Database Reference
In-Depth Information
that their customer loyalty is sustained for a long period of time. Finally, identifying how
much to spend on a customer is arguably the most important piece of the customer retention
puzzle. It is very easy for firms to over-communicate with a customer and spend more on his/
her retention than the customer will ultimately give back to the firm in value.
The decision patterns would incorporate
◾
Explaining customer retention or defection
◾
Predicting the continued use of the service relationship through the customer's
expected future use and overall satisfaction with the service
◾
Renewal of contracts using dynamic modeling
◾
Modeling the probability of a member lapsing at a specific time using survival
analysis
◾
Use of loyalty and reward programs for retention
◾
Assessing the impact of a reward program and other elements of the marketing mix
c. 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 objec-
tive of customer attrition modeling includes churn with time-varying covariates, mediation
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 possible by monitoring customer purchase behavior,
attitudinal response, and other metrics that help identify customers who feel underappreci-
ated or underserved. Customers who are likely to churn do demonstrate
symptoms
of their
dissatisfaction, such as fewer purchases, lower response to marketing communications, lon-
ger 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 versus 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 particular customer?
d. Customer win-back: This involves decisions on reacquiring the customer after the customer
has terminated the relationship with the firm. The objective of customer win-back modeling
includes customer 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, 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.
