Feeds:
Posts
Comments

Posts Tagged ‘Preference’

Measuring the value or equity of a retailer name from a customer’s point-of-view is usually challenging because of the diversity of products from various brands retailers offer on display and additional dimensions of performance that are specific to the retail store environment. For a long time retailers are not merely distributors that bring forward products to the consumers but offer products in their own names; furthermore, experiences of shoppers on-site of the store have a stronger influence on their purchase decisions. Thus, assigning customer values to retail names is often not a simple matter.

A similar problem with respect to the diversity of products may arise when trying to measure the value of a manufacturer corporate name to consumers, if the manufacturer uses its corporate name as an umbrella or even as a higher-level endorser for a wide range of products of different types. In this condition consumers may become confused as to which type of product they should refer and find it difficult to generalise their value judgements too broadly. Particularly, when trying to translate the subjective value to monetary terms, as often exercised with conjoint models of preference, one cannot plainly specify a price range that will be relevant to various types of products (e.g., TVs, stereo systems, washing machines, etc.) because the over-arching corporate name is too abstract. One has to conduct an evaluative study for each product category separately to obtain valid and relevant evaluations. The evaluation problem becomes several times more complex for a retail chain by accounting for the internal competition between manufacturers’ brands and a retailer’s own brand, and the other facets of the shopper experience in store (e.g., design and atmosphere, convenience, service).

  • As a case in example consider the branch of fashion retail. Castro, a leading Israeli homegrown fashion retail chain, has expanded greatly over the past 15 years (operating around 100 stores) and is a well-known and favourable name in many homes. According to a survey by Israeli business daily paper TheMarker with market research firm “Meida Shivuki” (27 Jan. 2012 (1)), Castro is the most familiar fashion retailer in Israel, remaining stable in this position six years in a row. Nearly half of adult Israelis have purchased an item or two in one of Castro’s stores in the past year. Fox, a low-cost retailer, is second, and Zara, a global Spanish-originated retailer, is in the third place (fourth year in a row). However, newly coming international retailers like H&M and Gap are tailing Castro, and more international brands like American Eagle or Banana Republic are expected to arrive soon. H&M climbed in awareness from 18.5% to 27% in six months, while Castro withdrew a little from 68% to 64%. Competition in the local fashion arena is becoming fierce, maintains TheMarker. In this setting, we may ask how well the value of Castro, from a consumer perspective, fares against rival international retail brands. One may also question what is the “attraction power” of Castro in terms of willingness-to-pay, and does it have to drag itself into a price war with the rivals to win?

In a conjoint analysis or choice study, brand is usually defined as one of the attributes describing a product, with several different brand names suggested as options (e.g., a choice set with four alternative products, each from another brand). This approach provides a single-numeric measure of value for each brand that some criticize as of too limited scope. Hence, further analyses on the subjective brand-equity values are advised, such as translating them also to monetary values of brand premiums by accounting also for consumers’ price sensitivity.

Moreover, in order to learn about the sources of brand values, we can analyse variation in brand values at the individual level vis-a-vis brand perceptions on several relevant dimensions of brand image (e.g., performance, reliability, or courteous service). Several techniques allow that, including with discrete choice modelling. It is worth mentioning in that context an unusual approach suggested more than 20 years ago of a brand-anchored model for evaluating the images of retailer brands (2). In that conjoint model, rather than including retailer names as options in a single brand attribute, retailer names are represented as options in each of several retail image attributes: For example, convenience of shopping is like at store of retailer “A”, “B”, or “C”.  This model does not offer overall values for each retailer, but it does suggest the relative values of a retailer name on each dimension of image. It’s like combining two-stage analyses proposed at the top of this paragraph in a single analysis. One conspicuous weakness of this approach is that respondents who do not know what would be the level of performance of an existing retailer on any of the dimensions will face difficulty in making reliable judgements of the retailer “portfolios” suggested to them or make a choice between them (the researchers have shown that consumers are likely to recall better the brands that score higher).

In the remaining of this post-article I suggest three alternative approaches for evaluating customer-based brand equity of retail chains in the framework of conjoint models, the first two apply a monetary currency whereas the third proposes distance as the currency of cost.

