There are many marketing concepts around the customer: cost of acquisition, retention, life cycle (also called lifetime value), attrition… In order to optimize their results, many companies choose to implement a segmentation of their customer contact base.
Different methods exist, one of the best known methods: RFM segmentation.
RFM segmentation is behavioral segmentation. It is not based on the characteristics specific to customers (gender, sex, place of residence, annual income, etc.), nor on their preferences (interests, tastes, etc.), but on their purchasing habits.
The principle is simple: a customer who has bought recently, who buys frequently and who generates a lot of income has a higher probability of buying again than another customer.
RFM segmentation is therefore based on three criteria:
RFM segmentation allows you to target your customers based on their purchasing behavior, with more relevant messages and improve the return on investment of your marketing actions.
There are many ways to segment your customer base. For the RFM method to be the most effective, it is important to take into account the particularities of your business and adapt your segmentation accordingly.
RFM segmentation identifies:
We propose here a segmentation into ten distinct groups. Depending on your sector of activity, the maturity of your company, the number of contacts in your customer base, etc. It is not necessarily necessary to implement them all.
Before setting up an RFM segmentation, the reference period must first be determined. Depending on the purchasing cycle and the average rate of consumption of your customers, it could be a period of a month, a quarter, a semester, or even more.
Once the period has been defined, it is then possible to calculate the RFM score according to the recency (the number of days since the last purchase), the frequency (the average purchase frequency over the period) and the monetary value (the average basket over the period).
Different RFM calculation methods can be used to normalize RFM values on a scale of 1 to 5. Below are the two most common methods.
With this method, you define a range for each score yourself. Thresholds are based on the nature of your business.
Example: According to Purchase Recency (Range / Score)
Easy to implement, this method has one main “flaw”: its fixed character.
You will need to review and adjust the ranges manually as your business grows to keep the RFM score relevant.
This method consists of dividing your data into 5 equal parts. This method involves slightly more complicated math. However, this solves the problems posed by the method of simple fixed range. The quintile method is scalable, it works with any sector. Scores are automatically assigned to customers.
This is the method we recommend you use to calculate the RFM score.
You can integrate your RFM segments into your emailing solution (like Dialog Insight, for example) and use them to create targeted campaigns and scenarios for each segment.
If you have a loyalty program, the “Potentially loyal customers” segment is an interesting segment to target. Their experience with your products and services should be enjoyable and memorable. Offer them time-limited promotions and there’s a good chance they’ll buy. Send them educational content to improve their engagement with your brand.
Hibernating customers and at-risk customers are the two segments you need to pay special attention to. Send them personalized emails to try to reconnect with them. You can also offer them discount coupons on repeat purchases or send them customer surveys to prevent them from going to the competition.
In this article, we’ve walked you through the basics of RFM segmentation. If you want to deepen this concept or test its effectiveness, Dialog Insight experts can support you. Contact us today.
