Suppose you are the marketing operation of a financial management company. The company has recently launched a series of new products, covering high, middle and low grades. You have a customer list at your hand.

Send unified promotion to all customers?

Save time and effort, but it is conceivable that the effect is not good……

Communicate with each customer, understand the needs before recommending?

Can achieve accurate recommendation, but you look at the mass of customer information, or shake your head……

If only you had a device that could tag your customers based on their track record, so that you could know which customers to focus on and recommend different products for different types of customers to maximize revenue.

So congratulations, today we are going to introduce “RFM customer analysis model” is the magic tool you are looking for!

What is the RFM model?

RFM model are the important customer value measuring tool, in RFM model, we take the * * (R) a recent consumption, consumption frequency (F) (M), consumption amount * * three dimensions of customer value segmentation, and then to different value customers on different labels, and then carry out personalized customer service, the limited resources reasonably allocated to customers of different value, Achieve maximum benefit!

  • R(Recency) : How long ago is the last purchase made by the customer
  • F(Frequency) : number of purchases made by a customer in a recent period
  • M(Monetary) : The amount of money purchased by customers in the recent period

In addition, the three indicators in the RFM model can be replaced, so the RFM model can be competent for any scene that uses three dimensions to evaluate and classify. The RFM model can be applied to almost any field, so it is a basic and necessary lesson for a data analyst/business person to learn RFM analysis.

RFM model principle is simple, Excel can also build, but need to write a lot of functions and code, the process is complex, very unfriendly to technical white….. So many data analysts have started to use BI visualization tools to build RFM models.

Here I would like to recommend a BI tool — FineBI, FineBI is a powerful self-service analysis platform, almost no technical foundation, xiao Bai can quickly start, easily build a variety of beautiful and practical data analysis models, and personal version of the permanent free trial.

Here is how to build an RFM analysis model in 10 minutes with FineBI.

(Download below)

First, the overall idea

RFM analysis requires complex data processing, but we use the self-service data set function of FineBI, which only requires simple drag and drop to complete data processing. The realization idea is shown in the figure below:

  • Create a self-service dataset and select the fields required for RFM analysis.
  • The data were processed and the three key indicators and their average values were obtained.
  • By comparing with the mean, vectorization of three indicators.
  • Customer classification according to feature vector.

Let’s finish it step by step.

Two, practical operation demonstration

This article takes the detailed DATA of RFM (click to download) as an example.

1. Create a self-service dataset

  • Enter the “Data Preparation” interface, select the “Style Data” service package, click Add table, and select add “Self-service Data Set”, as shown below:

Go to the self-service data set editing interface, choose Data List > Style Data > RFM Detail Data, add all the fields in the table, and name the self-service data set “RFM Analysis”, as shown in the following figure:

2. Calculate the key consumption indicators of each customer

To calculate the average value of the total consumption amount of customers: click “+” and select “New Column”, as shown below:

Name the new column “Average value of total Consumption amount of customers”, select “All values/Within group” and set as follows, and click “OK” to obtain the average value of total consumption amount of customers. As shown below:

3. Calculate the overall consumption index of customers

To calculate the average value of the overall consumption frequency of customers: click “+” and select “New Column”, as shown below:

Set the name of the new column as “average value of customer’s overall consumption frequency”, select “All values/Within the group”, and click “OK” to obtain the overall average index “Average value of customer’s overall consumption frequency”. The Settings are shown below:

To calculate the average time of the last consumption distance of the overall customer: click “+” and select “New Column”, as shown below:

Set the name of the new column to “Average time of the most recent customer consumption distance”, select “All Values/Within the Group” and click “OK” to obtain the average value indicator “Average time of the most recent customer consumption distance”. The Settings are shown below:

4. Vectorization of customer characteristics

Vectorization of customer characteristics based on whether key indicators are greater than the overall average level of customers.

In the formula IF(XXX > the average value of customer’s overall XXX,1,0), the value less than the overall average is set to 0, and the value greater than the overall average is set to 1, so that 1 maintains positive characteristics and 0 maintains negative characteristics.

Vectorization of consumption amount: Click “+” and select “New Column”, as shown below:

Name the new column “consumption amount vectorization”, enter the formula IF(MONEY> total average consumption amount of customers, 1,0), and click “ok”, as shown below:

5. Analysis of customer characteristics

Click Add to add new columns and use CONCATENATE() to CONCATENATE RFM vectorization values, as shown below:

At this point, a simple RFM analysis model was completed, and each customer was successfully tagged, followed by a visual display of customer categorization data through the dashboard.

Analysis Tool Sharing

Finally, the tools and data sets are ready for you, and you can get them all by clicking “RFM”!