A Common Set of Products (“basket”) — In this approach we present to respondents-shoppers a well-defined set of products and ask them to suppose that they are going to shop for this set in stores of several optional retailers. Since retailers frequently offer on display a large variety of products, this set should serve as a common reference for comparison with regard to price levels. In some domains, such as food and grocery, we may be able to construct a “basket” of particular product items, including specific brands, because most stores hold the same product brands. Yet, in other domains like fashion this task could be more daunting because retailers choose to offer more differentiated clothing designs and specialise in bringing clothing items from different designer names. In the case of fashion we may have to describe in more general terms an outfit composed of several items but be specific enough about the quality and style of the items (e.g., think of dressing a mannequin with an outfit).

Applying this approach, therefore, is more domain-contingent. Our aim is to estimate the price premium that shoppers are willing to pay in order to purchase a set of products from a particular retailer. However, there is greater risk in this approach of confounding the value of a retailer with the values of products included in the set of reference.

Retailers’ Qwn Brands — Many retail chains in various domains offer products in selected categories carrying their own retail name as brand or their unique private labels available only at their chain-stores. Emphasizing the retailer’s own brand of products helps to better focus attention on the retailer on all aspects of shopping from its chain-stores. It may be seen as a special case of the first approach, only that here respondents-shoppers are advised that all product items included in the set are carrying the retailer’s name or private label. Thus, the differences in quality between retailers with respect to their own branded products can be taken into consideration by the respondents-shoppers.

This method represents a more round-up approach for assessing the monetary premium shoppers are willing to pay when buying at a particular retail chain on ground of both products identified with the retailer and the experience of shopping at its stores. Yet, it is applicable only if all retailers proposed have salient brands of their own for comparison.

Distance from a Retailer’s Nearest Store — Taking on a different perspective, this approach breaks with the common use of monetary price as the currency of cost. As implied in the first two methods described above, the monetary currency may introduce quite difficult complications in the context of retailer evaluation. Nonetheless, there are types of cost consumers are likely to incur while making purchase decisions such as time and psychic effort or stress. Particularly in the context of retail, Sorensen relates to time and angst in addition to money as the three currencies of cost shoppers may incur while looking for products they require or desire in-store (3). However, even before entering the store, another type of cost may be the distance the shopper has to make to reach his or her favourite store. Distance is often suggested also as a measure of loyalty: How far are you willing to go in order to find your favourite brand or to shop in a store of your favourite retailer?

According to this approach, a “cost” attribute would inform respondents-shoppers, for instance that “the nearest store of Retailer A is 500 meters away from you”. This type of conjoint application measures the retailer’s brand premium in terms of extra distance shoppers are willing to go to reach one of its stores (relative to a competing retail chain). It is possible that some consumers would want to go further actually to find lower prices or better value, but that perception could also be engrained in the retailer’s image. Indeed, the conjoint model alone may not tell us whether a retailer’s brand is preferred due to price/value perceptions or shopping experience aspects. On the other hand, it provides a measure of loyalty that may fit more smoothly in the context of choosing a retailer and poses no pre-conditions on the specific products each consumer may wish to buy at the store.

Each of these three approaches to measuring customer-based brand equity of retailers may be more appropriate, sensible, and easier to apply in some domains rather than others. The third approach appears for example the more suitable in the domain of fashion. However, if pricing issues arise, the  first or second approaches may be more practical albeit with some greater difficulty. That is where experience and good judgement of managers and researches comes in.

Ron Ventura, Ph.D. (Marketing)

Notes:

(1) Hebrew readers may find the original article of TheMarker at http://www.themarker.com/consumer/1.1627373.

(2) “Reliability and Validity of the Brand-Anchored Conjoint for Measuring Retailer Images, Jordan J. Louviere and Richard D. Johnson, 1990, Journal of Retailing, 66 (4). pp. 359-382.

(3) “Inside the Mind of the Shopper (The Science of Retailing)”, Herb Sorensen, 2009, Pearson Education.

Read Full Post »