The Hidden Costs of Data Quality - Tackling Common Data Challenges in ML
summary
Claims a concrete ROI from fixing data: 20% missing emails found and filled, customer statuses reconciled with the CRM, and standardized formats. Uses open source tooling—pandas and dask, Great Expectations and Airflow—for cleaning, validation and pipelines. Results show measurable lifts: 3%→10% CTR, +35% conversions, −15% unsubscribes. Presents specific steps: assess, deduplicate, impute, enrich, validate, then retrain, and reports that those changes materially fix targeting and drive revenue.
AI-written summary · about
chapters (8)
- 0:00Welcome
- 1:26Data quality in ML
- 4:16Costs of poor data quality
- 10:57ML data challenges
- 16:18Data quality tools
- 18:44Case study: better data quality
- 26:32Best practices for data quality
- 29:44Q&A
related talks
49mData Cleansing & Beyond: How to Address Data Debt for AI
44mKeep it Clean: Why Bad Data Ruins Projects and How to Fix it
49mKeep it Clean: Why Bad Data Ruins Projects and How to Fix it
26mThe Hidden Foundation: How Effective Data Management Determines AI System Success
32mMaking Models Work: How an optimized ML lifecycle can drive better ROI
40mThe Hidden Cost of Data Debt
39mDoing Data: The Critical Process of Data Preparation (...)
11mTransforming Problem-Solving with Data-Driven Insights