Discovering insights from customer surveys at McDonald’s

on YouTube ↗22m

" summary

McDonald’s pipelines millions of survey and social comments through S3 into Amazon Comprehend, uses Coreferencing to resolve pronouns and Sub-comment Breakdown to split mixed sentiments, and augments Comprehend output with in‑house classifiers before loading results into RedShift. The system surfaces topic tags and net sentiment per sub-comment, reveals issues (drive‑thru slowdowns tied to special orders and incorrect sauces), powers dashboards for CSAT and promo performance, and scales across languages.

AI-written summary · about

" chapters (8)

1. 0:00Introduction
2. 0:47Amazon Comprehend overview
3. 9:34McDonald's text analytics journey
4. 13:10Handling mixed sentiment analysis
5. 14:37McDonald's architecture details
6. 17:49Visualizing customer insights
7. 18:57Promotional insights use case
8. 20:47Next steps and resources

" related talks

• 33mRAG + Apache Iceberg™: Structured Retrieval Made EasyVino Duraisamy, Josh Reini · AI By the Bay 2025
• 25mRevolutionizing Customer Experience Through AI & DataMohan Krishna Mannava · Conf42 Machine Learning 2025
• 20mAI-Powered Customer Insights for Financial ServicesNeha Thakur · Conf42 Machine Learning 2025
• 43mConsumer Intelligence and the Age of Predictive Marketing AnalyticsRami Deeb · Digital Marketing Europe 2022
• 08mAI-Driven Innovation for Enhanced Food Delivery OperationsPraveen Payili · Conf42 Cloud Native 2025
• 33mMaximize lifetime value of media contentEsther Lee, James Jameson · AWS Machine Learning Summit 2021
• 30mMobile Order Click-Through Rate (CTR) Recommendation with Ray on Apache Spark at Burger KingLuyang Wang · Ray Summit 2021
• 52mSerg Masís | AI Track | Compass AI & Data Summit 2025Serg Masís · Compass AI & Data Summit 2025