Optimizing the spatial distribution of marketers using GIS analytics

ABSTRACT
Marketing is a critical factor in the success of urban businesses, contributing significantly to commercial growth and competitive advantage. Production and distribution companies (such as those in the food, cosmetics, hygiene, detergent, and pharmaceutical industries) are increasingly seeking to reduce the costs associated with personnel and the travel of their marketing agents across urban areas. Optimal spatial distribution of marketers plays a vital role in minimizing human resource waste, time, and expenses, while enhancing the effectiveness of marketing activities. This study aims to optimize the spatial allocation of marketing agents using network analysis within a Geographic Information System (GIS) framework. To achieve this, a GIS-based network analysis method is proposed for clustering marketing territories based on the number and geographic distribution of retail outlets. The proposed method partitions retail locations into a set of geographical clusters in which the number of outlets is as evenly distributed as possible. Each cluster is then assigned to a specific marketing agent. The implementation of this method in District 6 of Tehran demonstrates that the resulting clusters and marketing visit routes within each cluster are optimized, with minimal boundary fragmentation—facilitating easier recognition by marketers and reducing operational errors. In other words, the proposed approach enables the formation of marketing clusters with an equal number of retail outlets and allows for the shortest possible intra-cluster visit routes.
Extended Abstract
Introduction
In today's world, marketing is recognized as one of the most important and prominent factors in the success of organizations or businesses. Marketing refers to communicating with customers, analyzing the market, designing advertising and sales strategies, attracting the attention of the audience, and creating an emotional connection with them. The importance of marketing is significant in various ways; including business growth and development, creating competition, identifying customer needs and preferences, and improving interaction and establishing effective communication with them. Manufacturing and distribution companies (such as food, cosmetics, detergents, pharmaceuticals, etc.) are looking to reduce human resources costs and marketing time. Since marketers are the main link between retailers and manufacturing and distribution companies, proper management and control of their performance will reduce distribution costs and, as a result, customer (retail) satisfaction. Meanwhile, the distribution of marketers in the region depends on the location of the retailers, so that there must be a balance between the number of marketers and retailers. The conventional method of assigning marketers to each region is to divide or cluster the retailers based on geographical areas and then assign each area to a marketer. In the past, distribution companies manually marked out territories on large paper maps, but today, various technologies have been developed that facilitate the management of marketers in providing services to customers and increase their efficiency.
Today, location-based marketing methods, utilizing concepts and tools based on Geographic Information Systems (GIS), have enabled the management, planning, and implementation of optimal marketing operations based on the geographic location of consumers and retailers. GIS can be considered as a computer-based information system consisting of hardware, software, people, data, methods, and models that is used to collect and store spatial and descriptive data, manage and display data, query and answer questions, process and analyze data, and generate geographic information and display this information. GIS is capable of integrating various types of spatial data and related descriptive data using analytical tools or spatial reporting to produce useful information to solve marketing problems. This information can help planners, decision makers, and managers of large and small businesses in the areas of sales territory allocation, geographic distribution of marketers, etc. In other words, spatial and descriptive data analysis related to the main components of an urban business system, including customers, competitors, and markets, is performed with greater accuracy, and owners and managers of small and large businesses are able to analyze and manage large volumes of information such as geographic location, customer addresses, range, and marketing routes in an area.
 
Methodology
One type of marketing method is location-based marketing. Location-based marketing uses the concepts and technologies of location information to manage, plan, and execute optimal marketing operations based on the geographic location of consumers. Jelokhani et al. (2019) state that GIS, as a powerful, popular, and cost-effective location-based tool, can play an important role in managing marketing activities such as market segmentation and spatial distribution of marketers. Accordingly, marketers can classify their company's market geographically based on various criteria and then determine their target market based on the strengths and weaknesses of their products and services.
The first step is to prepare the main roads and build the road network of the study area. In this step, using the capabilities of GIS network analysis, the road network has been created. In order to cluster retail locations and zone them, it is necessary to extract micro-areas from the street network. These micro-areas are the basis for creating larger geographic areas (clusters) for allocation to marketers. To determine the number of retail outlets in micro-areas, spatial connectivity is required. In other words, this analysis identifies the retail stores in each micro-area separately by spatially overlapping retail locations with micro-areas. Then, the center of gravity of each micro-area is determined and the retail stores within each micro-area are connected to that point. In order to aggregate micro-areas and extract larger geographic areas based on the number of retail stores, a regular grid of points was created. The average distances of the points of this network from the centers of gravity of the micro-areas were calculated through spatial allocation analysis, and based on this, and considering the number of marketers required, the center of the clusters was determined. It should be noted that during spatial allocation analysis, clusters are optimized based on a specific threshold. Ultimately, each micro-area is assigned to a cluster, and each cluster is assigned to a marketer.
 
Results and discussion
The micro-areas are grouped together to form clusters based on the minimum distance from the center. The algorithm ends where the sum of these distances is the minimum and the number of retail stores is approximately equal. Clusters represent geographic areas designated for marketers. Retail stores within each cluster are then assigned to a marketer. To quantitatively evaluate the proposed algorithm, the parameters of path lengths, standard deviation of path lengths in clusters, cluster compactness, and the number of fractures at the cluster boundary were used. The paths between points in each cluster were run and calculated separately for each method. A visual method was used to count the number of fractures, so that firstly, the outer clusters were not considered, and secondly, whenever the angle change within the cluster was more than 30 degrees, it was counted as a fracture. Obviously, the greater the number of clusters, the shorter the path lengths will be.
 
Conclusion
The results demonstrated that the proposed method outperforms existing approaches in several key aspects, including minimizing travel distance, reducing boundary fragmentation and discontinuity, and maximizing compactness. The reduced fragmentation along boundaries leads to improved accuracy in identifying and delineating marketing area borders. It should be noted, however, that this algorithm was applied only to District 6 of Tehran, and its performance may vary in other districts or cities with different road networks and urban structures.
 
Funding
There is no funding support.
 
Authors’ Contribution
Authors contributed equally to the conceptualization and writing of the article. All of the authors approved thecontent of the manuscript and agreed on all aspects of the work declaration of competing interest none.
 
Conflict of Interest
Authors declared no conflict of interest.
 
Acknowledgments
The authors would like to express their sincere gratitude and appreciation to all scientific advisors and contributors who participated in the research process.