Session: AI Powered Community Relations In App Management
summary
AppFollow builds an AI pipeline that ingests App Store, Google Play, social and email feedback, performs categorization, summarization and chat-with-reviews generation, and surfaces anomaly detection for fast product response. Real-world scale examples: Call of Duty Warzone ~240k reviews/month and WhatsApp ~300k in 30 days. Models used are model-agnostic but currently mostly OpenAI ChatGPT with some Claude tests. Hybrid automation raises effective reply rates from ~20% baseline to >40% (some titles near 100%), saving time and improving ratings and revenue.
AI-written summary · about
chapters (12)
- 0:00AI community relations intro
- 1:05Where communities live
- 3:57User reviews volume and impact
- 5:18AI sentiment analysis
- 6:01Review quality and hidden sentiment
- 7:51Reply rates and feature responses
- 9:19Response types and AI-human hybrid
- 10:46Future AI reply improvements
- 11:28One-to-many anomaly detection
- 12:31Measuring AI community management
- 13:58Key benefits of AI in community
- 15:01Q&A on AI models
related talks
20mAI & Automation Hacks For Ratings & Reviews
19mFrom Reviews to Ratings: Scaling Player Experience with AI Automation
18mSession: Player Value At Scale: Automating Replies That Actually Help
17mSession: Beat The Uninstall Button: How AI-Native Engagement Rewrites The Player Retention Playbook.
42mCurated Panel: A Practical Approach to AI Transformation Across Player Support
16mAI that boosts in-app purchases and monetisation
31mHow Agentic AI Is Reshaping Mobile App Quality
29mHow Agentic AI Is Reshaping Mobile App Quality