Title:
|
BEHAVIORAL MARKETING APPLIED TO SOME DIGITAL PRODUCT CATEGORIES |
Author(s):
|
J. F. S. Castel-branco |
ISBN:
|
978-972-8924-49-2 |
Editors:
|
Sandeep Krishnamurthy and Pedro IsaĆas |
Year:
|
2007 |
Edition:
|
Single |
Keywords:
|
CRM, RFM, segmentation, customer behaviour, digital products. |
Type:
|
Full Paper |
First Page:
|
106 |
Last Page:
|
113 |
Language:
|
English |
Cover:
|
|
Full Contents:
|
click to dowload
|
Paper Abstract:
|
Customer relationship management is decisive in management. In order to allocate resources within the
budget, firms must value their customers and treat them as assets. To do so firms must recognize that
customers are heterogeneous. The premise is that different customers must be treated differently. The
purpose of segmenting the customer base is to transform a large heterogeneous group into several smaller
homogeneous groups. However, as far as digital products are concerned in non-contractual context the only
information firms hold about their customers is the way they behave.
This paper intends to demonstrate that in the absence of any demographic and socio-economic variables,
behavioural information that can be automatically collected in the digital context is enough to group
customers according to their importance to the business and therefore to identify the most valuable ones. In
order to identify the most valuable customers we started by using the RFM model (recency, frequency,
money value). By inserting another behavioural variable extracted from the collected data, the RFM model
was extended. Another segment composed by super customers emerged from the model.
Using the Monte Carlo method we simulated a coupon discount campaign. This simulation allowed us
both to measure and compare the capabilities of the segmentation method through the campaign profitability.
We concluded that the use of the RFM model as a criterion to select a campaign target produces better results
than blind criteria in the non-contractual context of digital products. In addition, the extended model clearly
outperforms the results obtained with the classical RFM model. |
|
|
|
|