated with ketchup and baked beans. Likewise, Crayola has introduced a host of products beyond crayons, such as markers, paints, and other colorful arts and crafts products. Stretching the envelope even further, companies like GE and 3 M have entered notably disparate categories as they have grown their brands.

As companies seek to build multi-category brands by adding new products to their product lines, two questions loom large: (1) whether and (2) when the company should introduce products perceived by consumers as distant (dissimilar) from its existing core line of products (e.g., an apparel company introducing a food product). While the question of whether a brand should introduce a distant product is complex, there are certainly numerous examples of brands that have introduced distant products, both successfully and unsuccessfully. Hence, in this paper, we assume that a brand has made an educated and measured decision to enter a distant product category (i.e., the company sees enough potential in the new market—or diminishing potential in its current market—to be willing to take the risks associated with entering an unfamiliar category and attaching their brand name to their entry in it). Therefore, we focus on the second question: When should the brand introduce its first distant product?

While there is considerable research on brand extensions (see Völckner and Sattler 2006 for an extensive list), there is surprisingly little guidance for managers hoping to expand into distant categories. Consequently, this paper examines how the timing of introducing a brand’s first distant product (hereafter, FDP) influences consumers’ brand attitudes. We compare early introductions of FDPs (i.e., when the FDP is introduced before other, closer products) with late introductions of FDPs (i.e., when the FDP is introduced after other, closer products).

As we will detail, much previous research on how to grow or “stretch” a brand recommends entering distant product categories only after gradually introducing products increasingly more distant from the brand’s core. However, we argue and demonstrate that the assumptions and reasoning underlying this measured, deliberate process do not apply to companies planning to enter distant product categories more abruptly. Specifically, building on theories of mental categorization, we predict and demonstrate across three studies that, while introducing close products early is initially beneficial, early (vs. late) FDP introductions can result in more positive final brand attitudes; that is, brand attitudes when the brand has reached the end of its planned extension sequence. Critically, this holds true despite the fact that the brand ultimately offers identical product portfolios containing the same set of products regardless of the brand extension sequence employed.

We further demonstrate that the relationship between brand extensions and brand attitudes is mediated by consumers’ self-reported brand concept fluency and, accordingly, that this relationship between extensions and attitudes can be altered by established brand positioning tactics. We additionally demonstrate that the timing of FDP introductions can impact consumers’ preferences for the brand’s products and their likelihood of engaging with the brand.

In addition to testing our proposed theoretical process, a key objective of the paper is to provide managers insight and guidance regarding FDP introduction timing. Thus, we close the paper with discussions on (1) boundary conditions and moderators of the documented effect that are derived from our theoretical framework, (2) the limitations of the current work, and (3) directions for future research.

Conceptual background

Strategies for introducing distant brand extensions

Brand extension research is largely centered on two broad issues: (1) how consumers evaluate a brand extension and (2) how these evaluations, in turn, impact attitudes toward the parent brand of that extension. Relevant to the current work, research has found that distant extensions tend to be less favorably evaluated by consumers (e.g., Aaker and Keller 1990; Boush and Loken 1991) and therefore can damage or dilute the image of the parent (flagship) brand under certain circumstances (e.g., Gürhan-Canli and Maheswaran 1998; John et al. 1998). Indeed, much research argues that brands, new and established, benefit greatly from being consistent and reinforcing the brand identity (Aaker 2014; Keller 1999; Park, Jaworski, and MacInnis). Thus, the conclusion one might draw is that firms should not introduce distant brand extensions. Hence, it is important to understand why a brand might introduce a distant extension before discussing when it should introduce this extension.

Currently, much of the extant brand extension literature fails to consider that the increased risk of introducing a distant extension can be offset by gains related to the potential benefits from retaining a perception of relevance in the market (Aaker 2012; Beverland et al. 2015), tapping into a new target market, the presence of minimal competition in the distant extension category, creating more product variety for the brand, or a combination of these and other factors (see Chun et al. [2015] for a recent exception). Indeed, multi-category brands offer a host of financial benefits to a firm, such as expanding revenue streams, diversifying sales risk, and the potential for synergies and efficiencies. Moreover, introducing distant products as extensions is consistent with the advice given to marketing managers to avoid “marketing myopia” (Levitt 1960), and distant brand extensions can play an important strategic role for a brand, as evidenced by numerous companies successfully introducing products as brand extensions in diverse categories.

Some prior brand extension research certainly supports these brand breadth advantages. Boush and Loken (1991) showed that consumers evaluated far extensions from a “broad” brand more favorably than from a “narrow” brand. Dawar (1996) showed the memory drawbacks from trying to expand a narrow brand, finding that retrieval inhibition effects could reduce the activation of “less relevant” product associations for brands with a strong association to a single product, lowering perceived fit of extensions close to such products. Along the same lines, Meyvis and Janiszewski (2004) found that broad brands tended to have more accessible benefit associations than narrow brands and could therefore engage in more successful brand extensions than narrow brands, even when the narrow brands were more similar to the extension category.

Although this prior research documents some advantages of broader brands, questions remain as to what are the best possible ways to actually expand a brand to achieve that breadth in the first place. Table 1 lists successful and unsuccessful brand extensions, both distant and close to the core brands, based on marketplace success and conventional definitions of brand extension fit. These lists demonstrate two important characteristics of real-world extensions. First, if nothing else, they show that brand managers are willing to introduce distant brand extensions. More importantly, they demonstrate that both close and distant brand extensions can succeed or fail. This reality suggests that the influence on success of an extension’s distance (i.e., lack of fit) from the brand’s core (at least for the specific extension) is more nuanced than might be suspected. In other words, there are other factors that moderate the impact of extension distance on both the success of the extension and consumers’ attitudes toward the brand.

Table 1 Real-world brand extensions
Full size table

In sum, distant brand extensions are common and can be rationally motivated: “Distant” extensions are not inherently “bad” extensions. Instead, the important question is how to maximize the net impact of these extensions. Thus, we take it as given in this paper that the managers of a brand have made an informed, strategic decision to introduce a distant brand extension as part of an expanded product portfolio for the brand and turn our attention to the focus of this paper: When should the first distant product (FDP) be introduced?

