Allocate Support Cost and Diagnose Decline

Quick Overview

This question evaluates operational analytics competencies including cost allocation and attribution for support work, root-cause diagnosis of KPI regressions, and metric design for complaint-driver identification and prevention.

Allocate Support Cost and Diagnose Decline

Company: DoorDash

Role: Analytics Engineer

Category: Analytics & Experimentation

Difficulty: hard

Interview Round: Technical Screen

You are the analytics partner for the Customer Support team at a food-delivery company. You have the following data: agents(agent_id, monthly_salary, productive_hours, tenure, locale), tickets(ticket_id, created_at, resolved_at, issue_type, severity, channel, order_id, merchant_id, courier_id, csat, reopened_flag, escalated_flag), and ticket_touches(ticket_id, agent_id, minutes_spent, action_ts). 1) Design a fair method to allocate salary cost or performance-based compensation across support agents using completed complaint tickets. Explain how you would adjust for ticket difficulty, multiple agents touching the same ticket, quality, and gaming risk. 2) A core support KPI suddenly declines. Explain how you would determine whether the drop is caused by instrumentation changes, seasonality, queue mix, staffing, or a real operational issue. 3) Complaint volume is rising because merchants prepare the wrong order, couriers pick up the wrong bag, or couriers deliver to the wrong customer. What metrics would you track, how would you identify the main driver, and what preventive interventions would you recommend?

Quick Answer: This question evaluates operational analytics competencies including cost allocation and attribution for support work, root-cause diagnosis of KPI regressions, and metric design for complaint-driver identification and prevention.

|Home/Analytics & Experimentation/DoorDash
DoorDash logo
DoorDash
Oct 12, 2025, 12:00 AM
hardAnalytics EngineerTechnical ScreenAnalytics & Experimentation
4
0

You are the analytics partner for the Customer Support team at a food-delivery company. You have the following data: agents(agent_id, monthly_salary, productive_hours, tenure, locale), tickets(ticket_id, created_at, resolved_at, issue_type, severity, channel, order_id, merchant_id, courier_id, csat, reopened_flag, escalated_flag), and ticket_touches(ticket_id, agent_id, minutes_spent, action_ts). 1) Design a fair method to allocate salary cost or performance-based compensation across support agents using completed complaint tickets. Explain how you would adjust for ticket difficulty, multiple agents touching the same ticket, quality, and gaming risk. 2) A core support KPI suddenly declines. Explain how you would determine whether the drop is caused by instrumentation changes, seasonality, queue mix, staffing, or a real operational issue. 3) Complaint volume is rising because merchants prepare the wrong order, couriers pick up the wrong bag, or couriers deliver to the wrong customer. What metrics would you track, how would you identify the main driver, and what preventive interventions would you recommend?

Loading comments...