Objective and Scope:

The client sought to analyse videos posted on social media platforms by customers to assess customer sentiment, track perception shifts, identify reputational risks, and gauge key experience drivers. The study focused on:

• Tracking sentiment shifts across specific time periods and key events
• Identify core themes and emerging reputation risks
• Recommend actions to strengthen customer experience and brand trust

Approach:

We leveraged a structured, AI-enabled video analytics framework integrating Large Language Models (LLMs), Natural Language Processing (NLP), and Computer Vision. Over 2,000+ shorts & videos posted on different social media platforms were filtered and analyzed to extract relevant customer experience content. Multimodal analysis was conducted across text (titles, descriptions, transcripts), audio (tone and sentiment), and visual elements (facial expressions, environment recognition, on-screen text). The data was classified into thematic clusters and sentiment categories across different time periods, enabling trend analysis and identification of key reputation drivers.

Impact:

The solution enabled the client to gain structured visibility into dynamic competitor activity through:

• Centralized intelligence engine delivering a categorized, searchable repository of competitor announcements and market movements
• Improved benchmarking of competitor strategies by tracking acquisitions, partnerships, product launches, and regulatory updates in a structured manner
• Accelerated insight-led decision-making by reducing manual monitoring efforts and transforming unstructured news into actionable strategic signals