Intuition might suggest that the brand should delay introducing its FDP as long as possible. In fact, a pilot study found that 104 of 115 (90%) Executive MBA students had this exact intuition: When presented with the stimuli used in our studies, they overwhelmingly preferred the scenario in which the brand delayed introducing its FDP to the very end of the extension sequence (as opposed to it being the first extension introduced).

Previous research examining related but distinct topics (Dawar and Anderson 1994; Keller and Aaker 1992; Loken and John 1993; Swaminathan 2003) supports their intuitions, suggesting that brands should first introduce close extensions and delay introducing FDPs. Similarly, more recent research indicates that “weak” innovations (i.e., innovations resulting in a product more similar to the original) perform better when introduced earlier, while “strong” innovations (i.e., innovations resulting in a product more dissimilar to the original) perform better when introduced later (Heath et al. 2016; Study 3).

Yet, an important aspect of the research discussed above is that it examines contexts in which the brand has the desire and/or ability to slowly introduce products increasingly distinct from its core in an attempt to gradually reach one or more distant product categories (or, in the case of Heath et al. [2016], make increasingly radical innovations to its core product). These intervening extensions reduce the perceived dissimilarity of the distant category from the brand as a whole. Thus, by first gradually and systematically stretching the brand toward the target product category, there never really is a “distant” product introduced.

In contrast, we explore the advantages of having the brand expand its product portfolio more quickly by introducing a distant product much sooner and not gradually after numerous intervening extensions. To date, no research has examined this context and, therefore, managers making such decisions largely rely on their intuition and indirect evidence. Accordingly, the current work contrasts the impact of brand extension sequences in which the FDP is introduced early (prior to other, more similar extensions) versus late (after other, more similar extensions). Building on the premise that consumers treat brands and their product portfolios as categories (Boush and Loken 1991; Park et al. 1991), we draw on theories of mental categorization for insight into when FDPs should be introduced.

Mental categorization and consumers’ brand concepts

Mental representations are internal cognitive symbols of external, real-world stimuli (Markman 1999). When discussing mental categorization, the term concept is used to refer to the consumer’s mental representation of the class of stimuli (i.e., the mental category), whereas the term category is used to refer to the actual class of stimuli itself (Murphy 2004). As a concrete example, consider the category “dogs,” which includes all dogs that were, are, and will be. Of course, none of us has encountered all dogs past, present, and future. Yet we still have a concept of dogs, which allows us to—among other things—determine if newly encountered stimuli are or are not dogs.

The focus in the current work is on brand concepts—consumers’ mental representations of brands (Keller 2003; Park et al. 1986)—which are informed both by the products the brand sells and by other brand-specific information. We are specifically concerned with how consumers’ brand concepts evolve, particularly as a function of the portfolio of products the brand introduces and sells. As consumers encounter a brand’s extensions—that is, the brand’s exemplars (Medin and Schaffer 1978; Reinholtz et al. 2015; Rosch et al. 1976)—they learn about what is typical or prototypical of the brand (i.e., their brand concept is updated). This process highlights how brand concepts are meaningfully distinct from natural concepts such as dogs, trees, and the like.

As opposed to natural concepts, there is little ambiguity or subjectivity about which stimuli (i.e., products) are members of a brand concept: If the consumer knows a brand sells a product, that product is necessarily a part of that consumer’s brand concept. Thus, while a newly-encountered six-legged animal can easily be dismissed as not being a dog—and, thus, the concept of dog would remain unchanged—new brand extensions are necessarily part of the real-world brand category and thus impact the consumer’s brand concept.

The influence of a new brand extension on the consumer’s brand concept will be a function of how close (similar) or distant (dissimilar) the new product is relative to the brand’s current products (Love et al. 2004). Importantly, the ease with which a consumer is able to integrate the new exemplar and, subsequently, make sense of and understand the brand concept is likely to influence the consumer’s brand attitude, as we discuss next.

FDP introduction timing, brand concept fluency, and brand attitudes

Attitudes toward something are often influenced by how easily or fluently that thing is processed and understood (Alter and Oppenheimer 2009): Conceptually disfluent (vs. fluent) stimuli tend to be evaluated less favorably due to the negative emotions arising from disfluent processing being attributed to the source of the disfluency (Reber et al. 1998; Seamon et al. 1995; Winkielman and Cacioppo 2001). Brand concept fluency refers to the ease or difficulty with which the brand concept is processed and understood.

It is worthwhile to note at this point that brand concept fluency and brand extension fit (Keller 2002) are related but distinct concepts. Brand extension fit typically refers to the degree to which an extension is congruent with the brand’s other products, personality, or positioning (Mao and Krishnan 2006), and it can depend on factors unrelated to the qualities of the brand or the product itself (e.g., one’s culture; Monga and John 2007). In contrast, brand concept fluency relates to one’s understanding of the overall brand. Because the focus of the current work is on consumers’ attitudes toward the brand, we center our theorizing around brand concept fluency.

Why would FDP introduction timing impact brand concept fluency and attitudes? To answer this, we consider two specific brand-extension sequences. The first, early FDP introduction, occurs when the FDP is the first of the three brand extensions introduced. The second, late FDP introduction, occurs when the FDP is the last of the three brand extensions to be introduced. For instance, consider a brand that currently sells only sweatshirts, but plans to extend into the categories of exercise shorts, running shoes, and breakfast cereal (the FDP). An early FDP introduction strategy would introduce the breakfast cereal first, followed by shorts and shoes. In contrast, a late FDP introduction strategy would introduce cereal last.

Recall that mental categorization theories maintain that as newly-encountered exemplars—brand extensions in the current context—are encountered, the consumer integrates them into the brand concept and attempts to understand it. The ease with which this occurs is at least partly a function of the number and types of exemplars (i.e., the “cluster” of exemplars; Love et al. 2004) that constitute the preexisting brand concept.

Consider an early FDP introduction. When the breakfast cereal is introduced, it is noticeably distant from (i.e., does not fit with) sweatshirts, the lone exemplar in the brand concept at that point. Thus, the brand concept should be disfluent, which should engender a negative affective response that has a deleterious effect on brand attitudes. The subsequent closer brand extensions—shorts and shoes—more easily map onto (i.e., fit with) existing exemplars within the brand concept and, therefore, should be more easily integrated and brand concept fluency should increase. This, in turn, should lead to more positive brand attitudes.

