Evaluate merchant partnership for high-value customers
Company: Capital One
Role: Data Scientist
Category: Analytics & Experimentation
Difficulty: hard
Interview Round: Technical Screen
C1 (a credit-card issuer) is considering partnerships with a home‑sharing platform (Rent‑a‑Home, RH) and big-box retailers (Costco‑like) to attract high‑LTV cardholders. Define “high value” quantitatively (e.g., 12‑month net contribution after CAC and expected loss, or 24‑month CLV). List and justify the top factors you would analyze before signing: overlap and incremental reach, expected average spend and category mix, interchange by MCC, partner commission/revenue share, promo mechanics (e.g., 30% off RH), cannibalization of existing spend, breakage, approval/activation/usage funnels, credit risk and expected loss, fraud risk, operational/servicing costs, and legal/regulatory constraints. Propose a measurement plan: primary KPI(s), guardrails, experiment design (randomized offer with holdout), sample size and power assumptions, and how you’d detect/adverse‑select against “credit‑card gamers.” Specify data needed, key segments (new vs. existing, peak vs. off‑peak), and your go/no‑go thresholds (e.g., payback within 12 months and minimum IRR).
Quick Answer: This question evaluates skills in customer lifetime value modeling, marketing experiment design, acquisition economics, funnel and credit-risk analysis, fraud and operational-cost assessment, and measurement planning for partnership decisions.