Od analizy do strategii: jak sztuczna inteligencja przekształca klasyczną analizę SWOT
From Analysis to Strategy: AI-Enhanced Transformation of Traditional SWOT

Summary/Abstract: Aim: To empirically describe how users of the SWOTmaker application progress through the stages of AI-agent–supported SWOT/TOWS analysis, and to operationalize and characterize the degree of process completion (understood as progression through successive stages E1–E4) based on telemetry data. Methodology: An exploratory study using telemetry data from real user sessions (2022–2025). Descriptive statistics and behavioral segmentation were applied (X – ended at E2, Y – at E3, Z – full E1–E4). Results: Most sessions end at E3 (TOWS strategy generation), while only a minority reach E4 (action recommendations). SO strategies dominate and users largely accept AI suggestions without edits. Implications and recommendations: The findings suggest a need to design AI-SWOT interfaces with mechanisms that mitigate anchoring and premature closure (e.g., presenting countertypes/alternative strategy variants) and to introduce E3→E4 “nudges” that support translating strategy into actionable steps. The X–Y–Z segmentation can serve as a simple indicator of process completion within the tool and as a starting point for further research on the quality of human-in-the-loop (HITL) interaction, including validation using qualitative methods or expert evaluation. Originality/value: This study provides the first empirical evidence on real user behaviors within an AIdriven TOWS system. It introduces a new process completion metric (X–Y–Z segmentation) and demonstrates how generative AI transforms the classical SWOT model into a dynamic, interactive cognitive process.

• Details
• Contents