Zelios Analytics

View Original

How to segment your customers with RFM analysis

Nothing is more important in analytics than segmentation. Yet most of our reporting and analysis happens at an aggregate level. Different types of users visit your website with different intentions. This phenomenon mandates that you have a very effective and persistent segmentation strategy as part of your web analytics process. Using segmentation, you divide your audience into homogeneous groups in order to send them relevant communication that resonates with them.  One such data backed segmentation technique is RFM segmentation.
 

What is the RFM analysis?


RFM (Recency, Frequency, Monetary) analysis is a proven marketing model for behavior based customer segmentation. It groups customers based on their transaction history – how recently, how often and how much did they buy. Given that you have stored this information into your database (i.e. CRM system), you can then divide your customers into various categories or clusters (like loyal customers, big spenders, lapsed customer) to identify those who are more likely to respond to specific offers/campaigns as well as for future personalization services. 

Common practice when it comes to segmentation is to think that ‘big spenders’ are the most valuable clients. But what if they purchased only once or a very long time ago? Do they still use your product? It makes sense to reward all clients that keep buying your services/products on a regular basis, spending as much money as possible. Are you able to track that? The answer is positive.
 


Why RMF is crucial for marketing strategists?
 


RFM analysis can play a key role in increasing the profitability of advertising campaigns. It can be used in different channels and for various purposes such as the response rates to email campaigns, the transactions in an online shop, or the conversions of websites which offer a whitepaper for download. Campaign optimization, segmentation and deeper target group analysis can also be carried out based on RFM analysis. Most RFM models can be adapted to the respective conditions. However, they often also require a comprehensive CRM system that includes such analytical methods and a rather sophisticated data collection, which must be integrated and implemented using tracking methods. This is generally also possible in Excel, if the transaction is at least available.
 
Knowing your best (and worst) customers can give you important insights for a smart marketing strategy.  However, the use of these findings in marketing is often viewed critically. While RFM scores can help identify the customers with the greatest purchasing power, marketing activities should not only address that customer group, even if the bulk of sales are expected from it. Customer groups are arbitrarily divided. A type of profiling is used, which can be only poorly conveyed to the customers. If only paying customers are given special offers, there is a risk that other customers will feel discriminated against when they hear about it. Many marketing experts recommend focusing on the low-paying customer groups to increase their loyalty and purchasing power.
 
Hence, although RFM is a great tool, smart marketers know not to rely on RFM alone when developing a marketing plan. You also need to consider input from your sales team, feedback from your customers and the results of prior marketing initiatives to decide how best to market to your current customers and targeted prospects.
 

How to calculate the RFM score?
 


Once you have the RFM values from the purchase history, you assign a score from one to five to recency, frequency and monetary values individually for each customer. Five is the best (highest) value while one is the lowest. A final RFM score is calculated by combining individual RFM score numbers to create a three-digit number.  Thus, customers who purchased recently, are frequent buyers and spend a lot are assigned score of 555 – Recency(R) – 5, Frequency(F) – 5, Monetary(M) – 5. They are your best customers. 
On the other extreme are customers spending the lowest, making hardly any purchase and happened a long time ago – a score of 111. Recency(R) – 1, Frequency(F) – 1, Monetary(M) – 1. You can read more about the details of calculating the RFM score by reading this article.
 
 


Software for RFM analysis
 


RFM has become an integral part of marketing and business analytics. Hence if you are doing one-off evaluation of your customers’ shopping behavior, you can perform a manual or semi-automated RFM analysis using Excel. However, if you have a slightly large database, you don’t want to do all the complex calculations yourself. You can use Python or R to get your results in just a few seconds, thus doing this analysis at scale and fast. I have written an example of RFM analysis with Python that you can check on my Github account. In case you have your own data science team, it would be best to create a custom RFM model for your business using your preferred tools.
 

Bringing it all together, the RFM analysis is a scoring process with which you can sort customers into target groups (segments) and it aims to identify customers who are most likely to respond to a range of different marketing methods. RFM analysis is most frequently used in direct marketing, in campaign optimization and in email marketing. If you have mature customer data collection and an active email list, RFM analysis can create a strong return on investment. Have you conducted an RFM analysis?

Get in touch and tell me about your experiences!