The Expansion of Private Label Brands into the Premium Market in Turkey: The Case of File Market

Abstract

Purpose – This study focuses on BİM, one of Türkiye’s leading discount retail chains, and its premium-concept brand, File Market. The study aims to analyse the potential store expansion strategies of private label brands within the context of premium retailing by utilizing socioeconomic data and machine learning methods. In this context, the study seeks to estimate the potential number of stores for provinces where File Market is not yet present using a data-driven approach, and to identify the most suitable predictive models for this process.
Design/methodology/approach – In the study, socioeconomic indicators such as population, income, energy consumption, literacy, real estate prices, and vehicle ownership were collected for all 81 provinces and incorporated into the dataset along with the existing number of BİM and File Market branches. During the prediction process, Gradient Boosting, Linear Regression, Random Forest, Support Vector Machines, k-Nearest Neighbours, and Neural Network models were compared using the Orange Data Mining software. The performance of the models was evaluated based on MSE, RMSE, MAE, MAPE, and R² metrics.
Results – In line with the main objective of the study, the potential number of stores for provinces where File Market is not yet present was estimated using a data-driven approach. The analysis results indicate that Gradient Boosting and Linear Regression models exhibit higher accuracy and consistency compared to other methods. While the Linear Regression model provides strong explanatory power (R²=0.971), Gradient Boosting stands out for avoiding negative or meaningless predictions and for better capturing non-linear relationships. Both models have been identified as robust tools capable of providing strategic decision support in forecasting potential location numbers.
Discussion – The findings indicate that determining the potential number of stores in the retail sector through socioeconomic data and machine learning–based models can enhance the effectiveness of both operational and marketing strategies. The File Market case illustrates the potential of a premium positioning strategy to reach different income segments and strengthen brand image. This approach provides retailers with a competitive advantage by aligning the right customer profile with the ideal number of locations and contributes to sustainable growth.