During the early years of the COVID-19 pandemic in Mexico, the lack of a structured system for collecting data on vaccine side effects highlighted a critical gap in the national healthcare infrastructure. Today, social media has become a key platform where citizens share their post-vaccination experiences, often reporting physical and emotional discomforts following vaccine administration. Collecting and analyzing this information enables the visualization of the spatiotemporal impact of post-vaccination health across various regions of Mexico City. The main challenge lies in extracting this data in real time, understanding the dynamics of terms used to describe vaccination effects on social media, and subsequently identifying the known side effects of different COVID-19 vaccines. The most commonly reported side effects on Twitter, such as “headache,” “fatigue,” “fever,” “muscle pain,” and “chills,” reflect consistency between public perceptions and clinical observations, particularly for vaccines like AstraZeneca and Sputnik-V. Notably, “fever” and “headache” were concentrated in the central and southern regions of Mexico City. Our study proposes a framework for analyzing narratives obtained from Twitter using Natural Language Processing (NLP), spatial analysis, and data mining techniques. This approach complements official vaccination campaign data and can help shape future strategies and health policies.