Jurnal Ilmiah Komputer dan Informatika KOMPUTA
Edisi. .. Volume. .., Bulan 20.. ISSN : 2089-9033
Figure 2. Customer Segmentation Analysis
1 . Data Transactions have been defined and up to
date.
RFM customers are using this method , data on customer transactions should be defined in advance
and up to date . The transaction data which will be used as reference material in the delivery point.
Transaction data
obtained from
customer transactions in August 2014 - July 2015. The data
will be processed can be seen in Table 1. Table 1. Table Customer Transaction
2. Calculate the distance ranges of recency ,
frequency and monetary
In the process of calculating the distance range of recency , frequency , and monetary , previously done
first calculate the value of each of the three indicators recency , frequency and monetary .
a. recency Calculate recency may be obtained by using
equation 1 . The specified date is the date of the latest transaction is July 14, 2015 .
Then to get the range of recency can use statistical calculation formula n odd quartile see equation 2 .
Recency = latest transaction date - the date of the last transaction
1 2
Information : Qi = kuartil ke-i
n = banyak data The amount of customer data that is 119 data.
Calculated as follows :
Q1 = 1 119 + 1 4 = 30 all the data is data
with recency 30 to 56 days Q2 = 2 119 + 1 4 = 60
the data to 60 is data by recency 125 days
Q3 = 3 119 + 1 4 = 90 data is all the data
recency 90 is 210 days Based on the calculations above , we can determine
the distance range recency namely :
1. Recency ≤ 56 days then included customers with a 4 point scale
2. Recency ≤ 57 days ≤ 125 days then included customers with a 3-point scale
3. Recency ≤ ≤ 126 days 210 days then included customers with scale point 2
4. Recency ≥ 211 days then included customers with scale point 1
b. frequency Similarly, recency , to get a frequency range can use
statistical calculation formula n odd quartile see equation 2 . Calculated as follows :
Q1 = 1 119 + 1 4 = 30
data is all the data with frequency 30 is 1 times
Q2 = 2 119 + 1 4 = 60 the data to 60 is the
data with frequency 2 times Q3 = 3 119 + 1 4 = 90
data is all the data with a frequency of 90 is 3 times
Based on the calculations above , we can determine the distance range of frequency are:
Frequency = 1 time then including customers with scale point 1
Frequency = 2 times then including customers with scale point 2
Frequency = 3 times then including customers with 3 point scale
Frequency of 3 times then including customers with 4 point scale
c. Monetary Similarly, the recency and frequency , calculate the
distance range of monetary using the formula n odd quartile see equation 2 . Calculated as follows :
Q1 = 1 119 + 1 4 = 30
all the data is data with monetary 30 2,000,000
Q2 = 2 119 + 1 4 = 60 the data to 60 is the
data with monetary 4,800,000 Q3 = 3 119 + 1 4 = 90
data is all the data with monetary 90 is 19,000,000
Nama Pelanggan Transaksi
Terakhir Jumlah
Frekuensi Jumlah
Monetary
PT. Mandom
Indonesia 07032015
3 Rp 20.500.000
PT. Indosat 14072015
8 Rp 51.800.000
Yogya Group 26082014
1 Rp 2.000.000
PT. Mayora Indah 11022015
3 Rp 16.600.000
PT. Holcim
Indonesia 16122014
1 Rp 4.800.000
Jurnal Ilmiah Komputer dan Informatika KOMPUTA
Edisi. .. Volume. .., Bulan 20.. ISSN : 2089-9033
Based on the calculations above , can be determined within the range of monetary namely :
≤ 2,000,000 Monetary then including customers with scale point 1
2000001 Monetary ≤ 4,800,000 then including customers with scale point 2
4800001 Monetary ≤ 19 million customers with the included 3 point scale
Monetary 19,000,000 then including customers with 4 point scale
3. Calculation Point RFM
RFM has a range of calculations before, can be made RFM point scale . The point scale is obtained
between 1-4 points because there are four categories of customer groups. The following point scale RFM
can be seen in Table 2.
Table 2. Scale Point RFM
Based on Table 2. it can be seen that the total points obtained by summing attributes R + F + M
. If the minimum point of each attribute is 1, the minimum point total R + F + M 3. Meanwhile, if
the customer is the maximum point of each attribute is 4 , then the maximum point total R + F + M
customer is 12 . the next stage is the awarding points based on the RFM transaction data contained in
Table 1. This provision refers to the point RFM Table 2. granting point RFM can be seen in Table 3.
Table 3. Giving Point RFM
4. Grouping customers based point RFM
Based on the results of RFM analysis has been done , the next step is grouping customers based
point RFM . Point range and the number of categories of customers assumed by a discussion
with the manager . Before grouping customers must be determined beforehand pointnya range , the range
of points obtained will be used as reference material for the stage grouping .
Based on Table 2. obtained the highest point total is 12 , while the lowest point total is 3. Since
grouping is divided into 4 groups of categories of customers , then the distance determination
perkategori point range , namely : Results obtained division number 3 , this figure
is used as the value for the distance from point range perkategori. As for determining the point perkategori
ranges are as follows : 1. Most Valuable Customer that is taken from the
highest point total is 12 points . Because of the distance range is 3 , then made a point range of 10-
12. 2. Most growable Customer that is taken from the
highest point total is reduced lowest point total is 12- 3 = 9 points. Because of the distance range is 3 ,
then made a point range 7-9. 3. Migrator is taken from the second-highest total
point total is reduced lowest point ie 9-3 = 6 points . Because of the distance range is 3 , then made a
point range 4-6. 4. Below Zeros are taken from the lowest point total
of 3 points . At the customer grouping , there are 4 categories of
customer groups , while the pre-determined point range can be seen in Table 4 .
Table 4. Range Point Category Customer
From the results point above can perform grouping customers based on the ratings given in
accordance with the provisions of RFM is most valuable customer , most growable customer , zeros
and migratory below in Table 4. The results of the customer groupings can be seen in Table 5.
Table 5. Results Grouping Customers
5. The business strategy or service according to customer groups category
After segmenting customer data PT . Angga Media means using RFM analysis , it will be applied
to business strategy or the services that are tailored to specific customer categories . The strategy will be