Background: Technostress has become a significant concern in today’s technology-driven workplace, garnering increasing attention from organizational and health psychology. Several studies have shown its detrimental effects on workers’ well-being (e.g., burnout, perceived stress) and performance. The Technostress Creators Inventory (TCI) is one of the most used instruments to assess technostress. The original version of the scale consisted of five reflective indicators (i.e., techno-overload, techno-invasion, techno-complexity, techno-uncertainty, and techno-insecurity). However, its short version is frequently used, which comprises three dimensions (i.e., techno-overload, techno-invasion, and techno-complexity). Nevertheless, the TCI is characterized by some psychometric issues regarding: a) construct validity, as the dimensions of TCI often strongly correlated and they substantially overlap; b) criterion validity, since the TCI is often used as a general composite score or only specific subdimensions are scored; c) incremental validity, because the added value of the TCI dimensions in explaining the variance of relevant phenomena above and beyond other important predictors (e.g., workload) is not extensively investigated. Consistent with these premises, the present study aims to improve the factorial structure and the validity of the TCI, to better explain its specific dimensions and its effect on workers’ well-being and performance. Method: A sample of 778 Italian employees was utilized to compare two alternative confirmatory factor analysis (CFA) models, symmetrical bifactor and bifactor-(S-1) models to determine the best factorial structure of the short version of the TCI. Additionally, associations between these alternative models and anxiety, depression, perceived stress, burnout, and performance were examined to evaluate the criterion-related validity of the TCI. Results: Results suggested that the bifactor-(S-1) model with techno-overload as reference domain and two specific oblique factors showed the best fit to the data and good reliability and validity. Techno-overload as reference factor allows to better clarify the other two dimensions of the TCI (i.e., techno-complexity and techno-invasion) from both theoretical and practical standpoints. Additionally, our results show that the bifactor-(S-1) model was significantly related to burnout, perceived stress, anxiety, depression, and in-role performance, both concurrently and prospectively. Moreover, these relationships remained significant even after controlling for baseline quantitative workload level, gender, and job tenure. Conclusion: The bifactor-(S-1) model of short TCI can represent a promising tool for representing the factorial structure of technostress and increasing our knowledge regarding how techno-stressors may explain employee well-being and performance above and beyond other important predictors.