​
VOC Insight MCP

Turn consumer conversations into actionable insights.

Analyze multi-platform discussions to generate AI-powered VOC, sentiment, and competitor intelligence.

VOC Insight MCP helps teams understand how consumers discuss a brand, product, topic, or competitor across supported public social and review platforms. It organizes collection results into voice, sentiment, platform, competitor, risk, and evidence views that an AI agent can query.

​
What It Helps With

  • Brand and product VOC analysis
  • Positive and negative feedback discovery
  • Sentiment trend and driver analysis
  • Competitor voice-share comparison
  • Platform-level risk monitoring
  • Evidence post search and semantic retrieval
  • AI deep follow-up queries after collection completes
  • Persona generation, section drill-down, and simulated interviews

​
Platform and Compliance Scope

VOC Insight MCP is intended for compliant public social and review data analysis for global brands outside mainland China1. It does not support data collection or analysis from any China-based social media platform.

Default coverage includes TikTok, Instagram, YouTube, X, Facebook, Pinterest, and Trustpilot. Threads and Reddit can be added when needed. Amazon Reviews is not part of the default knowledge-space social scope; to analyze Amazon reviews, use the full brand onboarding flow with ASINs or Amazon product URLs.

​
How Reports Are Generated

VOC Insight MCP first collects data, then generates metrics, post lists, sentiment trends, competitor comparisons, and supports report-based deep follow-up from the collected dataset. AI deep follow-up happens after collection is complete; it is not part of the collection process itself.

When the user asks who the customers are or why they buy or churn, use the persona capability. Start with the free get_persona_voc overview, then call get_persona_section only for sections listed in availableSections; start generate_persona_voc only when no report exists. Generation is asynchronous, so check get_persona_generation_progress. ask_persona and ask_all_personas provide persona-based simulated answers and mark the result as simulated; they are not real user testimony.

💡 Usage tip: VOC Insight MCP is built for complete AI brand-insight analysis, not one-off data queries. Every analysis collects and processes a large volume of public data, so a single task typically consumes a significant number of points. See the Pricing section below for details.

​
Apply for VOC API Access

🔌 If you want to integrate VOC data collection into your own application or data analysis system, get in touch:

apply for VOC API access

​
Notes

1 This social analysis service is intended only for compliant public social-media and review-data analysis for global brands in markets outside mainland China, for the purpose of supporting global brand and product growth research. Customers are responsible for ensuring their use complies with applicable laws, regulations, and platform rules. Pangolinfo provides only the necessary data access and analysis capabilities and is not responsible for third-party data content or the results of reports generated from such data.