Now consider a late FDP introduction. The first two brand extensions—shorts and shoes—fit with the exemplar(s) already comprising the brand concept and should increase brand fluency and attitudes. Then, the introduction of the cereal stands in stark contrast to (i.e., does not fit with) the three sports-apparel exemplars currently comprising the brand concept (as opposed to one in the early FDP introduction sequence). Hence, integrating cereal into the brand concept should be more difficult than is the case with early FDP introductions, leading to a more marked decrease in brand concept fluency, and a sharper reduction in brand attitudes. Thus:

  1. H1:

    The pre-FDP versus at-FDP brand attitude change will be significantly more negative for late (vs. early) FDP introductions.

Thus, one potential downside of introducing only closely related extensions is that this may trap or typecast the brand into a narrow set of future extension options. If H1 holds, this has implications for consumers’ brand attitudes after all of the extensions have been introduced—their final brand attitudes. To be clear, final brand attitudes are neither conceptually distinct from other brand attitudes held throughout the sequence of extension introductions, nor are they measured differently. Instead, this term merely refers to attitudes held at a critical point in time—the moment when all extensions have been introduced and the brand product portfolio consists of the exact same set of products regardless of when the FDP was introduced. With that in mind, the implications of H1 are:

  1. H2:

    Final brand attitudes will be significantly lower after the completion of brand extension sequences using late (vs. early) FDP introductions.

As detailed in the preceding arguments, the influence of all brand extensions including the FDP on brand attitudes is predicted to be driven by consumers’ brand concept fluency:

  1. H3:

    The impact of all brand extensions including the FDP on brand attitudes will be mediated by consumers’ brand concept fluency: Higher (lower) levels of brand concept fluency will result in the more positive (negative) brand attitudes.

It is critical to reinforce at this point that introducing an FDP is not inherently bad even if it might have an immediate adverse effect on consumers’ attitudes toward the brand. The initial adverse impact on brand attitudes may simply be something that needs to be accepted or endured in order to reap later benefits on other metrics such as revenue, profit, brand diversification, and market share. The purpose of the current work is to determine how the manager can best manage the negative aspects of FDP introductions via the timing of those introductions.

Summary of intended scope and contributions

To recap, the current work focuses on a heretofore unexamined research question: When should a brand introduce its first distant product? This context differs from previous research that has examined brands evolving via the introduction of extensions increasingly dissimilar to its core to reach one or more distant product categories. Thus, the findings that follow are informative for the manager who does not have either the resources to, or the interest in, making a series of gradual extensions in an effort to eventually reach a distant category.

Prior research certainly documents the advantages of slowly introducing increasingly dissimilar extensions to expand a brand. Keller and Aaker (1992) showed how a successful intervening extension increased evaluations of a proposed extension for average quality core brands. Swaminathan (2003) observed similar effects: When an intervening extension was successful among non–brand loyal customers, experience with the intervening extension increased the likelihood of trial of a subsequent extension. The findings from Jap (1993) indicate that consistent extensions lead to higher brand concept accuracy and accessibility. Similarly, Dawar and Anderson (1994) found that introducing extensions in an ordered, gradual manner (i.e., increasingly different from the initial brand concept) allowed for greater coherence and purchase likelihood for target distant extensions. In sum, to the best of our knowledge, no previous research on brand extensions has investigated what is examined in the current research—when FDPs should be introduced in the absence of intervening extensions.

Critically, our theorizing suggests that the difference between brands gradually approaching versus abruptly entering distant categories is not merely superficial. On the contrary, while the research on gradually approaching distant categories recommends a small-step approach to maximize brand attitudes (i.e., monotonically introducing extensions more dissimilar or distant from the original core products over time), our reasoning suggests that those brands abruptly entering distant categories may benefit from doing so earlier versus later. In fact, our predictions run directly counter to lay, managerial, and academic intuition that has likely been informed by previous research on gradual approaches. Our predictions have been derived from fundamental aspects of mental categorization and brand concept fluency. Thus, the current work is intended to shed light on a largely ignored managerial decision—when to introduce the brand’s FDP—and reveal counterintuitive insights predicted on the basis on established principles of cognitive psychology.

Empirical evidence

We present three studies that test H1 through H3. All three studies demonstrate that late (vs. early) FDP introductions have a significantly more negative impact on brand attitude change (H1) and that final brand attitudes are significantly lower after late (vs. early) FDP introductions. Study 2 additionally demonstrates that the impact of FDP introduction timing can be moderated by brand-level positioning, consistent with our mental categorization account. Study 3 is a longitudinal survey collected over the course of four days. In addition to demonstrating the influence of FPD introduction timing on brand attitudes via brand concept fluency, Study 3 finds that the effect on final brand attitudes has an impact on actual product preference and brand engagement.

Study 1: The impact of early (vs. late) FDP introductions on brand attitudes

Study 1 was designed to test H1 and H2. To briefly recap, we contend that the negative impact of FDPs on brand attitudes operates through consumers’ brand concept fluency, which is partly a function of the exemplars in the brand concept when each brand extension is introduced. Therefore, an early FDP introduction—when the brand concept contains only a single exemplar distinctly different from the FDP—should produce less disfluency than a late FDP introduction—when the brand concept contains three exemplars which are distinctly different from the FDP and similar to each other. Hence, the negative impact on brand attitude change (relative to pre-FDP attitudes) from FDP introductions should be greater for late (vs. early) introductions (H1), which should result in significantly lower final brand attitudes (H2).

Method

Two hundred thirty-three paid Amazon Mechanical Turk (hereafter, AMT) participants (103 females, 120 males, avg. age = 32.38) were randomly assigned to one of two between-subjects conditions: early vs. late FDP introduction. Participants were told that they would be learning about a brand (“Brand A”) that started business selling sweatshirts and subsequently released one new product each year over the next three years. The FDP was the second (fourth) product introduced in the early- (late-) FDP introduction condition. In all conditions, the brand’s final portfolio of products was sweatshirts, exercise shorts, running shoes, and breakfast cereal (see Fig. 1). Thus, after the third extension (fourth overall product) the set of products offered by the brand was identical across conditions.

Fig. 1
figure 1

Product categories and extension sequences

Full size image

The brand’s product portfolio was presented serially, adding one product at a time, and participants rated the brand on two scales after learning about each additional product: (1) “How would you evaluate Brand A overall?” (1 = poor, 9 = good), and (2) “How favorably do you view Brand A?” (1 = not favorably at all, 9 = very favorably). The two scales were strongly correlated (r = .93, p < .001) and, accordingly, combined into a single measure of brand attitude. Table 2 lists dependent measures collected across the three studies.

