Evaluating the impact of TikTok short-video advertisements on impulse buying behavior among consumers in Türkiye and Pakistan: an extended UTAUT2 perspective

Purpose This study explains when acceptance of TikTok short-video advertising becomes impulse buying by separating planned advertising engagement from spontaneous purchasing and testing whether the pathways are robust across Türkiye and Pakistan. Design/methodology/approach Survey data from 529 TikTok users in Türkiye (n = 276) and Pakistan (n = 253) were analyzed using partial least squares structural equation modeling to assess the measurement and structural models and compare the proposed relationships across the two countries. Findings Habit was the strongest predictor of behavioral intention and affected impulse buying both directly and indirectly. Performance expectancy, hedonic motivation, and price value operated through engagement intention, whereas facilitating conditions predicted impulse buying directly. The effort-expectancy and price-value paths differed between countries. Comparisons involving non-invariant composites were treated as exploratory. Practical implications Marketers should combine repeated, useful, entertaining, and value-explicit content with reliable links, local payment compatibility, transparent total cost, and clear delivery and return information. Platforms should pair frictionless purchasing with disclosure, spending controls, and meaningful confirmation to protect deliberation. Originality/value The study extends Unified Theory of Acceptance and Use of Technology 2 from continued platform engagement to an unplanned consumption outcome. It distinguishes an exposure pathway, in which usefulness, enjoyment, price value, and habit operate through continued advertising engagement, from execution pathways, in which habit and facilitating conditions retain direct associations with impulse buying. The invariance-qualified comparison further identifies where these relationships remain stable and where their strength is context dependent.