Context-Aware Applications - Guide to Cloud Computing for Business and Technology Managers - page 507

3. Customer attrition or churn: This involves decisions on whether the
customer will churn or not, and, if so, what will be the probability of
the customer churning, and when. The objective of customer attri-
tion modeling includes churn with time-varying covariates, media-
tion effects of customer status and partial defection on customer
churn, churn using two cost-sensitive classifiers, dynamic churn
using time-varying covariates, factors inducing service switching,
antecedents of switching behavior, and impact of price reductions
on switching behavior.
Engaging in active monitoring of acquired and retained customers is the
most crucial step in being able to determine which customers are likely to
churn. Determining who is likely to churn is an essential step. This is pos-
sible by monitoring customer purchase behavior, attitudinal response, and
other metrics that help identify customers who feel underappreciated or
underserved. Customers who are likely to churn do demonstrate symptoms of
their dissatisfaction, such as fewer purchases, lower response to marketing
communications, longer time between purchases, and so on. The collection
of customer data is therefore crucial in being able to identify and capture
such symptoms and that would help in analyzing the retention behavior and
the choice of communication medium. Understanding who to save among
those customers who are identified as being in the churn phase is again a
question of cost vs. future profitability.
The decision patterns would incorporate
• When are the customers likely to defect
• Can we predict the time of churn for each customer
• When should we intervene and save the customers from churning
• How much do we spend on churn prevention with respect to a par-
ticular customer
4. Customer win-back: This involves decisions on reacquiring the cus-
tomer after the customer has terminated the relationship with the
firm. The objective of customer win-back modeling includes cus-
tomer lifetime value, optimal pricing strategies for recapture of lost
customers, and the perceived value of a win-back offer.
Identifying the right customers to win back depends on factors such as the
interests of the customers to reconsider their choice of quitting, the product
categories that would interest the customers, and the stage of customer life
cycle and so on. If understanding what to offer customers in winning them
back is an important step in the win-back process, measuring the cost of
win-back is as important as determining who to win back and what to offer
them. The cost of win-back, much like the cost of retention or churn, must
be juxtaposed with the customer's future profitability and value to the firm.