סיור לימודי של אהבה
סמינר בשיווק
Social Effects on Consumers' Adoption vs. Defection Decisions
SPEAKER: Vardit Landsman and Irit Nitzan from Tel-Aviv University
12 במרץ 2013, 14:00
בניין רקנאטי ,חדר 403
Social effects among consumers within a given social network have been thoroughly explored in the
context of product adoption (e.g., Katona, Zubcsek and Sarvary 2010; Goldenberg et al 2009). These studies
indicate that adoption decisions of neighboring members of the network have a positive effect on the focal
member’s decision to adopt the product. Recently, studies found similar evidence for network effects in the
context of service defection decisions, namely that defecting neighbors have a positive effect on customers’
likelihood to defect (Nitzan and Libai 2011; Dierkes, Bichler and Krishnan 2011).
Two open questions remain. First, no empirical finding to date enables the comparison between the
magnitudes of such social effects in the opposing processes of adoption and defection. Second, it is not yet
clear whether cross-process effects exist for these opposing decisions. That is, whether exposure to network
neighbors who defect the product affect one's adoption decision, and whether exposure to network neighbors
who adopt the product affect one's defection decision. This study aims to bridge these gaps in the literature
theoretically and empirically, by exploring the social influences at play in adoption and defection processes
concurrently, for a single (the same) product, in a given social network of customers (same population), in a
given (the same) market situation.
For this study we combine two unique dataset. The first data set contains communication data among
over one million customers of a cellular company over time that enables us to create a large-scale social system
composed of customers' individual social networks. The second data set is composed of adoption and
defection dates for an add-on cellular service offered to the network customers starting from the introduction
of this add-on.
Using these data we estimate a joint multi-event proportional hazards model to assess customers'
hazard of adopting and defecting this add-on service, taking into account adoption and defection decisions
within the social network together with information regarding the structure of that network.
This research can assist managers to better understand the social mechanisms that drive adoption and
defection decisions, and may help companies improve their prediction abilities and budget allocations with
respect to these two central consumption decisions.