Table 2 Dependent measures collected
Full size table

Results

A mixed-analysis ANOVA revealed a significant interaction between the within-subjects, repeated brand attitude measures (core-product vs. 1st vs. 2nd vs. 3rd extension) and the between-subjects FDP introduction factor (early vs. late; F(3, 693) = 54.16, p < .001). As expected and shown in Fig. 2, brand attitudes were not significantly different across the FDP introduction timing conditions after the core (initial) product (i.e., sweatshirts) had been revealed to the participants (F < 1). Participants in the late-FDP introduction condition then held more positive brand attitudes after each of the first two extensions than did those in the early-FDP introduction condition (F(1, 231) = 29.23, p < .001 and F(1, 231) = 24.16, p < .001, respectively). Most critically, this pattern reversed after the third extensions were introduced, and final brand attitudes were significantly more positive in the early- (vs. late-) FDP introduction condition (F(1, 231) = 24.16, p < .001) despite the set of products offered by the brand being the same in both conditions at this point, consistent with H2. Means and standard deviations for the brand attitude measures for all studies are presented in Table 3.

Fig. 2
figure 2

Brand attitude evolution (study 1). * FDP = First distant product. The (breakfast) cereal is the FDP in the figure

Full size image
Table 3 Brand attitude results for studies 1–3 and Web Appendix studies
Full size table

The relative impact of FDP introduction timing (late vs. early) on brand attitude change (pre-FDP vs. at-FDP) was also as expected. Supporting H1, the late FDP introduction resulted in a significantly more negative brand attitude change (M pre-FDP = 6.04 vs. M at-FDP = 4.88; Δ = −1.16) than did the early FDP introduction (M pre-FDP = 5.29 vs. M at-FDP = 4.67; Δ = −.62; F(1, 231) = 9.83, p < .002).

Discussion

The results of Study 1 support the contention that first introducing more similar, close extensions creates a brand concept more tightly defined by multiple exemplars distinctly different from the FDP, which results in late FDP introductions leading to greater disfluency than early FDP introductions. However, it may be argued that these results are the byproduct of order or product-specific effects. Specifically, it might be the case the breakfast cereal is merely more positively responded to when introduced after sweatshirts (as in the early-FDP condition) versus after running shoes (as in the late-FDP condition). In other words, it might not matter when the breakfast cereal is introduced but, instead, after which other product it is introduced.

To examine if this was the case, a follow-up replication study was run (N = 177). The lone difference between Study 1 and the follow-up study was that the latter indicated that Brand A had introduced running shoes as its first product and sweatshirts as its final non-FDP product (i.e., either third or fourth in the product-introduction sequence, depending on the FDP introduction condition). Thus, breakfast cereal was introduced after running shoes in the early-FDP condition, and after sweatshirts in the late-FDP condition (the opposite of the main study).

The same pattern of results was found in the follow-up study as in Study 1 despite having reversed the introduction timing of the running shoes and sweatshirts (see Table 3 for the complete results). First, a mixed-analysis ANOVA revealed a significant interaction between the within-subjects, repeated brand attitude measures (core-product vs. 1st vs. 2nd vs. 3rd extension) and the between-subjects FDP introduction factor (early vs. late; F(3, 525) = 31.46, p < .001). Second, final brand attitudes were significantly more positive in the early- (vs. late-) FDP introduction condition (M early = 5.41 vs. M late = 4.54; F(1, 175) = 11.07, p < .001), supporting H2. Third, supporting H1, the late FDP introduction resulted in a significantly more negative brand attitude change (M pre-FDP = 5.77 vs. M at-FDP = 4.54; Δ = −1.23) than did the early FDP introduction (M pre-FDP = 5.46 vs. M at-FDP = 4.71; Δ = −.75; F(1, 175) = 4.59, p < .04). Thus, the results of the follow-up study suggest that Study 1’s results are not the mere byproduct of order or product-specific effects.

Study 2: The role of overall brand positioning

Study 1 showed that distant brand extensions can come at the cost of diminishing brand attitudes in the short run. Critically, those short-term costs were more severe for late FDP introductions, as our mental categorization–based theory predicts.

To this point, we have construed brand concepts only in terms of the products sold by the brand (i.e., its exemplars). Of course, brands are more than just their products and, consequently, brand concepts typically include information beyond the product exemplars mentally associated with that brand (Keller 2003). Here we examine how one type of non-product information—specifically, the brand’s positioning—can either accentuate or attenuate the exemplar-based effects documented in Study 1.

Marketers position or frame their brands in many ways. Pricing, slogans, logos, spokespeople, packaging, marketing channels, sponsorships, cross-branding, social media, and corporate social responsibility programs all influence how consumers perceive a brand and, consequently, what they expect from that brand. Of particular relevance here are tactics that influence the impact of brand extensions on consumers’ brand concept fluency.

A manager may position a brand narrowly around the brand’s core competencies (i.e., around its extant exemplars) or choose a broader positioning that is inclusive of a wider range of products or services. If broader positioning tactics are used, a wider range of products should be viewed as close to, or at least consistent with, the brand concept. Accordingly, a product that is difficult to integrate into the brand concept in the absence of such brand positioning (i.e., solely as a function of the products offered by the brand) can be more readily integrated into that concept if the brand is more broadly positioned. Thus, brand attitudes should be less negatively impacted by the introduction of an FDP when the brand is positioned broadly. Formally:

  1. H4:

    Relative to a control condition, pre-FDP versus at-FDP brand attitude changes should be significantly smaller for both late and early FDP introductions when the brand is positioned broadly.

In the absence of strong brand-positioning cues, the brand concept is largely defined by the products sold by the brand. Therefore, early FDP introductions are less detrimental to brand attitudes than are late FDP introductions for reasons already discussed (H1–H3). However, early FDP introductions should have a more detrimental impact on brand attitudes when the brand is narrowly positioned because the narrow positioning makes integrating the FDP into the brand concept even more difficult and, thus, makes the brand concept even more difficult to understand. Hence:

  1. H5:

    Relative to a control condition, pre-FDP versus at-FDP brand attitude changes should be significantly larger for both late and early FDP introductions when the brand is positioned narrowly.

