Search results

1 – 4 of 4
Article
Publication date: 27 November 2017

Serhat Peker, Altan Kocyigit and P. Erhan Eren

Predicting customers’ purchase behaviors is a challenging task. The literature has introduced the individual-level and the segment-based predictive modeling approaches for this…

1189

Abstract

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.

Article
Publication date: 6 May 2017

Serhat Peker, Altan Kocyigit and P. Erhan Eren

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…

4038

Abstract

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.

Details

Marketing Intelligence & Planning, vol. 35 no. 4
Type: Research Article
ISSN: 0263-4503

Keywords

Article
Publication date: 24 January 2022

Şevval Seray Macakoğlu, Burcu Alakuş Çınar and Serhat Peker

In the recent years, the rapid growth of the grocery retailing industry has created a great heterogeneity in prices across sellers in the market. Online price comparison agents…

Abstract

Purpose

In the recent years, the rapid growth of the grocery retailing industry has created a great heterogeneity in prices across sellers in the market. Online price comparison agents which are key mechanisms to solve this problem by providing prices from different sellers. However, there are many sellers in the grocery industry do not offer online service, and so it is impossible to automatically retrieve price information from such grocery stores. In this manner, crowdsourcing can become an essential source of information by collecting current price data from shoppers. Therefore, this paper aims to propose Kiyaslio, a gamified mobile crowdsourcing application that provides price information of products from different grocery markets.

Design/methodology/approach

Kiyaslio has been developed through leveraging the power of crowdsourcing technology. Game elements have also been used to increase the willingness of users to contribute on price data entries. The proposed application is implemented using design science methodology, and it has been evaluated through usability testing by two well-known techniques which are the system usability scale and the net promoter score.

Findings

The results of the usability tests indicate that participants find Kiyaslio as functional, useful and easy to use. These findings prove its applicability and user acceptability.

Practical implications

The proposed platform supports crowd sourced data collection and could be effectively used as a tool to support shoppers to easily access current market product prices.

Originality/value

This paper presents a mobile application platform for tracking current prices in the grocery retail market whose strength is based on the crowdsourcing concept and incorporation of game elements.

Details

International Journal of Web Information Systems, vol. 18 no. 1
Type: Research Article
ISSN: 1744-0084

Keywords

Article
Publication date: 4 April 2016

Serhat Peker, Seyma Kucukozer-Cavdar and Kursat Cagiltay

The purpose of this paper is to statistically explore the relationship between web usability and web presence of the universities. As a case study, five Turkish universities in…

1799

Abstract

Purpose

The purpose of this paper is to statistically explore the relationship between web usability and web presence of the universities. As a case study, five Turkish universities in different rankings which were selected from Webometrics rankings were evaluated and compared.

Design/methodology/approach

Two different methods were employed for performing usability evaluation of the selected universities: a user testing was used to measure the user performance on the selected tasks and a questionnaire to assess the user satisfaction on the website use. Both usability evaluation methods were applied on the pre-determined tasks for each university by participation of 20 subjects. After the usability evaluation, the universities were ranked in terms of usability results and finally, the relationship between web usability and web presence of universities was statistically investigated by using Kendall’s rank correlation.

Findings

Several common usability problems which were asserted by related previous studies were identified at the end of usability evaluation of university websites. The usability results also revealed that selected Turkish university websites suffer from numerous usability problems. Further, a strong positive correlation (p < 0.05) between the usability of the university websites and their web presences were identified. Hence, the participants showed a higher success and satisfaction while performing the tasks on the university websites which have strong web presences.

Practical implications

The findings from this study have practical implications for universities. Correlation results showed that universities can improve their web usability by giving importance to their web presence volumes. Universities can estimate their web usability levels by investigating their web presence rankings and they can also raise their rankings in Webometrics ranking system by improving the usability of their websites. Moreover, university web developers can design more usable and more user-friendly websites by avoiding usability and design problems identified through usability evaluation.

Originality/value

Different from the prior research efforts focussing on usability of educational web pages, this study contributes to the growing literature by statistically exploring the relationship between web presence and web usability of universities. This study is also precious from the point of view that it is one of the first attempts to evaluate and compare usability levels of a set of universities’ websites from Turkey.

Details

Program, vol. 50 no. 2
Type: Research Article
ISSN: 0033-0337

Keywords

1 – 4 of 4