Paper Number
ICIS2025-1854
Paper Type
Complete
Abstract
Stakeholders increasingly question companies’ Environmental, Social and Governance (ESG) claims, exposing a gap between self-reported metrics and public perception. We present a design-science artifact – a real-time ESG validation framework that mines social-media discourse to supply organizations with stakeholder feedback. The modular pipeline combines domain-adaptive topic modeling with transformer-based sentiment and emotion classification. A multi-industry case study (IT-consulting, food-and-beverage, tobacco) demonstrates that the system captures sector-specific ESG themes and detects sentiment shifts following announcements, controversies or initiatives. Social media emerges as a sensitive barometer: positivity dominates in sectors aligned with sustainability goals, while tobacco discourse remains negative. Event-based temporal analysis and crowdsourced annotation confirm the framework’s accuracy. The study advances information systems research by offering (1) a replicable tool operationalizing social media for ESG accountability, (2) design principles uniting socio-technical relevance and machine learning, and (3) a methodological basis for examining when public discourse legitimizes or penalizes symbolic disclosure.
Recommended Citation
Mishra, Nidhi; Diener, Moritz; Heinz, Daniel; Spitzer, Philipp; and Cudré-Mauroux, Philippe, "Bridging Corporate Claims and Public Perception: Real-Time Validation of ESG Initiatives with Social-Media Analytics" (2025). ICIS 2025 Proceedings. 8.
https://aisel.aisnet.org/icis2025/da_bus/da_bus/8
Bridging Corporate Claims and Public Perception: Real-Time Validation of ESG Initiatives with Social-Media Analytics
Stakeholders increasingly question companies’ Environmental, Social and Governance (ESG) claims, exposing a gap between self-reported metrics and public perception. We present a design-science artifact – a real-time ESG validation framework that mines social-media discourse to supply organizations with stakeholder feedback. The modular pipeline combines domain-adaptive topic modeling with transformer-based sentiment and emotion classification. A multi-industry case study (IT-consulting, food-and-beverage, tobacco) demonstrates that the system captures sector-specific ESG themes and detects sentiment shifts following announcements, controversies or initiatives. Social media emerges as a sensitive barometer: positivity dominates in sectors aligned with sustainability goals, while tobacco discourse remains negative. Event-based temporal analysis and crowdsourced annotation confirm the framework’s accuracy. The study advances information systems research by offering (1) a replicable tool operationalizing social media for ESG accountability, (2) design principles uniting socio-technical relevance and machine learning, and (3) a methodological basis for examining when public discourse legitimizes or penalizes symbolic disclosure.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
07-DataAnalytics