Method

Five hundred ninety paid AMT participants (245 females, 345 males, avg. age = 34.19) were randomly assigned to one of six conditions in a 3 (positioning: broad vs. narrow vs. none/control) × 2 (FDP introduction: early vs. late) between-subjects design. FDP introduction timing was manipulated exactly as it was in Study 1 (see Fig. 1).

Brand positioning was manipulated via three cues: (1) slogan, (2) logo, and (3) retail outlet. In the broad condition, the cues were selected to encompass both health and wellness and thus both food and athletic apparel. In the narrow condition, the cues were selected to focus solely on athletic apparel. Specifically, in the broad condition, the slogan was “The Healthy Lifestyle People,” the logo was that pictured on the left in Fig. 3, and the exclusive retailer of the brand’s products was Target (Target’s logo was also displayed). In the narrow condition, the slogan was “The Athletic Apparel People,” the logo was that pictured on the right in Fig. 3, and the exclusive retailer of the brand’s products was Sports Authority (Sports Authority’s logo was also displayed). Participants in the control condition were not provided any of these cues.

Fig. 3
figure 3

Brand logos used in Study 2

Full size image

Participants reported their brand attitudes on the same two scales used in Study 1 after each product was introduced. In addition, after all the products had been introduced, participants responded to the following question on a 1 (negative) to 9 (positive) scale for each product: “How would you rate your overall attitude to this product sold by Brand A?”

Results

The two brand attitude items were again averaged into a single measure and submitted to a 3 (positioning: broad vs. narrow vs. control) × 2 (FDP introduction: early vs. late) × 4 (core-product vs. 1st vs. 2nd vs. 3rd extension) mixed-analysis ANOVA. Replicating Study 1, a significant interaction was found between the within-subjects, repeated brand attitude measures and the FDP introduction timing factor (F(3, 1752) = 102.90, p < .001). More important, this two-way interaction was qualified by a three-way interaction with the brand-positioning factor (broad vs. narrow vs. control; F(6, 1752) = 16.40, p < .001). These interactions show that (1) the way in which brand attitudes evolved varied based on when the FDP (breakfast cereal) was introduced and (2) the relationship between FDP introduction timing and brand attitude evolution depended on how the brand was positioned. The results are shown in Fig. 4, and the complete set of means and standard deviations is presented in Table 3.

Fig. 4
figure 4

Brand attitudes by brand positioning and First distant product (FDP) conditions (study 2). * FDP = first distant product. The (breakfast) cereal is the FDP in the figure

Full size image

Of critical importance, within the control condition, the late FDP introduction changed brand attitudes significantly more negatively (i.e., the difference between pre- and at-FDP brand attitudes) than did the early FDP introduction (Δlate = −1.11 vs. Δearly = −0.66, F(1, 196) = 6.88, p < .01), replicating the results of Study 1 and supporting H1.

Examining the patterns in Fig. 4, we find strong evidence for H4. Specifically, when the FDP was introduced early, the change in brand attitudes in the broad-positioning condition was significantly less negative (and actually slightly positive; M pre-FDP = 5.97 vs. M at-FDP = 6.15; Δ = .18) than in the control condition (M pre-FDP = 5.39 vs. M at-FDP = 4.73; Δ = −.66; F(1, 584) = 21.46, p < .001). Likewise, when the FDP was introduced late, the change in brand attitudes in the broad-positioning condition was significantly less negative (M pre-FDP = 6.46 vs. M at-FDP = 6.19; Δ = −.27) than in the control condition (M pre-FDP = 6.35 vs. M at-FDP = 5.24; Δ = −1.11; F(1, 584) = 22.47, p < .001). Thus, broadly positioning the brand mitigated the negative impact of the FDP on brand attitudes both when it was introduced early and late.

The support for H5 was mixed. On the one hand, when the FDP was introduced early, the change in brand attitudes in the narrow-positioning condition was significantly more negative (M pre-FDP = 5.98 vs. M at-FDP = 4.76; Δ = −1.31) than in the control condition (Δ = −.66; F(3, 584) = 13.71, p < .001), supporting H5. However, unexpectedly, there was no difference between the narrow-positioning (M pre-FDP = 5.98 vs. M at-FDP = 4.76; Δ = −1.31) and control (Δ = −1.11) conditions when the FDP was introduced late (F < 1). This suggests that introducing three similar products in the control condition influenced participants’ brand concepts in a manner and extent similar to the narrow-positioning manipulation (i.e., the narrowness of the brand was established by the introduction of three highly similar exemplars).

Participants’ final brand attitudes followed the pattern that would be expected given the impact of brand positioning on brand attitude change just discussed. Specifically, final brand attitudes were significantly less positive after a late (vs. early) FDP introduction in both the control (M late = 5.24 vs. M early = 5.68; F(1, 584) = 3.36, p = .067) and narrow-positioning conditions (M late = 5.20 vs. M early = 6.02; F(1, 584) = 11.65, p < .001). In contrast, FDP introduction timing did not significantly influence final brand attitudes when the brand was positioned broadly (M late = 6.19 vs. M early = 6.43; F(1, 584) = 1.08, p > .30).

Lastly, there were no significant differences between FDP introduction timing conditions in terms of participants’ attitudes toward Brand A’s sweatshirts, exercise shorts, or running shoes. However, there were main effects of FDP introduction timing (F(1, 584) = 25.35, p < .001) and brand positioning (F(2, 584) = 122.45, p < .001) on attitudes toward the FDP (breakfast cereal). These main effects were qualified by a significant interaction (F(2, 584) = 4.26, p < .02). The complete product-attitude results are presented in Table 4.

Table 4 Attitudes toward first distant product (FDP = Breakfast Cereal) by condition (Study 2)
Full size table

Within the control and narrow-positioning conditions, attitudes toward the FDP were significantly more positive after an early (vs. late) FDP introduction (F(1, 584) = 23.15, p < .001 and F(1, 584) = 10.18, p < .001, respectively). Conversely, FDP introduction timing did not influence attitudes toward the FDP in the broad-positioning condition (F < 1). Thus, the pattern of attitudes toward the FDP mirrors that of final brand attitudes, suggesting that final brand attitudes impacted participants’ feelings about the FDP, but not the other products offered by the brand in this study. We further examine product-level effects in Study 3.

