Whenever a customer buys something from an outlet or shopping mall, the cashier asks for their phone number, one might think that they are just using it for their record, but a lot of information can be derived from a particular transaction of the customer using which they segment their customers and target them using a different strategy.
Whenever the cashier uses the number, they have the customer’s past purchase history displayed. First is the date of purchase throughout the time they have been shopping at a mall, next is the number of purchases over a period and your overall spending through that period. Since every phone number is unique, it acts as a unique ID for the customer.
To do the segmentation and analyze the customer purchasing behavior we perform RFM analysis.
R stands for Recency which means the freshness of customer activity, be it purchases or visits. Customers that made a purchase recently are more responsive to promotions.
F stands for Frequency which means the frequency of customer transactions or visits. Customers that purchase frequently are likely to purchase again and stay engaged
M stands for Monetary value which means the purchasing power of a customer. Monetary value differentiates heavy spenders from low-value purchasers.
80% of business comes from 20% of users.
Pareto Principle is at the core of the RFM model. Focusing your efforts on critical segments of customers is likely to give a much higher return on investment. RFM is an excellent segmentation method used to predict customer responses, improve customer interactions and increase profit
Customers can be divided into segments based on the RFM Score.
To get the RFM score we require the following datapoint:
- UserID (Any unique identifier)
- Date of Purchase
- Amount Spent.
We can calculate recency by subtracting the current date with the last date of purchase, the frequency we can get by grouping Date of Purchase by UserID similarly if we do the sum of monetary value after grouping for getting the total value spent per customer.
The example below presents the way we calculate it using a scenario. The example below represents the way to calculate the RFM score for two months.
From the above image we see that customer A usually spends 5k once a month so the last time they had made a purchase would have been 30 days ago, hence we put the R-value for customer A as 30. Similarly, we do it for B and C. The Frequency value of customer A is 2 since he spends 5k once a month and we are considering the data for 2 months so the value would be 2. Since B purchases 10 times per month, the value would be 20 for two months and since C only purchases once per two months, the F value would be 1. Coming to M value, since A spends 5k per month, for two months the value would be 10k, For B, 500*10*2=10,000, and since C spends 10k once in 2 months so the value will be 10k.
For segmenting we give them a score from 1 to 5 where 5 represents the highest value segment. We create segments by dividing the highest value by 5 and create segments based upon it. For R Score, the highest R-value is 60 Days which means 60/5=12, we create segments like 0-12 will be assigned as 1, 13-24 will be assigned as 2, 25-36 as 3, and so on. For Customer A the 30 Days value will come in the 3rd segment so we will assign the value as 3. For B, the R-value is 3 which will come in segment 1. C will be assigned to the last segment which is 5.
Using RFM analysis we can retain more customers and get an idea of how the brand is performing based upon the percentage of each segment. RFM, if utilized effectively, can help brands to achieve higher revenue and a loyal customer base.
To get an in-depth analysis of your customers and their purchase life cycle, reach out to our team at Merkle Sokrati.