What is an author doing?

An author is doing a research paper title “Deriving and exploring behavior segments within a retail loyalty card program” to identify the customer cluster that deriving by the royalty card program, so the management can use the information to implement the right strategy to the right customer.

Why is the author doing what he/she does?

  • The author notice that even though the rapidly increasing number of retail loyalty programs. Most of the company still not notice or pay attention to the difference in the behavior patterns of the customers within the loyalty programs.
  • The author wants to know that is there ability to create customer segments within a retail loyalty program. If it can, what are the drivers of difference among them that make each of them unique and whether it need difference managerial implications (strategy) to manage in order to yield higher customer satisfaction and company profitability.

How does the author do the project?

The author use data of the one-year test of retail loyalty card program provided by one major US retailer. The author formed the segment by numerical taxonomy process. First, the author using cluster analysis to generate a range of potential market structures (grouping 57,650 loyalty card program members in to 2 through 11 group) for classifying each of the loyalty card patrons on a set of managerially relevant variables. Each group was ranking their segment by least patrons to highest patrons.

Second, select the most appropriate market structure by using scree testing (test important) and discriminant analysis. The result was either five or six segments might yield the best market structure, stable and reliably predict group membership. However the author choose six segments because it create the unique characteristics of the best group and allow more definition within the three most interesting groups.

Third, the author names six segments (clusters) by its characteristic;

Cluster 6: ideal, highly loyal shoppers – most loyal group of card holder spent over twice as much per person as members of the cluster 5. The smallest group only 1% of total cardholder population.

Cluster 5: half loyal shopper – Around 9% of total cardholder population. Spent three times higher than cluster 3.

Cluster 4: late but enthusiastic followers – This group is the most impacted by the loyalty card program. Since they increased participation of the program (their purchasing 600 percent during the second half of the year).

Cluster 3: shopper who lost their enthusiasm – This group unsatisfied with the program because their rate of use continue to drop.

Cluster 2: very infrequent card shoppers – Infrequent use of card, this group did not reach the spending threshold required for the rewards by the loyalty program

Cluster 1: shopper who wanted to like it but not – Around 39% of total cardholder population. This group tried the program, stayed with it for some period of time, but did not like it enough to continue and this group did not reach the spending threshold required for the rewards by the loyalty program.

Finally after the author name six clusters and it characteristic, the author regress a set of patronage-related variables onto the clusters using multinomial logistic regression. To find what are the drivers of difference among clusters. The author able to find out that the distance to the store, distance to the nearest competitor and the number of very early adopters in the cardholder’s neighborhood interaction field were the most importance driver accounting for overall difference between the most loyal cluster and other clusters. The author also makes the suggestion on which strategy should use to approach each clusters and the limitation of this research in order to make a better research next time.

In conclusion, it appear that the ability to create segments within loyalty card program and understand the drivers of differences among them can be great benefit to the company by using difference strategy that match the difference customer need in each segment to increase the customer satisfaction, to know which customer the company choose focusing on and to increase the profit to the company.