Discussion

The results of Study 2, which were consistent with our mental categorization account, also have implications for managers: Flexibility in choosing when to introduce an FDP may be gained via positioning one’s brand more broadly. While our results do not consider other consequences of broad versus narrow brand positioning, they do highlight the opportunity to mitigate potential damage to brand attitudes when introducing products distant from the brand. Importantly, failure to position a brand (represented by the control condition) produced results similar to that of a narrow positioning. This suggests a broad brand positioning is needed to counteract the negative effects of a FDP on brand attitude. Additionally, there is some evidence that FDP introduction timing impacts responses toward the brand’s products as well, though this was limited to the FDP (breakfast cereal) in this study. The robustness of these product-level results is examined in Study 3.

Study 3: Examining the underlying process and behavioral responses

Studies 1 and 2 examined the evolution of brand attitudes as extensions were serially introduced. However, they did so using a single-session survey methodology. Thus, there was no significant delay between the introductions of the brand’s extensions to the participants. Additionally, although the brand attitude results have been consistent with our mental categorization and brand concept fluency account, this process has not been directly examined. Lastly, none of the dependent variables to this point have been incentive-compatible (i.e., behavioral)—they have all been attitude measures about hypothetical brands and their products. Study 3 was designed to address these weaknesses.

Method

Four hundred paid AMT participants were randomly assigned to either the early or late FDP conditions. The materials in this study were very similar to those in the preceding studies—the same product categories were used and the same brand attitude measures were collected—with the following important differences.

First, participants were told that the brand they would be evaluating was a real brand considering entering the U.S. market and which, therefore, was surveying U.S. consumers. They were also told that, because we were interested in their responses to specific aspects of the brand, the brand’s name, logo, and country of origin would be withheld: Instead, the brand would simply be referred to as “Brand A.” The brand was presented as being ostensibly real so that we could collect behavioral measures regarding the brand (described below). Participants were additionally told that the brand currently sold multiple products and that they would be shown these products in the order in which they had been introduced.

Second, instead of evaluating the brand repeatedly on one occasion, participants completed four surveys over the course of four days. Administering multiple surveys over multiple days to the same Amazon Mechanical Turk participants was accomplished using the www.turkprime.com panel-study service. The first survey introduced the core product—sweatshirts—to all participants. The second survey, completed on the second day, reminded the participants of the brand’s first product and then introduced its second product. The third survey, completed on the third day, reminded the participants of the brand’s first and second products and then introduced its third product. Finally, on the fourth day, participants were reminded of the brand’s first three products and the final product was introduced. The order in which the specific products (sweatshirts, shorts, shoes, and cereal) were introduced was determined by the condition to which the participant was randomly assigned (see Fig. 1).

Third, in addition to reporting their brand attitudes after learning of each new product, participants also responded to four questions measuring their brand concept fluency (Table 2) each day. While these brand concept fluency items were not taken from previous research, they were developed to be consistent with the principles of processing fluency, which tend to be very consistent across sources of (dis)fluency (Alter and Oppenheimer 2009). Specifically, if a brand concept is well- or easily-understood, then the consumer should not only be able to say they understand the brand concept but also feel confident about what the brand is likely to do in the future.

As a concrete example, a consumer who has a fluent brand concept for Chipotle should not only have a clear understanding of what the brand currently is (a fast-casual Mexican food restaurant chain), but also what it is likely to be or do in the future (e.g., introduce additional Mexican foods on its menu). Hence, our brand concept fluency measures explicitly included measures pertaining to participants’ expectations for the brand’s future actions. Collecting responses on these measures allowed us to test whether shifts in brand concept fluency drive (mediate) the link between brand extensions and brand attitudes.

Fourth, once all of the extensions had been introduced and evaluated, participants rated the perceived quality of each of the brand’s products (1 = low quality, 9 = high quality), as opposed to indicating their attitudes toward the products (as was the case in Study 2). We chose to measure perceived quality of the products—as opposed to attitudes toward those products—because we would subsequently be collecting incentive-compatible brand–product preference measures (described below) and quality-perception measures could provide converging (as opposed to redundant) evidence of the consequences of FDP introduction timing.

Fifth and finally, we collected incentive-compatible (behavioral) measures for product preferences and brand engagement. First, regarding preferences, participants were told they would receive a “thank-you payment” for participating in the four surveys and that they could indicate their preference between a small cash payment of $3 and each of the four Brand A products: (1) one Brand A sweatshirt, (2) one pair of Brand A exercise shorts, (3) one pair of Brand A running shoes, or (5) five boxes of Brand A breakfast cereal. Participants were informed that one of these four choices would be randomly selected and they would be awarded their preferred option (product or cash). Thus, participants ostensibly needed to sacrifice a cash payment to receive a Brand A product, rendering this measure incentive-compatible. In reality, since Brand A was not a real brand, all participants were awarded $3 at the end of the survey.

Participants were then told that upon launching in the U.S., Brand A wanted to reach out to potential customers via email. They were then asked to indicate which, if any, of the following communications they would be willing to receive (a proxy for brand engagement): (1) daily emails, (2) weekly emails, (3) monthly emails, (4) semi-annual emails, or (5) they would not be willing to receive any emails.

After responding to all measures on day four, participants were (1) informed that the brand was not real, (2) debriefed as to why the brand had been presented as real, (3) informed that they would be paid the additional $3 thank-you payment for participating, (4) assured that they would not be contacted via email due to their participation in these surveys, and (5) thanked for their participation.

Results: Brand attitudes

Unsurprisingly, there was some attrition over the four days, and the ultimate number of participants that completed all four surveys was 295 (74%; 167 females, 128 males, avg. age = 36.99). Critically, there was no differential attrition across conditions, as evidenced by 146 early FDP introduction participants and 149 late FDP introduction participants completing all four surveys. We limited our analysis to those participants who completed all four surveys.

The two brand attitude items were again averaged into a single measure and submitted to a 2 (between-subjects FDP introduction factor: early vs. late) × 4 (within-subjects, repeated attitude measure: core-product vs. 1st vs. 2nd vs. 3rd extension) mixed-analysis ANOVA. Replicating the previous studies, there was a significant interaction between the within-subjects, repeated brand attitude measures and the FDP introduction timing factor (F(3, 879) = 29.56, p < .001; see Fig. 5 and Table 3). As before, brand attitudes were not significantly different across the FDP introduction timing conditions after the core (initial) product (i.e., sweatshirts) had been revealed to the participants (F(1, 293) = 1.23, p > .26). Somewhat unexpectedly, there remained no difference in brand attitudes after the first brand extensions had been introduced (F < 1).

