Artificial Intelligence Algorithms in Financial Diagnostics and Strategic Sales Management in B2B Retail

Abstract

This paper explores the strategic application of artificial intelligence (AI) algorithms in financial management and sales planning within the retail sector under conditions of high uncertainty. The study focuses on the B2B segment of household appliances and addresses the limitations of traditional financial forecasting models in volatile economic environments. A comparative empirical analysis was conducted based on the implementation of AI-driven systems in Beko/Whirlpool operations across Central and Eastern Europe (Ukraine, Poland, Bulgaria), highlighting measurable improvements in key financial indicators. Specifically, AI integration led to a 41% reduction in overdue receivables, increased margin rates by 3.4–5.1 percentage points, and improved sales forecast accuracy up to 93%. The research employs advanced machine learning models—including XGBoost and LSTM—and benchmarks them against traditional methods such as ARIMA and linear regression using MAE, RMSE, and MAPE metrics. The findings demonstrate the advantages of AI in managing accounts receivable, dynamic pricing, and client scoring, as well as in streamlining decision-making through real-time BI dashboards. The paper also outlines the evolving role of Key Account Managers (KAMs) as data-driven financial analysts. Practical recommendations are offered for companies seeking to implement AI-based controlling systems, with emphasis on adaptability, operational transparency, and KPI optimization. The study contributes to the discourse on financial digitalization by offering a scalable model for integrating predictive AI into retail finance under macroeconomic turbulence.