GMC Risk Assessment & Mitigation 2 — Questions and Answers
Question 1: A growth team wants to run 10 simultaneous A/B tests on their homepage. What is the primary risk of this approach?
- Increased server load from traffic splitting
- Interaction effects between tests contaminating results (Correct answer)
- Running out of unique variant combinations
- Regulatory compliance violations
Correct answer: Interaction effects between tests contaminating results
Running multiple simultaneous tests risks interaction effects where one test's variant influences another's outcome, leading to unreliable data.
Question 2: Which metric best indicates that a new growth channel is becoming over-reliant in your acquisition mix?
- Channel CAC rising above blended CAC
- A single channel contributing over 50% of new users (Correct answer)
- Organic traffic declining month over month
- Email open rates dropping below 20%
Correct answer: A single channel contributing over 50% of new users
When a single channel exceeds 50% of new user acquisition, the business faces concentration risk if that channel's performance deteriorates.
Question 3: A SaaS company plans to triple its paid ad spend next quarter. What risk assessment step should come FIRST?
- Redesign landing pages for higher conversion
- Audit current unit economics to confirm positive LTV:CAC ratio (Correct answer)
- Hire additional customer success staff
- Negotiate bulk discounts with ad platforms
Correct answer: Audit current unit economics to confirm positive LTV:CAC ratio
Scaling spend before validating unit economics risks amplifying losses; confirming a healthy LTV:CAC ratio ensures the scaled spend is profitable.
Question 4: What does 'novelty effect' risk mean in the context of growth experiments?
- Users avoid new features out of unfamiliarity
- Early positive results are inflated because users respond to newness, not actual value (Correct answer)
- Novel experiments require longer approval cycles
- New product categories have unpredictable demand curves
Correct answer: Early positive results are inflated because users respond to newness, not actual value
The novelty effect causes users to engage more with changes simply because they are new, making early experiment results misleadingly positive.
Question 5: A growth marketer launches a viral referral program that goes unexpectedly mainstream. Which risk is MOST immediately critical to address?
- Brand dilution from too many new users
- Infrastructure inability to handle the traffic spike
- Referral fraud and incentive abuse at scale (Correct answer)
- Decline in existing user NPS scores
Correct answer: Referral fraud and incentive abuse at scale
Viral referral programs that scale rapidly are highly vulnerable to fraud and incentive abuse, which can drain budget and distort acquisition metrics instantly.
Question 6: In risk-adjusted growth planning, what is a 'sensitivity analysis' used for?
- Measuring how emotionally sensitive users are to price changes
- Modeling how key outcomes change when individual input assumptions vary (Correct answer)
- Determining statistical significance thresholds for experiments
- Assessing team morale risk during high-growth sprints
Correct answer: Modeling how key outcomes change when individual input assumptions vary
Sensitivity analysis tests how much a result (e.g., revenue) changes when a single input assumption (e.g., conversion rate) is altered, revealing which risks matter most.
Question 7: Which mitigation strategy is most appropriate when a growth channel shows high variance in monthly performance?
- Double down on that channel to average out the variance
- Pause the channel until performance stabilizes
- Diversify the channel mix to reduce overall portfolio variance (Correct answer)
- Increase bids to improve performance consistency
Correct answer: Diversify the channel mix to reduce overall portfolio variance
Diversifying across multiple channels reduces the impact of any single channel's variance on overall growth performance.
GMC Risk Assessment & Mitigation 2
A growth team wants to run 10 simultaneous A/B tests on their homepage.
What is the primary risk of this approach?