analisis rfm menggunakan k-means untuk segmentasi perilaku nasabah dari bank lokal di Indonesia = An rfm analysis using k-means for customer behaviour segmentation of a local bank in Indonesia

Pangouw, Yanry Yohan Pieter (2023) analisis rfm menggunakan k-means untuk segmentasi perilaku nasabah dari bank lokal di Indonesia = An rfm analysis using k-means for customer behaviour segmentation of a local bank in Indonesia. Masters thesis, Universitas Pelita Harapan.

[img]
Preview
Text (Title)
Title.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (458kB) | Preview
[img]
Preview
Text (Abstract)
Abstract.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (2MB) | Preview
[img]
Preview
Text (ToC)
ToC.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (3MB) | Preview
[img]
Preview
Text (Chapter1)
Chapter1.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (5MB) | Preview
[img] Text (Chapter2)
Chapter2.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (8MB)
[img] Text (Chapter3)
Chapter3.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (14MB)
[img] Text (Chapter4)
Chapter4.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (8MB)
[img] Text (Chapter5)
Chapter5.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (2MB)
[img]
Preview
Text (Bibliography)
Bilbliography.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (4MB) | Preview
[img] Text (Appendices)
Appendices (5).pdf
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (5MB)

Abstract

In the highly competitive financial industry, strategic customer segmentation is crucial for customer retention and profitability enhancement. This study aims to address this imperative at BCD Bank, a local bank in Indonesia. BCD Bank faces challenges in effectively segmenting its savings account holders due to the lack of nuanced understanding from traditional segmentation methods. This study utilised the Recency, Frequency, Monetary (RFM) model to assess transaction data from savings accounts and identify essential behavioural attributes. The application of K-means, a machine learning algorithm, further refined these segments, uncovering detailed behavioural trends. To ensure the statistical significance and interpretability of the formed clusters, the Silhouette Coefficient was integrated, yielding a solid validation with a score of 0.594423. The application of this approach resulted in the identification of four distinct customer groups: 'Potential', 'Passer-by', 'Prime', and 'At-risk'. 'Potential' customers, making up approximately 27% of the observations, showed a good amount of recency and frequency in their transactions, coupled with high monetary value. 'Passer-by' customers accounted for around 5.6% of the data and had lower scores across all RFM attributes. The 'Prime' customers were the largest segment, comprising around 56% of the observations, with the highest scores across all the RFM attributes. Finally, the 'At-risk' segment, accounting for roughly 10% of the customers, showed medium recency, low frequency, and medium monetary value. The findings suggest actionable strategies to address the characteristics of each customer group, potentially enhancing customer loyalty and profitability for BCD Bank. For instance, promotional offers could be targeted towards 'Potential' and 'Prime' customers to reinforce their loyalty, while re-engagement strategies could be implemented to bring 'Passer-by' and 'At-risk' customers back to the bank. / Dalam industri keuangan yang sangat kompetitif, segmentasi pelanggan yang strategis sangat penting untuk retensi pelanggan dan peningkatan profitabilitas. Studi ini bertujuan untuk mengatasi kebutuhan ini di Bank BCD, sebuah bank daerah di Indonesia. Bank BCD menghadapi tantangan dalam melakukan segmentasi efektif terhadap pemegang rekening tabungan karena kurangnya pemahaman nuansa dari metode segmentasi tradisional. Penelitian ini memanfaatkan model Recency, Frequency, Monetary (RFM) untuk menilai data transaksi dari rekening tabungan dan mengidentifikasi atribut perilaku yang penting. Aplikasi dari K-means, sebuah algoritma pembelajaran mesin, lebih lanjut menajamkan segmen-segmen ini, mengungkapkan tren perilaku yang detail. Untuk memastikan signifikansi statistik dan interpretasi dari klaster yang terbentuk, Koefisien Silhouette diintegrasikan, menghasilkan validasi yang solid dengan skor 0.594423. Penerapan pendekatan ini menghasilkan identifikasi empat grup pelanggan yang berbeda: 'Potential', 'Passer-by', 'Prime', dan 'At-risk'. Pelanggan 'Potential', yang mencakup sekitar 27% dari observasi, menunjukkan jumlah yang baik dari segi kedekatan dan frekuensi dalam transaksinya, ditambah dengan nilai moneter yang tinggi. Pelanggan 'Passer-by' mencakup sekitar 5,6% dari data dan memiliki skor yang lebih rendah di semua atribut RFM. Pelanggan 'Prime' adalah segmen terbesar, terdiri dari sekitar 56% dari observasi, dengan skor tertinggi di semua atribut RFM. Terakhir, segmen 'At-risk', yang mencakup sekitar 10% dari pelanggan, menunjukkan kedekatan sedang, frekuensi rendah, dan nilai moneter sedang. Temuan ini menyarankan strategi untuk mengatasi karakteristik dari setiap grup pelanggan, yang berpotensi meningkatkan loyalitas pelanggan dan profitabilitas untuk BCD Bank. Misalnya, penawaran promosi dapat ditujukan kepada pelanggan 'Potential' dan 'Prime' untuk memperkuat loyalitas mereka, sementara strategi penyelamatan pelanggan dapat diimplementasikan untuk membawa kembali pelanggan 'Passer-by' dan 'At-risk' ke bank.

Item Type: Thesis (Masters)
Creators:
CreatorsNIMEmail
Pangouw, Yanry Yohan PieterNIM0167121000301671210003@student.uph.edu
Contributors:
ContributionContributorsNIDN/NIDKEmail
Thesis advisorMurwantara, I MadeNIDN0302057304made.murwantara@uph.edu
Uncontrolled Keywords: RFM ; Customer Segmentation ; K-Means ; Silhouette Coefficient ; Customer Analytics ; Local Bank
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: University Subject > Current > Faculty/School - UPH Karawaci > School of Information Science and Technology > Master of Informatics
Current > Faculty/School - UPH Karawaci > School of Information Science and Technology > Master of Informatics
Depositing User: Users 29691 not found.
Date Deposited: 11 Aug 2023 09:35
Last Modified: 11 Aug 2023 09:35
URI: http://repository.uph.edu/id/eprint/57493

Actions (login required)

View Item View Item