Fig. 5
figure 5

Brand attitude evolution (study 3). * FDP = first distant product. The (breakfast) cereal is the FDP in the figure

Full size image

Consistent with our predictions, however, brand attitudes were significantly higher in the late- (vs. early-) FDP introduction condition after the second brand extensions had been introduced (F(1, 293) = 16.65, p < .001). Most critically, this pattern reversed after the third extensions were introduced, at which point final brand attitudes were significantly more positive in the early- (vs. late-) FDP introduction condition (F(1, 293) = 14.53, p < .001) despite the set of products offered by the brand being the same in both conditions at this point, consistent with H2.

Of critical importance, the late FDP introduction resulted in a significantly more negative brand attitude change (M pre-FDP = 6.43 vs. M at-FDP = 5.62; Δ = −.81) than did the early FDP introduction (M pre-FDP = 6.04 vs. M at-FDP = 5.85; Δ = −.19; F(1, 293) = 16.37, p < .001).

Results: Brand concept fluency mediation

After we reverse coded items 2 and 4 (Table 2), the four brand concept fluency measures were highly correlated across the four days of responses (minimum α = .842). Thus, these items were combined into a single brand concept fluency measure and submitted to the same 2 (FDP introduction: early vs. late) × 4 (post core-product vs. 1st vs. 2nd vs. 3rd extension) mixed-analysis ANOVA.

The same general pattern of results found for brand attitudes (Fig. 5) was found for brand concept fluency (Fig. 6), and the interaction between the within-subjects, repeated brand attitude measures and the FDP introduction timing was significant (F(3, 879) = 288.74, p < .001), as expected. However, the critical question was whether participants’ brand concept fluency mediated the relationship between extension introductions and brand attitudes.

Fig. 6
figure 6

Brand concept fluency evolution (Study 3)

Full size image

To determine whether the interaction between FDP introduction timing (early vs. late) and the current stage in the extension sequence (post core-product vs. 1st vs. 2nd vs. 3rd extension) on brand attitudes was mediated by participants’ stage-specific brand concept fluency, we employed Hayes’s (2013) SPSS PROCESS Macro (model 8). To be clear, both brand concept fluency and brand attitudes were measured after each product was introduced. Thus, we are able to determine if the evolution of brand concept fluency (over the four product introductions) drives the evolution of brand attitudes as each product is introduced. For instance, the product introduced by the brand at the third stage in the sequence (i.e., the second “extension” after the core product) was shorts (shoes) in the early- (late-) FDP introduction condition. Attitudes and brand concept fluency differed between conditions at this point. The question then is, can the overall pattern of brand attitudes across FDP introduction timing conditions be predicted by the overall pattern of brand concept fluency? If so, this would indicate that brand concept fluency was a primary driver of brand attitudes.

As predicted, the indirect path from the focal interaction, through participants’ brand concept fluency, to their brand attitudes was significant (as indicated by the 95% confidence interval not including zero; LLCI = .1253, ULCI = .2382). Thus, brand concept fluency mediated the effect of specific brand extensions on brand attitudes, as our mental categorization account predicted (H3).

Results: Perceived product quality

After participants had been presented with all four of the brand’s products, they rated the perceived quality of each of them. In Study 2 attitudes toward the brand’s products varied as a function of FDP introduction timing only in the case of the FDP itself. In contrast, in Study 3, participants’ product-level quality perceptions were significantly influenced for all of the brand’s products: Late (vs. early) FDP introductions resulted in lower perceived quality for the brand’s sweatshirts (M late = 5.64 vs. M early = 6.05, F(1, 293) = 6.18, p < .02), exercise shorts (M late = 5.64 vs. M early = 5.96, F(1, 293) = 4.16, p < .05), running shoes (M late = 5.70 vs. M early = 6.05, F(1, 293) = 4.27, p < .04), and breakfast cereal (M late = 5.40 vs. M early = 5.78, F(1, 293) = 4.38, p < .04). These results are directionally consistent with participants’ final brand attitudes.

Results: Product preference and brand engagement

Participants’ final tasks were to make ostensibly real choices (1) between potentially receiving Brand A products versus cash and (2) regarding their willingness to be contacted, via email, by Brand A.

We first analyzed participants’ product preferences by summing the number of products each participant preferred over a cash payment. The resulting means were not significantly related to the manipulated FDP introduction timing conditions (F < 1). However, following Zhao et al. (2010), we tested if there was indirect-only mediation (i.e., mediation in which there is a significant indirect effect, but no direct effect) from the manipulated FDP introduction timing, through participants’ final brand attitudes, to their ultimate willingness to choose Brand A products over cash. Hayes’s (2013) SPSS Macro (model 4), revealed a significant indirect path from FDP introduction timing to participants’ preferences via their final brand attitudes (95% confidence interval; LLCI = .0134, ULCI = .1385). Thus, indirect-only mediation was found for participants’ incentive-compatible product preference choices.

Participants’ willingness to be contacted by the brand via email (a proxy for brand engagement) was analyzed by dummy coding participants’ responses (1 = they were willing to receive emails at some frequency, 0 = they were not). Like participants’ product preferences, their willingness to be contacted by the brand was not significantly predicted by the manipulated FDP introduction timing factor (χ2 < 1). However, once again, the indirect path from FDP introduction timing to participants’ willingness to be contacted via their final brand attitudes was significant (95% confidence interval; LLCI = .1086, ULCI = .4808). Thus, as with product preferences, indirect-only mediation was supported for participants’ willingness to be contacted by the brand; that is, their brand engagement.

Discussion

Study 3 helps answer several lingering questions. First, the effect of FDP introduction timing on brand attitudes found in Studies 1 and 2 was replicated using a design with longer intervals between the participants’ exposure to the brand’s various products. Second, direct measures of brand concept fluency were shown to mediate the relationship between brand product introductions (extensions) and brand attitudes, as our mental categorization account predicted. Third, FDP introduction timing was found to impact product-level quality perceptions for all products offered by the brand. Fourth and finally, FDP introduction timing significantly, albeit indirectly, impacted participants’ product preferences and brand engagement. Collectively these results both replicate the results of the previous studies and bolster the substantive implications of our findings in general.

