Customers are not created equal. Breaking customers into homogenous groups - called segments, helps with 2 objectives -
- Help with understanding customer better. Aka know your customer.
- Improve targeting and communication to drive better results.
It sounds easy and powerful.
Only problem with segmentation is it goes out of hand very soon. You can segments on geographic, demographic, technological, behavioural and what not. Also on each segment (may be 100+), you need to make different targeting. Not worth my time and effort.
RFM (Recency, Frequency and Monetary) framework of segmentation based on customer behaviour is one of the best approach to take for segmentation to keep your sanity and get results.
RFM stands for segmenting your customer base on Recency, Frequency and Monetary behaviour taken together. Looking all of them together for a single customer is a key, otherwise it will be imbalanced.
How recent the interaction / behaviour shown by the customer.
Taking purchase behaviour, when was the last order was placed? Taking visit behaviour, when was the last visit done by the customer on website / app?
Again, taking purchase behaviour, how frequently customer places the order in a given time frame? Taking visit behaviour, how frequently customer visits the website / app in a given time?
Taking purchase, we can include the total life time value of the customer. Taking visits, we can include the total ad revenue generated by this customer.
To get the RFM analysis for your customers, here are the steps -
- Take the customer data in simple excel or database with following fields
- Customer ID
- Last behaviour done date / time. Like last order date.
- Total Interactions during time frame / Total orders in last year
- Total monetary worth during time frame / Customer life time value.
- Give scale to the