Data Analysis & Decision Making Flashcards
7 cards from real GMC practice questions. Tap to flip, then mark Knew It or Still Learning — missed cards come back until you master them.
Read the first 7 Data Analysis & Decision Making flashcards as text
Which scenario represents a 'novelty effect' threat to A/B test validity?
Answer: Existing users perform unusually well on a new design simply because it is different, inflating early results
Novelty effect occurs when users engage more with a new variant simply because it's unfamiliar, not because it's genuinely better — this inflates initial results.
A growth team uses last-touch attribution but suspects it undervalues top-of-funnel channels. Which model would better reflect multi-channel impact?
Answer: Position-based (U-shaped) attribution
Position-based attribution assigns more credit to first and last touches, capturing both initial awareness and final conversion while acknowledging middle-funnel channels.
What does a high 'time to value' (TTV) metric typically signal in a SaaS growth context?
Answer: Users take too long to reach their first meaningful outcome, risking early churn
High TTV means users wait too long to experience core product value, increasing the likelihood they abandon before activation.
In a growth context, what is the key distinction between 'correlation' and 'causation'?
Answer: Correlation shows two variables move together; causation proves one variable directly causes the other
Correlation identifies a relationship between variables, but only controlled experiments or causal inference methods can establish that one variable causes another.
A company calculates its payback period as 18 months. What does this mean for growth investment decisions?
Answer: The company recovers its customer acquisition cost in 18 months, which is acceptable for enterprise but risky for high-churn consumer products
An 18-month payback period means CAC is recovered after 18 months of revenue; this is viable for enterprise with long contracts but problematic if churn happens before that point.
Which SQL aggregate function would a growth analyst use to find the median session duration from a large dataset?
Answer: PERCENTILE_CONT(0.5) WITHIN GROUP (ORDER BY session_duration)
PERCENTILE_CONT(0.5) calculates the continuous median in standard SQL; most dialects use this function since a native MEDIAN() is not universally supported.
A growth team is deciding between doubling down on a channel that drives high volume but low LTV users versus a channel that drives low volume but high LTV users. What framework guides this decision?
Answer: Compare CAC-to-LTV ratio for each channel and favor the one with the highest LTV:CAC ratio
The LTV:CAC ratio measures the return on acquisition investment per channel; a higher ratio indicates more efficient, sustainable growth regardless of volume.