Purpose: The purpose of this paper is to propose a new RFM model called length, recency, frequency, monetary and periodicity (LRFMP) for classifying customers in the grocery retail industry; and to identify different customer segments in this industry based on the proposed model. Design/methodology/approach: This study combines the LRFMP model and clustering for customer segmentation. Real-life data from a grocery chain operating in Turkey is used. Three cluster validation indices are used for optimizing the number of groups of customers and K-means algorithm is employed to cluster customers. First, attributes of the LRFMP model are extracted for each customer, and then based on LRFMP model features, customers are segmented into different customer groups. Finally, identified customer segments are profiled based on LRFMP characteristics and for each customer profile, unique CRM and marketing strategies are recommended. Findings: The results show that there are five different customer groups and based on LRFMP characteristics, they are profiled as: “high-contribution loyal customers,” “low-contribution loyal customers,” “uncertain customers,” “high-spending lost customers” and “low-spending lost customers.” Practical implications: This research may provide researchers and practitioners with a systematic guideline for effectively identifying different customer profiles based on the LRFMP model, give grocery companies useful insights about different customer profiles, and assist decision makers in developing effective customer relationships and unique marketing strategies, and further allocating resources efficiently. Originality/value: This study contributes to prior literature by proposing a new RFM model, called LRFMP for the customer segmentation and providing useful insights about behaviors of different customer types in the Turkish grocery industry. It is also precious from the point of view that it is one of the first attempts in the literature which investigates the customer segmentation in the grocery retail industry.
The ability to leverage data science can generate valuable insights and actions in organisations by enhancing data‐driven decision‐making to find optimal solutions based on complex business parameters and data. However, only a small percentage of the organisations can successfully obtain a business value from their investments due to a lack of organisational management, alignment, and culture. Becoming a data‐driven organisation requires an organisational change that should be managed and fostered from a holistic multidisciplinary perspective. Accordingly, this study seeks to address these problems by developing the Data Drivenness Process Capability Determination Model (DDPCDM) based on the ISO/IEC 330xx family of standards. The proposed model enables organisations to determine their current management capabilities, derivation of a gap analysis, and the creation of a comprehensive roadmap for improvement in a structured and standardised way. DDPCDM comprises two main dimensions: process and capability. The process dimension consists of five organisational management processes: change management, skill and talent management, strategic alignment, organisational learning, and sponsorship and portfolio management. The capability dimension embraces six levels, from incomplete to innovating. The applicability and usability of DDPCDM are also evaluated by conducting a multiple‐case study in two organisations. The results reveal that the proposed model is able to evaluate the strengths and weaknesses of an organisation in adopting, managing, and fostering the transition to a data‐driven organisation and providing a roadmap for continuously improving the data‐drivenness of organisations.
Purpose: Predicting customers’ purchase behaviors is a challenging task. The literature has introduced the individual-level and the segment-based predictive modeling approaches for this purpose. Each method has its own advantages and drawbacks, and performs in certain cases. The purpose of this paper is to propose a hybrid approach which predicts customers’ individual purchase behaviors and reduces the limitations of these two methods by combining the advantages of them. Design/methodology/approach: The proposed hybrid approach is established based on individual-level and segment-based approaches and utilizes the historical transactional data and predictive algorithms to generate predictions. The effectiveness of the proposed approach is experimentally evaluated in the domain of supermarket shopping by using real-world data and using five popular machine learning classification algorithms including logistic regression, decision trees, support vector machines, neural networks and random forests. Findings: A comparison of results shows that the proposed hybrid approach substantially outperforms the individual-level and the segment-based approaches in terms of prediction coverage while maintaining roughly comparable prediction accuracy to the individual-level method. Moreover, the experimental results demonstrate that logistic regression performs better than the other classifiers in predicting customer purchase behavior. Practical implications: The study concludes that the proposed approach would be beneficial for enterprises in terms of designing customized services and one-to-one marketing strategies. Originality/value: This study is the first attempt to adopt a hybrid approach combining individual-level and segment-based approaches to predict customers’ individual purchase behaviors.