RFM (Recency, Frequency, and Monetary) ANALYSIS

Introduction

RFM analysis is a marketing analytic technique that can be used to infer basics information needed to evaluate performance of a product or the company as a whole on the basis of customers’ response to offers. Figure 1 shows the various marketing analysis that can be done on IBM SPSS.

Figure 1: Various possible Marketing Analysis on IBM SPSS.

For instance, have you ever wondered why you are glued to shopping on a particular website? Well, your data might have gone through analysis whereby your tastes are identified and presented to you when you need them most or least. How? This blog can show some of the analyses that can be used for such a study!

Recency (R) is crucial in determining which clients or customers are most likely to respond to a new offer.

Frequency (F) is the second-most significant component. Customers who have purchased more frequently are more likely to reply than those who have purchased less frequently.

Monetary (M) is total expenses, the third most crucial aspect. Customers who have made more purchases or expenses in the past (across all sales) are more likely to reply than customers who have made fewer purchases.

Data

The data used for this blog can be found from the link:

https://www.kaggle.com/datasets/rodsaldanha/arketing-campaign

RFM in SPSS

Analyze >>> Direct Marketing >>> Choose Technique

In the next section, we have to choose the technique to be used, in this case Understand My Contacts >>> Help Identify my best contacts (RFM Analysis)

The next section requires us choose the form of our data, such as Transaction data or Customer data. In our case, we have both the transaction and customer data but we shall be going for customer’s data as shown below.

Results

Bar chart and Heat map are produced to observe the classification of the groups from the data. The categorized bar chart is subdivided into five (5) classes. From the part of the frequency, Class 2 has the highest frequency, followed by the Class 3. Class 1 has the least frequency. The information in the categorized bar chart is better known in the heat map.

From the heat map, it is evident that based on the recency of transaction of the customer Class 1 takes the lead. This suggests that the customers that earn between $70000 - $100000 patronized the company more than the others. Thus, this set of customers can be targeted for promo by placing offers on their products data.

Conclusion

In this blog, we have introduced RFM analysis using IBM SPSS. Having used a marketing campaign data the analysis shows that the customers with income between $70000 - $100000 are more likely to respond to offer than the lower earning customers.

Related blogs: Forecasting, Control-Chart