CTO Data Strategy & Analytics 3 — Questions and Answers
Question 1: A CTO is evaluating whether to build an internal ML platform or use a managed MLOps service. The strongest argument for building internally is:
- Faster time to first model deployment
- Lower total cost of ownership in all cases
- Unique data security or compliance requirements that cloud vendors cannot meet (Correct answer)
- Easier recruiting of ML engineers
Correct answer: Unique data security or compliance requirements that cloud vendors cannot meet
Building internally is justified when regulatory, sovereignty, or security constraints prevent use of external managed services.
Question 2: What is data lineage and why is it critical for a CTO?
- The chronological order of database schema migrations
- A record of data's origin, movement, transformation, and consumption across systems (Correct answer)
- The versioning history of machine learning models
- A log of user access events for audit purposes
Correct answer: A record of data's origin, movement, transformation, and consumption across systems
Data lineage tracks the full lifecycle of data, enabling impact analysis, regulatory compliance, and debugging of data quality issues.
Question 3: An organization wants to monetize its data externally. Which governance concern is most critical?
- Ensuring the data warehouse has sufficient storage
- Obtaining proper consent and anonymizing PII before external sharing (Correct answer)
- Maximizing the volume of data shared to increase revenue
- Migrating all data to a public cloud provider first
Correct answer: Obtaining proper consent and anonymizing PII before external sharing
Monetizing data externally requires rigorous privacy protection, consent management, and PII removal to comply with regulations like CCPA and GDPR.
Question 4: Which KPI is most useful for tracking the health of an organization's data quality program?
- Number of data engineers on staff
- Percentage of critical data fields passing defined quality rules (Correct answer)
- Total terabytes of data stored
- Number of BI reports published monthly
Correct answer: Percentage of critical data fields passing defined quality rules
Tracking the percentage of data fields passing quality rules directly measures whether data meets the standards required for reliable decisions.
Question 5: A CTO is designing a data retention policy. Which factor should drive retention periods for operational data?
- Available storage capacity on current servers
- The longest retention period required by applicable regulations plus business need (Correct answer)
- The default retention setting of the database vendor
- The preference of the data engineering team
Correct answer: The longest retention period required by applicable regulations plus business need
Retention periods must satisfy the most stringent regulatory requirement that applies, combined with genuine business operational needs.
Question 6: What is the primary advantage of using a feature store in machine learning operations?
- It automatically trains and deploys ML models without human intervention
- It provides a centralized repository of reusable, versioned features for training and serving (Correct answer)
- It replaces the need for a data warehouse in analytics workflows
- It encrypts model weights to prevent intellectual property theft
Correct answer: It provides a centralized repository of reusable, versioned features for training and serving
A feature store standardizes and shares computed features across teams, preventing duplicate work and ensuring training-serving consistency.
Question 7: In a data strategy context, what does 'data democratization' mean?
- Allowing all employees to modify raw production data
- Making data accessible and understandable to non-technical business users through self-service tools (Correct answer)
- Storing all data in open-source systems only
- Distributing data storage evenly across geographic regions
Correct answer: Making data accessible and understandable to non-technical business users through self-service tools
Data democratization empowers business users to access, analyze, and derive insights from data without requiring engineering support for every query.
CTO Data Strategy & Analytics 3
A CTO is evaluating whether to build an internal ML platform or use a managed MLOps service.
The strongest argument for building internally is: