Τμηματοποίηση πελατών στις υπηρεσίες του τουρισμού

Ιακωβίδη, Μαργαρίτα

Customer segmentation in the tourism services sector

Master Thesis

Author

Ιακωβίδη, Μαργαρίτα

Date

2026-01

Abstract

Customer segmentation is a cornerstone of modern marketing and business analytics, enabling firms to better understand differences in customer behavior and value and to design interventions with greater precision and stronger evidence. By forming customer groups with shared characteristics, organizations can improve the effectiveness of marketing actions, allocate resources more rationally, and enhance customer experience and loyalty. Within this context, the RFM model is widely used as a rule-based scoring scheme that captures three key dimensions of transactional behavior: how recent the last transaction or visit is (Recency), how often transactions occur (Frequency), and the overall monetary value associated with the customer (Monetary). Each customer is scored on these dimensions using predefined scoring rules and, based on the resulting score combination, is assigned to a segment with clear managerial meaning, supporting decision-making and the systematic management of customer relationships. This study examines the use of the RFM model in the context of tourism services, with a particular focus on hotel businesses in Greece. The analysis indicates that, in this setting, the standard RFM scheme is not sufficient, as it does not consistently yield segments with stable and unambiguous managerial interpretation. In addition, strong seasonality directly affects the meaning of Recency, since the time elapsed since the last stay often reflects the timing of the visit within the tourist season rather than an actual intention to return. Moreover, the Monetary dimension in standard RFM captures primarily gross revenue and may not reflect net contribution, as it does not account for key drivers such as distribution-channel costs and the operational burden associated with servicing the stay. A key limitation is linked to the Frequency dimension as traditionally defined, which is not always well suited to capturing customer value and behavior under conditions of strong seasonality, where service consumption is expressed not only through repeat visits but also through the length of stay. For this reason, an adapted RML scheme is proposed, in which Frequency is replaced by Length of Stay (L), so that segmentation better reflects the characteristics of hotel consumption and provides more actionable support for managerial decisions. The proposed scheme was applied to data from a real high-end hotel property in Greece. Following data collection, cleaning, and the construction of a consistent final dataset, quantitative clustering methods were employed, visualization techniques (tree diagrams) were used to aid interpretation, and a composite performance indicator was developed to capture each customer’s overall contribution. This process resulted in four distinct segments with clear operational profiles: (a) customers with older visits, shorter stays, and lower spending, forming a large-volume but lower-average-value segment; (b) customers of medium to higher value with medium-length stays, not necessarily recent but constituting the core of demand; (c) recent customers with mid-range spending and an average length of stay, representing the main body of recent visits; and (d) high-value customers with longer stays, whose recency varies but who account for a disproportionately strong contribution to overall performance and capacity utilization. Overall, the empirical findings show that, in the setting examined, customer differentiation is not meaningfully captured through the “frequency” of repeat visits, as return behavior is limited. Segmentation becomes clearer when it emphasizes the recency of the relationship, total economic contribution, and length of stay—factors that align directly with value creation in small, seasonal tourism businesses. In particular, length of stay emerges as a critical dimension for both understanding demand and translating segmentation into managerial insight, since the customer–property relationship is expressed primarily within the stay rather than through repeat purchasing. Finally, clustering results were complemented by a unified performance indicator that enables comparable customer ranking and focuses attention on the highest-performing customers, offering a practical basis for management directions while respecting the constraints of the available data.

Postgraduate Studies Programme

Οικονομική και Επιχειρησιακή Στρατηγική

Department

Σχολή Οικονομικών, Επιχειρηματικών και Διεθνών Σπουδών. Τμήμα Οικονομικής Επιστήμης

Number of pages

151

Language

Greek

URI

https://dione.lib.unipi.gr/xmlui/handle/unipi/18906

Collections

Τμήμα Οικονομικής Επιστήμης

Show full item record

Except where otherwise noted, this item's license is described as
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα