Delays in product delivery or failure to promptly meet demand during the sales process negatively impact economic efficiency. This article explores the use of time series analysis and machine learning algorithms to detect delays in sales. Based on data analysis, the study identifies patterns, temporal dependencies, and other influencing factors of such delays. ARIMA, SARIMA, and LSTM models are utilized and their performance metrics are compared. The research results provide opportunities to forecast potential delays and minimize their impact on business processes.