The study is oriented toward a comprehensive understanding of how artificial intelligence (AI) technologies in 2024–2025 are redefining the substance and operating logic of commercial functions. The object of inquiry is a qualitative shift from fragmented initiatives that automate isolated operations toward an end-to-end, systemically organized integration of agentic AI and hybrid forecasting models that provide continuous decision support and coordinate commercial processes. The empirical–theoretical foundation is built on a comparison of the academic corpus indexed in Scopus/WoS with analytical materials produced by leading consulting organizations (Gartner, McKinsey, Deloitte). Within this approach, the analysis clarifies how AI assistants influence key performance indicators (KPIs), including sales conversion, customer retention rate, the dynamics of return reduction, and aggregated financial impact expressed through total revenue. Particular attention is given to causal mechanisms linking the deployment of intelligent assistive layers to changes in efficiency metrics, interpreted not in isolation but through their operational interdependence. A dedicated layer of analysis addresses neural architectures for demand forecasting, including compositional solutions in the LSTM-mTrans-MLP class, treated as an instrument for improving forecast robustness under nonlinear demand patterns and high data variability. In parallel, the methodological contour for detecting process dysfunctions and losses via Process Mining is developed in detail, enabling reconstruction of actual execution trajectories for commercial procedures, identification of deviations from target models, and formalization of inefficiency zones at the level of event logs and the organizational logic associated with them.
Daryna Oliinyk (Thu,) studied this question.