Home   |   Login   |   Register      

 

    

 

The Relationship Between Distribution and Market Share

Paul W. Farris, James Olver, and Cornelis de Kluyver, 1988 [88-103]

View/Order Item >>
Comments from members >>

Subject
The linkage between the distribution a product receives and the market share it achieves.

Type of Report
Technical presentation of model; formulation of propositions; test against empirical data

Objective
To develop a model of the distribution-market share relationship that explicitly accounts for: 1) the effects of compromised and uncompromised consumer demand, and 2) the effects of manufacturer-trade support on stocking levels.

Method
Identifies relevant areas of the literature and conceptualizes a model wherein consumer "pull" and manufacturer "push" interact to affect market share; defines the constructs used in the conceptualization; formalizes the model and sets forward two propositions of trade behavior under specific assumptions; tests the model with empirical data; examines the implications of the postulated between distribution and market share; suggests opportunities for further research.

Audience
This technical report is directed toward modelers and others with considerable quantitative proficiency.

Principal Findings
The model developed in this paper relates distribution and market share. In doing so, it uses very specific definitions for consumer preference, resistance to compromise, distribution, and in-store attractiveness. These terms are further related in a conceptual "push-pull" model that treats "push" and "pull" as effects of marketing mix decisions rather than as traditional classifications of marketing strategies. Understanding these differences in definition are essential to appreciating the findings of the research.

In general terms, distribution and market share, as treated by the authors, are related in a nonlinear fashion. As higher levels of distribution coverage are reached, increases in distribution depth in the form of additional stock-keeping units, a larger share of shelf, or special displays/positioning can significantly affect market share. As market share increases, distribution breadth increases more-than proportionately, reflecting the trade's tendency to stock those brands which sell well--and also the tendency to give those brands a larger share of shelf space and to stock more stock-keeping units. Conversely, as distribution is increased, the greater availability of the product, combined with the trade support for it, results in larger-than-proportional increases in market share.

Unavailability of a smaller brand, either because of stockouts or restricted distribution, benefits larger brands more than smaller ones. This result is implicit in the stocking behavior of retailers: Given the choice between a large brand and a small brand (and given equal levels of trade support), retailers will preferentially choose the brand for which there is greater demand--and consequently the one with the larger market share.

Increases in consumers' willingness to search for a brand --and the trade support for that brand--combine synergistically to affect brand availability and market share. Thus, consumer franchise-building activity pays off in two ways: First, such actions serve to enhance consumers' willingness to search for the brand and raise demand for the brand. Second, this increase in "pull" allows increases in distribution intensity because of retailers' desires to stock brands for which there is greater consumer preference. Increases in trade support act synergistically by encouraging retailers to provide a larger share of shelf, stock more stock-keeping units, or otherwise merchandise the brand more intensively.

Implications

Because retailers base their stocking decisions on potential sales, changes in consumer preference for a brand exhibit multiplier effects. In other words, there are direct, immediate effects of preference on sales, which are followed by changes in distribution depth and breadth. These distribution changes also cause changes in sales and tend to amplify (multiply) the "pure" effects of preference.

Additionally, conventional ways of measuring distribution coverage, such as percentage of All Commodity Volume (ACV) may be inappropriate at higher levels of distribution. Issues such as consumers' preference, in-store merchandising effects, and relative share-of-shelf or share-of-stockkeeping units should explicitly be considered to fully capture the synergistic effect of "push" and "pull" on distribution and market share.

From management's point of view, the more useful measure may be incremental change in Percentage of Category Volume (PCV), particularly when: 1) the average PCV in the category is low, 2) consumers have a significant preference for the brand, and 3) consumer resistance to compromise is low.

Researcher Comments
The researchers provide the following context for considering their model:

"Our model is considerably more complex than many experimental designs and regression models that have been used to relate distribution and market share. The greater complexity was motivated by our observation that it is necessary to consider distribution as a consequence of the interaction between 'push' and 'pull' before meaningful conclusions can be drawn about how distribution affects market share. This view highlights the role of distribution in creating product availability from a consumer perspective, as well as the role of consumer preference in stimulating retail stocking behavior.

"In practice, the activities and effects modeled in the 'push-pull' paradigm do not occur simultaneously, nor would one expect them to have similar carryover effects. Additionally, short-term influences on distribution intensity, such as price promotions, were not explicitly investigated. Finally, a negative influence of market share growth on distribution in the short term is not uncommon, as when excess demand results in a higher-than-usual level of stockouts at the retail level.

"These limitations notwithstanding, the most important test of the model is whether it yields new insights into the basic forces that link distribution and market share. The results presented here clearly demonstrate that simple linear or multiplicative representations are naive and often misleading. The evidence suggests that meaningful models of distribution coverage should explicitly account for both consumer and trade behavior and should use more refined measurement concepts than those currently used."

About the Authors
Paul Farris, James Olver, and Cornelis de Kluyver are, respectively, Professor, Assistant Professor, and Professor of Business Administration at the Colgate Darden Graduate School of Business Administration at the University of Virginia

---

Comments from members

Please login to view and/or submit comments

---

View/Order Item

Online version
Your Price: not available
Hard copy
List Price: $5.00 Your Price: $5.00

Please login for full access and discounted pricing.