General discussion

Summary and implications

As marketers attempt to leverage their brands as engines for organic growth by extending into diverse product categories, brand product portfolio management issues become critical. In particular, marketers need guidance as to the effects on their brands of introducing extensions, which vary in similarity, at different times. With this in mind, three studies have shown that the brand’s decision of when to introduce its first distant product (FDP) in a series of planned brand extensions can impact several managerially relevant dependent variables. Specifically—and counter to lay, managerial, and some theory-based intuitions—early (vs. late) FDP introductions tend to result in more positive final brand attitudes, stronger preferences for the brand’s products, and a greater willingness to engage with the brand in the long run, even though the final set of products offered by the brand is the same regardless of when the FDP is introduced.

Theoretically, these results add to our understanding of how consumers think of (mentally represent) brands. Moreover, they shed light onto the evolution of brand concepts over time via brand extensions. While much research has examined the process of gradually extending brands to distant categories, the current research has examined the theoretical processes underlying consumers’ responses to distant extensions that are not approached gradually. Moreover, as opposed to examining brand extensions as one-off instances, the current studies examined brand attitude evolution as the product portfolio for the brand expanded. While examining and understanding the impact of individual brand extensions is important, the current work highlights the value of taking a broader, more strategic perspective on brands and their product portfolios.

Of equal importance is the insight that these results provide managers. Currently, brands wishing to enter distant categories without gradually introducing products less and less similar to their core products have little to no guidance on the matter. Indeed, much research suggests that companies (i) should not introduce distant extensions and, if they choose to do so, (ii) should delay the introduction of distant extensions as long as possible. A pretest revealed that managers shared this same intuition regarding when to introduce distant extensions.

When to introduce distant extensions

Building on established psychological theory, our results demonstrate that the opposite can often be true in terms of consumers’ brand attitudes: Distant brand extensions should be introduced earlier rather than later when marketers are unable or unwilling to introduce multiple intervening extensions to reach the distant category. Importantly, Study 2 demonstrated that brand-level positioning can afford marketers a degree of flexibility regarding the introduction timing of the brand’s FDP, consistent with our theoretical framework of mental categorization (several other substantive and theoretical boundary conditions are examined in three additional studies presented in the Web Appendix). Thus, in terms of managing brand attitudes (and plausibly a wider array of behaviors toward the brand, as found in Study 3), managers now have theory-based guidance regarding when FDPs should be introduced that stands in stark contrast with their likely intuitions on the matter.

Distant extensions are not inherently “bad” extensions

While distant extensions often entail greater risk than proximal extensions, they can also offer greater potential gains: Short-term losses in terms of brand attitudes can be offset by the long-term benefits on other metrics such as revenue, profit, brand diversification, sustained relevance, and market share. This perspective on distant extensions is not derived merely from wishful hypothetical reasoning. Numerous brands have introduced distant extensions (Table 1), and academics have argued that brands can sustain relevance by being innovative and extending the brand beyond its traditional identity (Aaker 2012; Beverland et al. 2015). Some have failed and some have succeeded, which speaks to the inherent risk-reward nature of such brand extensions. For the purposes of the current investigation, we have assumed that the brand has made a measured decision to introduce an FDP among a series of planned extensions (although we do not presume that all brands are forward looking despite the advantages of being so) and focused on mitigating the risks associated with such brand extensions.

Limitations and directions for future research

Of course, there are limitations to our studies. First, they were controlled experiments which did not allow for actual experience with the brand or its extensions. Perhaps these results will persuade a manager to perform a real-world test of our hypotheses. Currently, researchers’ and managers’ intuitions are that it is better to introduce close before distant extensions. At a minimum, our results should lead managers to question conventional wisdom when deciding the order in which to introduce brand extensions.

Another limitation of the current research is that it primarily examined contexts where the majority of the brand’s products are centered in one product category (i.e., “close” products) with a minority in another, distinct category (i.e., “distant” products). Future research should examine the evolution of more “balanced” brands (initial results found in Study A3, in the Web Appendix suggest that the effect, although weakened, still persists). Likewise, the FDP throughout these studies came from a product category objectively distinct from the brand’s core. While this is the most straightforward, and probably externally valid, manner of establishing the FDP’s distance from the brand’s core, future research might instead examine FDPs that are of the same basic category (i.e., hold the category constant), but whose perceived distance is manipulated via exogenous cues of the product’s innovativeness or reliability (somewhat akin to Heath et al.’s 2016 work).

Future work can go in several other directions as well. It may be fruitful to examine the process more closely to better understand how consumers update or evolve their brand concepts after extensions have been introduced. Considering early FDP introductions, for instance, one might examine whether the FDP is truly integrated into the brand concept—which would extend the brand concept—or, alternatively, if the post-FDP extensions (which are closer to the original brand concept) allow the consumer to effectively “write-off” the FDP as an anomaly. Although the brand positioning results of Study 2 is more supportive of the former interpretation, more detailed process evidence could be revealing. Should it be the case that introducing an FDP broadens the brand concept, this should affect the response to future introductions of distant products. Ascertaining the optimal level of breadth (as driven by brand extensions or positioning) is an interesting topic to pursue, though one outside the scope of the current work.

Additional “incentive compatible” experiments can also be run to test the robustness of the current results. Even better, market tests could be conducted. Short of that, a meta-analytic review of past extensions could uncover significant patterns of results, although assembling the data to do so would be difficult. It would also be difficult to control for the many confounds which would be present. If different strategies can be identified and stock market efficiency assumed, then financial value could be modeled as a function of extension strategy.

More broadly, the effect of order may decay over time as new cohorts are exposed to the brand. (Interestingly, later cohorts are likely to acquire information about the brand’s past extensions in a concurrent manner, which Study A1, presented in the Web Appendix, shows will reduce the impact of FDP introduction timing.) Examining the pattern across cohorts is interesting both theoretically and practically, as would be considering the introduction of additional, subsequent distant products.

Finally, it might be the case that FDPs, given the right set of circumstances and factors, bolster brand attitudes. While our theoretical framework is mute on this topic, one could envision scenarios in which a “boring” brand is revitalized by the introduction of a distant product. A more mundane instance might relate to a brand abandoning old technologies for new. Hopefully this paper will stimulate additional behavioral and quantitative work (both analytical and empirical) in this area.