- RFM is a marketing analysis technique that stands for Recency, Frequency, and Monetary Value.
- Recency: How recently a customer has made a purchase
- Frequency: How often a customer makes a purchase
- Monetary value: How much money a customer spends on purchases
- RFM analysis numerically ranks a customer in each of these three categories, generally on a scale of 1 to 5 (the higher the number, the better the result). The “best” customer would receive a top score in every category.
- These three RFM factors can be used to reasonably predict how likely (or unlikely) it is that a customer will do business again with a firm or, in the case of a charitable organization, make another donation.
Reference
What Is Recency, Frequency, Monetary Value (RFM) in Marketing?
SuperStore is a global retail company with a large customer base. On the occasion of Christmas and the New Year, the Marketing Department wants to run marketing campaigns to show appreciation to customers who have supported the company over time, as well as to tap into potential customers who could become loyal clients.
However, the Marketing Department has not yet categorized this year's customers due to the large dataset, making it impossible to process manually as in previous years. Thus, they are seeking assistance from the Data Analysis Department to implement a customer segmentation classification task in order to execute tailored marketing programs for each customer group.
The Marketing Director has also proposed using the RFM model. Given the massive amount of data, they hope the Data Department can build a workflow for evaluating segmentation using Python programming.
Table: ecommerce retail
| Field | Type of column |
|---|---|
| InvoiceNo | Dimension |
| Stockcode | Dimension |
| Description | Dimension |
| Quantity | Measure |
| InvoiceDate | Measure |
| UnitPrice | Measure |
| CustomerID | Dimension |
| Country | Dimension |
Table: segmentation
| Field | Type of column |
|---|---|
| Segment | Dimension |
| RFM Score | Measure |
In this section, I outline 5 steps for EDA:
- Descriptive statistics
- Check data types
- Handle incorrect values
- Handle missing values
- Handle duplicates
Please check the attached code file for more details.
After completing the calculation of the Recency - Frequency - Monetary variables, we will detect outliers and process them to ensure more accurate data.
Now, all CustomerIDs are segmented according to RFM analysis
INSIGHTS
| Segment | Insights |
|---|---|
| Champion Customer | With the number of customers in this group accounting for 16% and their spending accounting for 39%, this group consists of customers who have recently transacted, spend the most, and purchase frequently. The business needs to implement loyalty campaigns with special and personalized values. Additionally, the business can suggest high-value Christmas/Tet combos for this group based on their historical purchase data. |
| Loyal and Potential Loyalist Customer | With the number of customers in these two groups accounting for 23% and their spending also accounting for 23%, this group consists of customers who have recently spent at a moderately high level, with frequent or at least more than one purchase. To enhance the cart value of this group, the business can propose Christmas or Tet gift programs when purchases reach a certain value. |
| New Customers & Promising Customer | With the number of customers in these two groups accounting for 11% and their spending accounting for 4%, this group consists of the most recent new customers, with low purchasing power and infrequent purchases, or significant purchasing power but also infrequent purchases. The business needs to implement campaigns that create goodwill from the outset to retain customers. Sending holiday greetings during Christmas or Tet, along with offering a voucher for their next purchase, can stimulate buying interest. Additionally, actively interacting and making suggestions based on historical choices will help retain customers. |
| Need Attention, At Risk, Cannot Lose Them Customer | With the number of customers in these three groups accounting for 17% and their spending accounting for 23%, this group consists of customers who have not made recent purchases; however, they previously had a relatively high cart value and frequent purchase frequency. This is a group of customers that the business cannot afford to lose. There needs to be a campaign to identify the reasons for customer departure and to propose solutions regarding quality and service. |
| About to Sleep, Hibernating Customers, Lost Customers | With the number of customers in these three groups accounting for 31% and their spending accounting for 10%, this group consists of customers who have not made purchases in a long time, have weak purchasing power, and infrequent purchase frequency. The situation indicates that the business should be somewhat alarmed due to the high number of customers in this segment. The business can use email or cold calls to survey opinions and propose discount programs to attract customers back. However, this is very challenging; during this phase, the business should prioritize the Champion segment more. |
=> During the Christmas/Tet holiday, the business should focus on the Monetary metric, as this is the time when customers are more willing to spend compared to other times of the year, due to their need to shop for Tet celebrations and gifts. The business can propose medium-to-high value combos and suggest gifts for orders that meet the specified target value.
INSIGHTS
Strong Growth in High-Value Customer Segments (Champions, Loyal, Potential Loyalist)
- Champions group showed strong month-over-month growth. Notably, the total spending from this group significantly outpaced its numerical growth, indicating that these customers had exceptionally high basket values. In November and December 2011, revenue from this group accounted for approximately 80%-90% of total sales, further solidifying their importance to the business.
- Loyal group steadily increased throughout the year but saw a sharp decline in both customer count and spending in the last two months. This decline could be due to two possible factors: a positive scenario where many Loyal customers upgraded to Champions, or a riskier scenario where some customers stopped purchasing and fell into lower-tier segments. Given the substantial increase in Champions and the overall decrease in lower-tier groups, the first scenario seems more likely.
- Potential Loyalist and Promising groups exhibited steady growth until the end of the year, a positive sign of customer engagement and progression through the loyalty tiers.
Signs of New Customer Growth and Potential Risks
- New Customers group started appearing significantly in November 2011, indicating a successful customer acquisition effort.
- Lower-tier customer groups declined gradually over the months, with some groups reaching minimal proportions by the year’s end. However, the presence of the At Risk group from December 2010 to September 2012 suggests that customer retention challenges persisted. While the total contribution of this group to revenue remained significant and their numbers eventually declined by year-end, their presence highlights an area for improvement to prevent future churn spikes.
RECOMMENDED STRATEGIES
- Loyalty Enhancement: Develop tailored rewards to keep Champions engaged and incentivize high-value spending.
- Onboarding & Engagement: Implement structured engagement plans for New Customers to increase their chances of moving into Loyal or Champions groups.
- Risk Mitigation: Use predictive analytics to identify early warning signs of at-risk customers and proactively re-engage them before churn occurs.
Through the two charts showing the distribution of customer numbers and spending levels of segments based on location, we can see that 90% of customers are from the United Kingdom. Therefore, we can choose this as the primary area to enhance growth strategies.