Marketing Data Science
summary
Claims that marketing must become data driven and shows a telecom case where a model using 304 features yields 13× conversion lift versus traditional selection. Demonstrates tooling and pipelines: IBM Watson personality API, tone analyzer and NLU feeding Jupyter dataframes, clustering (PCA visualization) from a 30–50 customers demo. Cites industry numbers (about 25% ad impressions blocked, ~85% of purchase journey self-serve) and argues for incremental agile adoption of models and marketing automation APIs.
AI-written summary · about
chapters (16)
- 0:00Marketing data science intro
- 1:30Marketing vs data science clash
- 3:17Market value and acquisition
- 5:29Shift to buyer control
- 7:23Evolving marketing strategies
- 9:39Data beats experience
- 12:12Demo: Segmenting with Watson
- 18:29Clustering by personality
- 24:35AI for personalized campaigns
- 25:39What is marketing data science?
- 27:04Digital transformation
- 29:35Future of marketing data science
- 30:40Q&A: AI tools adoption
- 34:19Companies' data-driven reactions
- 36:09Active vs passive data collection
- 38:18Q&A: Data sources and agile
related talks
27mUnlocking Product Growth With Big Data, Data Science And Machine Learning
11mDrug Launch Success Through Prompt-Driven Sales Data Analytics
27mData storytelling in the era of generative AI
40mBridging bytes and behavior with Data Science and UX Research
24mStorytelling With Data
37mLife of a data scientist
27mEmpowering Business to make model-based marketing decisions
19mData science demystified