Monitoring SDG 11 targets is crucial for making informed decisions and supporting multidimensional transitions in European cities. Among all the goals, SDG 11 emerges as a cornerstone for cities, offering a comprehensive framework to tackle their multifaceted challenges. Composite indicators and indices, as suited evaluation tools to monitor city progress or decline, allow sustainability problems to be included in local agendas by aggregating multi-dimensional variables at different time spans through data-driven approaches. The primary concerns about using indicators as evaluation tools to compare performances are inherent to inconsistencies related to different assessment frameworks and methods, data downscaling from global to local levels, choice of aggregation rules to obtain synthetic results, and data gaps. This contribution, in particular, focuses on data gaps by elaborating on a testing case, while critically discussing related issues. The research was addressed to identify normative, assessment, and methodological gaps in monitoring progress towards SDG 11 at global, European, and Italian levels. Application of Machine Learning algorithms to predict null values within an SDG 11 regional dataset was implemented to compare three Italian regions according to 18 common indicators. The contribution is part of the Research Project of National Relevance “GLOSSA - GLOcal knowledge System for Sustainability Assessment of urban projects”, coordinated by Polytechnic of Turin (Italy), and it supports its first-step knowledge phase aimed at identifying gaps in SDG 11 indicators downscaling and monitoring.