Overcoming the limitations of social media analytics: Dynamic capabilities and workaround strategies for inclusive and resilient public service continuity

This study addresses the limitations of using social media analytics in disaster management, arguing that human analysis remains indispensable, especially when dealing with vulnerable populations. While algorithms are powerful, most meaningful insights still come from people — through context, judgment, and trust-based engagement. Effective disaster management requires tools and approaches that include all groups in society, particularly when relying on data and artificial intelligence. Social media analytics are widely used to understand public needs during crises, but they tend to oversimplify what people are saying, missing important differences such as age, language, occupation, and community background. This becomes a serious issue during emergencies, when public organizations need to understand and respond to the needs of diverse populations: relying too heavily on aggregated data can lead to vulnerable groups — such as linguistic minorities, the elderly, or young people — being overlooked.

Drawing on interviews with public officials and software developers, the study finds that current analytics tools do not automatically detect or highlight the needs of these vulnerable groups in social media conversations. As a result, public organizations often develop workarounds and flexible strategies to better understand and engage with these populations, helping ensure that a wider range of voices is heard and that services remain effective during crises. From this follow the study's key implications: human analysis remains relevant when dealing with uncertainties; social media analytics tools need to be redesigned to better capture diverse community needs; crisis communication strategies should actively account for vulnerable and underrepresented groups; and public organizations benefit from being flexible and innovative when standard tools fall short. Overall, the study highlights the importance of combining technology with human judgment and adaptability to build more inclusive and resilient disaster management systems.