The decline of commercial areas as communities, due to online shopping and global retail chains, poses a significant economic and social challenge. High vacancy rates in commercial areas are a growing concern in many countries, including the UK and Japan. Revitalizing these areas requires community-based management strategies. Consumer behavior analysis, particularly focusing on transaction data, offers a powerful tool for such revitalization. However, many traditional shopping streets, especially in Japan ("shotengai"), lack the digital infrastructure (like POS systems) necessary for collecting comprehensive consumer behavior data. This study aims to overcome this limitation by proposing a novel approach that leverages paper-based community currencies to gather cross-cutting consumer behavior data and inform effective community-based management strategies. The study seeks to answer how low-digitalized shopping communities can realize community-based management and how conventional retailing management methods using consumer behavior analysis can be extended to community-based management.

Existing literature highlights the importance of community-based management in revitalizing commercial areas. Studies have explored consumer behavior analysis using transaction data from individual stores, often utilizing POS systems. However, these approaches are unsuitable for low-digitalized communities. While methods like questionnaires and interviews exist for collecting community data, they often lack the direct link to purchasing behavior crucial for effective management. Community currencies have been used to promote local consumption and cohesion, but their application to consumer behavior analysis for community management remains largely unexplored. This study builds on this existing work by proposing a novel approach to data collection and analysis using paper-based community currencies, thereby bridging the gap between existing research and the needs of low-digitalized communities.

The study proposes a novel management approach using data collected through two types of paper-based community currencies: FT (from-to) and CA (customer attributes). FT currencies track the flow of currency between stores, revealing co-occurrence relationships between stores used by the same customer. CA currencies are linked to customer attributes (e.g., student, faculty), providing data on store usage patterns based on customer characteristics. Two field experiments were conducted in distinct Japanese shopping districts. **Study 1 (FT-type):** A four-week experiment in Tokyo's Shin-Okubo/Okubo area involved 17 out of 55 eligible stores. Customers received an FT currency for purchases exceeding 200 yen. Data from 498 transactions were collected. ABC analysis identified frequently visited stores, and association analysis revealed co-occurrence relationships between stores. **Study 2 (CA-type):** This experiment, conducted near Waseda University, used CA currencies distributed to students and faculty members based on their attributes. The experiment used the "Machi-Navi" application and a dedicated reception desk to link customers with their respective CA-type currencies. 338 transactions were collected over two separate four-week periods. ABC analysis was applied separately for student and faculty data, revealing stores frequently visited by each group. The same analysis was also used for aggregated data not considering the customer's attribute. Data analysis involved ABC analysis (classifying stores by visit frequency) and association analysis (identifying relationships between stores based on co-occurrence). The study extended these methods to the community level, moving beyond their traditional application to individual stores.

**Study 1 (FT-type):** ABC analysis identified stores 7, 10, and 12 as frequently visited. Association analysis revealed strong co-occurrence relationships between certain stores, suggesting potential synergies and collaboration opportunities. Store 12, a pharmacy, acted as a hub connecting other stores. **Study 2 (CA-type):** ABC analysis identified different sets of frequently visited stores for students and faculty members separately and also for aggregated data. This highlights the importance of customer segmentation for targeted community management strategies. Stores frequently visited by students and faculty differed, indicating distinct preferences and providing opportunities for tailored interventions. Both studies demonstrate that the proposed methods of data collection and analysis are viable in low-digitalized communities. The Pareto principle (80/20 rule) appeared to hold for Study 2 (CA-type), suggesting that a small proportion of stores account for a large proportion of transactions. Study 1 did not show the same result.

The findings demonstrate the feasibility of applying consumer behavior analysis methods, traditionally used for individual stores, to community-based management in low-digitalized settings. The use of paper-based community currencies provides a practical and cost-effective solution for data collection, overcoming the limitations of digital technology adoption. The identification of frequently visited stores and co-occurrence relationships between stores offers valuable insights for targeted interventions to boost community-wide economic activity. Customer segmentation, as revealed by the CA-type currency experiment, further enhances the precision of community management strategies. The varying validity of the Pareto Principle in the studies suggests further investigation in different contexts is warranted.

This study successfully demonstrates the application of consumer behavior analysis to community-based management in low-digitalized settings. The use of paper-based community currencies allows for effective data collection without requiring advanced digital infrastructure. ABC and association analyses revealed valuable insights for managing commercial areas by identifying key stores and relationships. Future research should explore the effectiveness of management interventions based on these findings, extend the analysis to other methods, and enhance data collection to include product-level information. This study provides valuable insights into community management, offering a practical and effective alternative for areas with limited digital infrastructure.

The study's limitations include the quality of data collected using the single-use paper-based currencies, restricting analysis to relationships between two elements (stores or store and customer attribute). The use of only ABC and association analyses limits the scope of potential insights, and the differing results regarding the Pareto Principle require further investigation. Furthermore, the small sample sizes of the field experiments might limit the generalizability of the findings.