Talk Invitation
Department of Information Management
Copy Alert: A Method and Metric to
Detect Visual Copycat Brands
Michel Wedel
PepsiCo Chair in Consumer Science, University
of Maryland, USA
Friday, March
th
7 ,
9:30 - 10:30am
E.Sun Hall, Building I, College of Management
The authors propose a method and metric to quantify consumer
confusion between leading brands and copycat brands due to the
visual similarity of their packaging designs. The method has three
components. First, image-processing techniques establish the
objective similarity of the packages of leading and copycat brands
based on their colors and textures. Second, a perceptual decision
task (triangle test) assesses the accuracy and speed with which
consumers can identify differences between brands from rapidly
(300 msec.) flashed images of their packages. Third, a competing
accumulator model describes the buildup of evidence on each of
the alternative brands during consumers' perceptual decisions,
and predicts the accuracy and response time of brand
identification. Jointly, this establishes the impact that the visual
features of copycat packaging has on consumer confusion. The
method is applied in a test of experimentally-designed and market
copycats in fifteen product categories. A three-tiered metric (CopyAlert, Copy-Warning, Copy-Safe) establishes to what extent
copycat brands imitate the package designs of target brands, and
which visual features are responsible for this.
Talk Invitation
Department of Information Management
Usage Experience with Decision Aids
and Evolution of Online Purchase
Behavior
Jie Zhang
The Harvey Sanders Fellow of Retail Management,
University of Maryland, USA
Friday, March
th
7 ,
11:00 - 12:00pm
E.Sun Hall, Building I, College of Management
This study investigates how usage experience with various decision
aids available in an online store contributes to purchase behavior
evolution in a new Internet shopping environment. In the context
of online grocery stores, we categorize four types of decision aids:
those for 1) nutritional needs, 2) brand preference, 3) economic
needs, and 4) personalized shopping lists, and construct a Nonhomogeneous Hidden Markov Model of store visit incidence and
shopping trip spending. The same model is also applied to study
purchase incidence and quantity decisions in eight product
categories. We find that consumers evolve through distinct states
of purchase behaviors and exhibit stronger tendency to use
habitual decision heuristics over time. While their average levels of
price and promotion sensitivities increase first and then decrease,
individual consumers show divergent patterns. Moreover,
consumers’ usage experience with decision aids contributes to
their behavior evolution, and the effects differ by the specific
decision aids and behavioral state. In general, the impact of usage
experience with decision aids on purchase behavior evolution
appears to be stronger for higher-purchase-frequency and nonfood categories. We demonstrate that targeted promotion
activities based on our models can improve store-level and
category-specific sales for an online retailer.