Consumers rank variety of assortment right behind location and price when n
aming reasons why they patronize their favorite stores. Consumers care abou
t variety because they are more likely to find what they want when going to
a store that offers more varied assortments. When tastes are not well form
ed or are dynamic, perceived variety matters even more because of the desir
e to become educated about what is available while maintaining flexibility.
Variety perception also matters when the variety-seeking motive operates.
Retailers care about variety because customers value variety. Therefore, it
is important to understand how people perceive the variety contained in an
assortment and how these perceptions influence satisfaction and store choi
ce. Remarkably, except for a recent study by Broniarczyk et al. (1998), the
re has been no research aimed at understanding the variety perception proce
ss itself.
We offer a general mathematical model of variety based on the complete info
rmation structure of an assortment, defined both by the multiattribute stru
cture of the objects and their spatial locations. We impose a psychological
ly plausible set of restrictions on the general model and obtain a class of
simpler estimable models of perceived variety. We utilize the model to dev
elop assortments that vary widely in terms of their information structure a
nd study the influence of three factors on variety perceptions: (a) informa
tion structure of each assortment (i.e., the attribute level differences be
tween objects); (b) level of organization of the objects and hence their re
lative spatial positions; and (c) task orientations, promoting either analy
tic or holistic processing. We also investigate the influence of variety pe
rception and organization on stated satisfaction and store choice. To summa
rize our major findings:
1. Information structure has a big impact on variety perceptions, though di
minishing returns accompany increases in the number of attributes on which
object pairs differ.
2. People are more influenced by local information structure (adjacent obje
cts) than nonlocal information structure. Proximity matters.
3. Organization of the display can either increase or decrease variety perc
eptions. When people engage in analytic processing, organized displays appe
ar to offer more variety. When processing is holistic, random displays are
seen as more varied.
4. Both variety perceptions and organization drive stated satisfaction and
store choice. People are more satisfied with and likely to choose stores ca
rrying those assortments that are perceived as offering high variety and th
at are displayed in an organized rather than random manner.
Our work provides a basic framework for thinking about variety. By helping
retailers to understand the factors that drive variety perception, it may b
e possible to design more efficient, lower cost assortments without reducin
g variety perceptions and the probability of future store visits.