RFM ANALYSIS AND SEGMENTATION
Companies want to know their customers and understand how they think, feel, and most importantly how they make purchasing decisions using data. The success of a business lies in offering the right product to the right consumer. This makes sense, if a company knows, their customers’ behaviors and properties, it can determine the marketing strategy more accurately.
In today’s digital world, marketing has become a precise and customer-centric process. Based on the size of companies, they may have lots of customers. In this sense, a company must analyze its customers and understand their behaviors.
Customer segmentation can help companies divide their customers into unique groups to analyze their needs and communicate with them effectively. This leads to a competitive advantage for companies, as well as improves their profits on their ad expenses. In this series of articles, I will demonstrate to you how to segment customers by using RFM analysis.
What is RFM Analysis?
The RFM is an abbreviation of recency, frequency, and monetary.
Recency: refers to the date of last purchase from the specific date.
Frequency: refers to the total number of orders
Monetary: refers to how much money the customer spends in a given period?
RFM analysis is a marketing technique that is used to quantitatively rank and group customers based on recency, frequency, and monetary values. These RFM metrics are important indicators of a customer’s behavior because the frequency and monetary value affect a customer’s lifetime value, and recency affects retention. I will explain customer’s lifetime value and how to segment customers by using it in detail in a later section of this article. RFM Analysis assigns each customer numerical scores based on recency, frequency, and monetary values to provide an objective analysis of our customers.
RFM analysis helps companies find answers to the following questions:
Which customers do contribute to the company’s profit most?
Who are the new customers of the company?
Which customers could churn?
Who has the potential to become valuable customers?
Which customers of the company can be retained?
Let’s demonstrate how RFM Works;
After that, we sort calculated monetary, frequency, and recency values from largest to smallest then split the sorted values 5 even part.
Recency: calculated by subtracting the customer’s last purchase date from the calculation day.
Frequency: total number of purchases completed by customers in a fixed time period.
Monetary: How much money is spent by a customer in a fixed time period.
After that, We sort calculated monetary, frequency, and recency values sort from largest to smallest then split the sorted values 5 even part.
Recency Score: we have sorted customers by recency values, Since customers are assigned scores from 1–5, the latest shoppers 20% of customers receive a recency score of 5, the next recent 20% a score of 4, and so on.
Frequency Score: we have sorted customers by frequency values, since customers are assigned scores from 1–5, the most frequency 20% of customers receive a frequency score of 5, the next frequency 20% a score of 4, and so on.
Monetary Score: we have sorted customers by monetary values since customers are assigned scores from 1–5, big spenders 20% of customers receive a monetary score of 5, the next monetary 20% a score of 4, and so on.
After calculating these scores separately we calculate the RFM_SCORE by combining these individual scores.
This output of code demonstrates how it works and looks.
Customer Segmentation Process:
The following two-dimensional graph is generally used for the segmentation process of RFM. We use recency and frequency scores to visualize segments.
Interpretation of some of these segments;
cant_loose:
There are 71 people in this segment, on average, they made their last purchase 331 days ago, and spent an average of 3881 USD. Since these customers, who contribute greatly to the company’s income and whose numbers are quite low, have high recency values, the company should remind itself via SMS and direct customer visits. As we can see from the table, a group of 71 people has high income and frequency. If this number is low, the advertising and campaign expenses that the company will prepare for its customers will be relatively low. If their recency values increase, they will enter the champions segment and become one of the customers who make the biggest contribution to the company’s revenues.
at_risk:
There are 750 people in this segment, they made their last shopping on average 377 days ago. This group has a shopping frequency of 3.9 and has an average of 1383 USD expenditures. If they are not taken care of by the company, they will completely go away from the company. In order to keep them as customers of the company, some promotions should be introduced via notifications from communication tools like SMS, and there is also a need to investigate why they have not shopped for a long time. The company can also call these customers about their promotions and try to understand their reasons for not purchasing from the company. There is a marketing proverb: “ the cost of retaining a customer is much cheaper than acquiring new customers”.
loyal_customers:
There are 1147 people in this segment who made their last shopping on average 54 days ago, have a good shopping frequency of 9.8, and spent an average of 4199 USD. These customers can increase the amount of shopping if it is introduced some discounts for the product group they are interested. Additionally, the recency values should be increased by examining the shopping product groups and cross-product sales. Customer-specific product presentations can be implemented via SMS and mobile applications. If the recency values are increased with this strategy for this customer group, they can enter the champions group and help to increase the profitability of the company to a great extent.
The cons of the RFM:
The RFM consists of the historical data about customers, and it mainly refers to the previous behaviors of the customers. Hence, it may not accurately indicate future activities.
RFM segmentation is a powerful method for customer segmentation; the RFM scores consist of three main approach frequency, recency, and monetary. Therefore, the RFM segmentation provides a good insight into customers. However, it doesn’t take into account many other factors about the customers such as what kind of items are purchased by the customer. It also doesn’t reflect customer demographic attributes such as gender, age, and so on.
Thanks for reading. Please let me know if you have any feedback and question.
