View in EDS

Customer segmentation for private market investments: exploring investment behaviour to develop user profiles for Finexity

Saved in:
Bibliographic Details
Title: Customer segmentation for private market investments: exploring investment behaviour to develop user profiles for Finexity
Authors: Schütter, Theresa
Contributors: Rongjiao, Ji, RUN
Publication Year: 2025
Subject Terms: Data science, Cluster analysis, Behavioural analysis, User profiling, K-Means, Hierarchical clustering, DBSCAN, UMAP, High-dimensional data, Private market investments, Tokenization, Investment platform, Real world assets, Domínio/Área Científica::Ciências Sociais::Economia e Gestão
Description: The rise of private market investments, fueled by blockchain-enabled tokenisation, represents a new trend in the financial sector. Thereby, fintech platforms offering these investments must effectively segment investors into distinct groups to retain them and finance respective assets. To identify and analyse customer segments and pinpoint premium investors, clustering methods k-means, hierarchical clustering, and DBSCAN were applied to four feature sets: demographic, behavioural, combined demographic-behavioural, and UMAP-transformed combined features. While behavioural features produced the most interpretable results, UMAP transformed combined features delivered the most accurate segmentation. These findings provide actionable insights for implementing tailored marketing strategies for each customer segment.
Contents Note: TID:203927680
File Description: application/pdf
Language: English
Availability: http://hdl.handle.net/10362/181483
Rights: embargoed access
Accession Number: rcaap.com.unl.run.unl.pt.10362.181483
Database: RCAAP
Description
Abstract:The rise of private market investments, fueled by blockchain-enabled tokenisation, represents a new trend in the financial sector. Thereby, fintech platforms offering these investments must effectively segment investors into distinct groups to retain them and finance respective assets. To identify and analyse customer segments and pinpoint premium investors, clustering methods k-means, hierarchical clustering, and DBSCAN were applied to four feature sets: demographic, behavioural, combined demographic-behavioural, and UMAP-transformed combined features. While behavioural features produced the most interpretable results, UMAP transformed combined features delivered the most accurate segmentation. These findings provide actionable insights for implementing tailored marketing strategies for each customer segment.