DEPENDENCE OF ADVERTISING PERFORMANCE IN THE DIGITAL ENVIRONMENT ON METEOROLOGICAL FACTORS

Abstract

The article is devoted to examining the dependence of targeted advertising performance in the digital environment on meteorological factors as external situational determinants of consumer behavior. The relevance of the topic is driven by the growing role of behavioral analytics and the need to incorporate exogenous variables into the evaluation of digital marketing campaign effectiveness, particularly those operating in real time. The empirical basis consists of daily aggregated advertising activity data for Kyiv covering the period from May 1 to May 29, 2025, obtained from Meta Ads Manager and the CRM system of an online store, combined with official meteorological data. The methodological framework is based on correlation analysis of paired samples, which made it possible to quantitatively assess the direction and strength of linear relationships between variables. An anonymous survey of respondents regarding their subjective perception of the impact of weather conditions on online activity and purchase decisions served as an additional interpretative component. The findings indicate the presence of noticeable associations between weather parameters and selected advertising KPIs. In particular, a stable positive relationship was identified between temperature and the number of purchases, suggesting increased conversion activity on warmer days. At the same time, precipitation demonstrates a moderate positive correlation with the number of link clicks, which may reflect higher online engagement under conditions of reduced mobility. A negative relationship between temperature and link CTR suggests a potential decline in the proportion of click-throughs in warmer weather, despite an increase in absolute conversion figures. This divergence is interpreted as evidence of funnel-stage heterogeneity in user responses to situational factors. The results are exploratory in nature and provide a foundation for further research using extended time series, multifactor modeling, and cross-regional comparisons.