Skip to content
Go back

Profiling the User of Real Estate Tax Tech Market (Poster Award)

Published:  at  09:00 AM
... views

Conference & Awards: Kim, D., Kim, Y., Kang, G., Shin, J., Kim, D., & Moon, H. (2025). Profiling the User of Real Estate Tax Tech Market. Journal of The Korean Data Analysis Society, 27(1). Awarded the Poster Encouragement Award.

Profiling the User of Real Estate Tax Tech Market. 2025 Winter Conference of Journal of The Korean Data Analysis Society

Figure 1. Poster

As the tax tech market expands, understanding the exact behavioral profiles of digital service users becomes crucial for targeted marketing and service optimization.

📌 The Problem: Lack of Empirical User Data

Despite the rapid digital transformation in the tax sector (e.g., the National Tax Service’s ‘One-Click Service’), empirical research analyzing the actual users of private tax solution services remains severely limited. Without understanding who is using the service and how they transact, real estate tax tech companies struggle to provide customized experiences and optimize their marketing strategies.

⚙️ The Method: RFM and K-Means Clustering

To decode user behavior, this study analyzed nationwide real estate capital gains tax filing data alongside actual user data from a commercial tax solution (NEWEYE Corp.).

  1. Demographic Goodness-of-Fit: Compared the demographic distribution of nationwide filers with the service’s actual users.
  2. RFM Transformation: Converted users’ raw transaction history into Recency, Frequency, and Monetary (RFM) variables.
  3. Clustering: Applied the K-Means clustering method (optimized at k=5 via Silhouette score) to segment the diverse user base into distinct subgroups based on their behavioral patterns.

🚀 The Impact: Data-Driven Market Strategies

The analysis revealed significant demographic shifts and provided actionable business intelligence:

This research provides critical empirical evidence for tax tech companies to deploy data-driven, personalized service strategies optimized for each user segment’s specific needs.


Share this research on:

Previous Research
[Working Paper] Consumer Choice Modeling for Smart Home Demand Response Services
Next Research
Chatbot Design for Structuring User Input Data via LLM (Poster Award)