The subject of our article is the pattern of companies’ behavior, being the result of two effects: the inertia effect (actions based on routines) and the prediction effect (actions based on expectations). The aim of the analysis was to identify and compare the behavior pattern of enterprises in periods of normal volatility and extraordinary instability of the business environment, visible in qualitative economic indicators. The econometric study was conducted for three groups of enterprises: small, medium-sized, and large in a period of normal volatility (01.2000 – 02.2020) and in a period of extraordinary instability (03.2020 – 06.2023). In the study we used indicators (diagnoses and forecasts) of the general economic situation of enterprises (GES). The basis for the conclusions was a multiplier analysis based on ARDL models. The results indicate that a long-run pattern of behavior reflecting the inertia and predictions effects on the assessment of GES existed only in groups of small and medium-sized enterprises during the period of normal volatility. In large companies, no pattern of behavior was identified in any of the analyzed periods.