Design and evaluate P2P payments in messaging

Quick Overview

This question evaluates a data scientist's competency in product analytics, causal inference, experimental design under network interference, monetization measurement (ARPU impact), and risk mitigation for a peer-to-peer payments feature relevant to a Data Scientist role.

Design and evaluate P2P payments in messaging

Company: Meta

Role: Data Scientist

Category: Analytics & Experimentation

Difficulty: hard

Interview Round: Onsite

You're evaluating launching a Venmo‑like peer‑to‑peer payments feature in a large messaging app. 1) State the single most important business goal (objective function) and 2–3 concrete success metrics (precise definitions/units) for the first 90 days; include at least one guardrail protecting core messaging health. 2) Propose two monetization approaches (e.g., per‑transaction fee vs interchange/float) and how you would measure causal impact on ARPU without subsidy‑driven volume inflation. 3) Design a beta rollout that handles network interference and leakage: specify the randomization unit (user, dyad, ego‑network, or geo), exposure definition, how you’ll enforce access when a non‑assigned friend wants the feature, and how you’ll estimate ITT vs TOT. 4) If an RCT is infeasible, write the exact difference‑in‑differences specification (define treatment/control, pre/post windows, the estimand, required parallel‑trends checks, and clustering of SEs) and name two falsification tests. 5) List the top three risks (product, fraud/regulatory, ecosystem) with a measurable leading indicator and a mitigation you’d ship before full launch.

Quick Answer: This question evaluates a data scientist's competency in product analytics, causal inference, experimental design under network interference, monetization measurement (ARPU impact), and risk mitigation for a peer-to-peer payments feature relevant to a Data Scientist role.

|Home/Analytics & Experimentation/Meta
Meta logo
Meta
Oct 13, 2025, 9:49 PM
hardData ScientistOnsiteAnalytics & Experimentation
3
0

P2P Payments in a Large Messaging App — Design, Measurement, and Risk Plan

You are a data scientist at an at-scale messaging platform evaluating a Venmo-like peer-to-peer (P2P) payments feature.

Tasks

  1. Business goal and success metrics (first 90 days)
    • State the single most important business goal (objective function) for the first 90 days.
    • List 2–3 concrete success metrics with precise definitions and units.
    • Include at least one guardrail protecting core messaging health.
  2. Monetization approaches and measuring impact on ARPU
    • Propose two monetization approaches (e.g., per-transaction fee, interchange/float).
    • Describe how you would measure the causal impact on ARPU while preventing subsidy-driven volume inflation.
  3. Beta rollout with network interference and leakage
    • Specify the randomization unit (user, dyad, ego-network, or geo) and the exposure definition.
    • Explain how access will be enforced when a non-assigned friend wants the feature.
    • Explain how you will estimate ITT vs TOT.
  4. If an RCT is infeasible — Difference-in-Differences (DiD)
    • Write the exact DiD specification: define treatment/control, pre/post windows, the estimand, required parallel-trends checks, and clustering of standard errors.
    • Name two falsification/placebo tests.
  5. Top risks and mitigations
    • List the top three risks (product, fraud/regulatory, ecosystem) with: a measurable leading indicator and a mitigation you would ship before full launch.
Loading comments...