Author: Philip Godfrey
Marketers typically have extensive data on their existing customers – such as purchase history, browsing history, prior campaign response patterns and demographics – that can be used to identify specific groups of customers that can be addressed with offers very relevant to each.
This is where RFM analysis, or customer segmentation, can help us.
My last blog covered the basics of RFM, if you missed it or would like to read it again, you can find it here.
For this blog, we will go through the steps for performing RFM analysis, from identifying the data we need, through to labelling customer segments and assigning them to our example data. We will end with some actionable insights, to make sure we can take this analysis back to business stakeholders to understand our customers in greater detail.
Preparing data for RFM analysis
1. Define a timeframe for the analysis – this could be the last 10 years as an example
2. Export the data - at a minimum we need:
a. Customer ID
b. Last purchase date (r)
c. Total number of purchases (f)
d. Total purchases (m)
|
ID |
Last Purchase Date - R |
Total Number of Purchases - F |
Total Purchases (£) - M |
|
1 |
01/01/2022 |
1 |
£100 |
|
2 |
01/01/2020 |
2 |
£100 |
|
3 |
01/01/2021 |
5 |
£500 |
3. Assign scores for R, F and M values - these could be ranked 1-10 for each value
Scores are assigned a ranking from 1-10 for each value, but these can change depending on your business needs.
Recency
|
Last Purchase date |
R Score |
|
01/01/2013 |
1 |
|
01/01/2014 |
2 |
|
01/01/2015 |
3 |
|
… |
… |
|
01/01/2022 |
10 |
Frequency
|
Number of Purchases |
R Score |
|
1 |
1 |
|
2 |
2 |
|
3 |
3 |
|
… |
… |
|
10 |
10 |
Monetary
|
Total Purchases (£) |
R Score |
|
1 |
1 |
|
50 |
2 |
|
100 |
3 |
|
… |
… |
|
100,000 |
10 |
4. Sum
scores to create an RFM values – using
the ranked scale above, this would mean the lowest score could be 3/30, with
the highest score being 30/30
|
ID |
R Score |
F Score |
M Score |
RFM Score |
|
1 |
10 |
1 |
3 |
14 |
|
2 |
7 |
2 |
3 |
12 |
|
3 |
9 |
5 |
5 |
19 |
5. Prioritise high scorers
If we organise our customers into the highest RFM score, these are customers who have likely purchased from us recently, made a higher number of purchases over the time period, and in total spent a good amount of money with the company.
|
ID |
Last Purchase date - R |
Total Number of Purchases - F |
Total Purchases (£) - M |
RFM Score |
|
3 |
9 |
5 |
5 |
19 |
|
1 |
10 |
1 |
3 |
14 |
|
2 |
7 |
2 |
3 |
12 |
6. Labelling customer segments and assign
The segments and labels below are frequently used as a starting point, but you can come up with your own segments and labels that is better fits for your customer base and business model.
Champions: Bought recently, buy often, and spend the most
Loyal customers: Buy on a regular basis. Responsive to promotions.
Potential loyalist: Recent customers with average frequency.
Recent customers: Bought most recently, but not often.
Promising: Recent shoppers but haven’t spent much.
Needs attention: Above average recency, frequency, and monetary values. May not have bought very recently.
About to sleep: Below average recency and frequency. Will lose them if not reactivated.
At risk: Some time since they’ve purchased. Need to bring them back!
Can’t lose them: Used to purchase frequently but haven’t returned for a long time.
Hibernating: Last purchase was long back and low number of orders. May be lost.
|
ID |
Last Purchase date - R |
Total Number of Purchases - F |
Total Purchases (£) - M |
RFM Score |
Segment |
|
3 |
9 |
5 |
5 |
19 |
Champions |
|
1 |
10 |
1 |
3 |
14 |
Recent customers |
|
2 |
7 |
2 |
3 |
12 |
Can’t lose them |
7. Review performance of customer base
Optionally, we might want to view this across our entire customer base to understand what type of customer we have and if there’s anything we need to address.
8. Actionable Insights
Now we understand our customer base in more detail, it’s time to start taking actions.
· At-risk customers can be targeted with offers and discounts.
· Recent customers can be sent information about other products that they could be interested in.
· Champion customers could be given greater access to products and used as a mechanism for feedback, before launching it to other customers.
The best news is all of this can be done simultaneously by the business.
9. Implementing RFM Analysis in Oracle Cloud
Oracle Cloud offers a range of tools and technologies that enable businesses to carry out RFM analysis efficiently:
Data Integration: Oracle Cloud allows seamless integration of various data sources and systems, ensuring a comprehensive view of customer transactions and interactions.
Data Analytics: Leveraging Oracle Analytics Cloud (OAC) or Oracle Autonomous Data Warehouse (ADW), businesses can easily perform RFM analysis on customer data. Advanced analytical capabilities enable insightful visualizations and reporting, enabling businesses to gain a deeper understanding of customer segments.
Automated Segmentation: Oracle Unity Customer Data Platform enables the automated segmentation of customers based on RFM value analysis. This is a built-in offering as part of the Customer Analytics area of CDP, using Machine Learning and Oracle generated personas. This segmentation provides businesses with data-driven customer profiles, making it easier than ever to tailor marketing efforts and deliver personalized experiences.
Comments
Post a Comment