Analytics Leadership & Innovation Flashcards
7 cards from real CDO practice questions. Tap to flip, then mark Knew It or Still Learning — missed cards come back until you master them.
Read the first 7 Analytics Leadership & Innovation flashcards as text
What is the primary benefit of a feature store in a mature machine learning program?
Answer: It enables consistent, reusable features across training and serving
Feature stores reduce duplication and training-serving skew by sharing curated features.
A CDO is launching a generative AI assistant for employees. Which control most directly reduces the risk of confidential data leaking to external model providers?
Answer: Enterprise agreements with no-training/no-retention terms plus data loss prevention controls
Contractual data protections and DLP controls limit exposure of sensitive data sent to external AI services.
Which statement best describes 'hidden technical debt' in machine learning systems?
Answer: Model code is a small part of a system; surrounding data pipelines and dependencies create ongoing maintenance costs
Most ML system complexity lies in data dependencies, pipelines, and configuration, which accumulate maintenance burden.
Which approach best monetizes data indirectly rather than selling it externally?
Answer: Using analytics to improve pricing, retention, and operational efficiency
Indirect monetization creates value by improving internal decisions and products rather than selling data.
When setting OKRs for an analytics team, which is the best example of a key result?
Answer: Reduce forecast error (MAPE) for demand planning from 18% to 10% by Q4
Key results should be specific, measurable, and time-bound outcomes.
An A/B test showed a statistically significant result after the team checked results daily and stopped as soon as p < 0.05. What is the main concern?
Answer: Repeated peeking inflates the false positive rate
Stopping early after repeated peeks increases the chance of a false positive unless sequential methods are used.
Which practice best ensures fairness in an AI-driven credit decisioning model in the U.S.?
Answer: Testing for disparate impact across protected classes and documenting adverse action reasons
Fair lending laws like ECOA require disparate impact testing and explainable adverse action notices.
How this page was made: written with AI tools and published by Nathan Beja, founder of Practice Test Geeks, who is responsible for its content. We have not yet matched this exam's facts to an official source, so confirm fees, dates and rules with the exam body. Not reviewed by a subject-matter expert. Editorial policy