Browsing by Subject "Structural models"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Essays on the application of multitasking in marketing channels.
    (2010-06) Banerjee, Ranjan
    Firms routinely engage channel partners to create and deliver valued products and services to end customers. There are a large variety of such partners performing several different tasks across the stages of the buying process, including prospecting for customers, making product recommendations, educating customers, and closing the sale. Managing multiple delegated tasks poses a significant challenge because efforts on individual tasks are often not directly observable. Multi-tasking (or the presence of multiple tasks) is ubiquitous in marketing channels. However, there is a dearth of empirical literature which sheds light on problems of task allocation and contract design in situations where firms desire to have multiple tasks performed through channel members. In this dissertation, we attempt to address this gap through three essays which provide theoretical and empirical insights into issues of task allocation, contract design and the relationship between contract design and channel performance in multi-task settings. In essays 1 and 2, we consider a situation where a firm wishes to have multiple tasks performed through a number of retail stores which differ in terms of locational characteristics. We build and estimate a model which is geared to this setting, and counterfactually compute the performance implications of alternative channel structures. Our results suggest a performance rationale for the deployment of multiple channels, and validate theoretical predictions about the dampening of incentive effectiveness in the presence of a non-measured second task. In essay 3 we consider a two stage process where telecallers generate leads which are then passed on to salespeople for conversion. We use a unique combination of a structural model and a field intervention to estimate relevant parameters for both salespeople and telecallers. Our analysis provides a generalizable methodology for incentive design in these settings and suggests that the nature of interdependence between tasks crucially affects incentive loading between intermediate and final outputs.