Debug a spike in complaints from families on Ads at global scale

Problem Statement Description

Product context: Uber is a mobility and delivery platform; its products include rides, Uber Eats, grocery and retail delivery, freight, driver and courier tools, and marketplace pricing.

Uber has detected a sudden global spike in complaints from family users related to Ads across its mobility and delivery experiences. The complaints may be coming from parents, family profile owners, teen-account guardians, or households using Uber for rides, Eats, grocery, or other marketplace services. The issue is occurring at global scale, which raises concerns around user trust, safety perception, ad relevance, targeting quality, and the impact of monetization on core Uber experiences.

You are the PM responsible for leading the root-cause investigation. Your task is to frame the anomaly, determine whether the spike is real or caused by measurement/reporting changes, identify which users, surfaces, geographies, ad formats, campaigns, or workflows are affected, and decide how to prioritize mitigations while protecting marketplace reliability and user trust.

The investigation should account for Uber’s operating complexity: multiple countries, languages, regulatory environments, family-related account types, diverse ad inventory, third-party advertisers, and interactions between rides, delivery, and membership experiences. You should also consider how complaints are captured across support tickets, in-app feedback, app-store reviews, social channels, safety escalations, and internal quality signals.

The experience should consider:

- How to define the complaint spike, including baseline period, denominator, affected population, and severity of complaints.

- How to segment by family profile type, teen/guardian flows, household usage, geography, product surface, platform, app version, and ad format.

- How to validate instrumentation, taxonomy, support-routing changes, translation issues, duplicate tickets, or reporting delays before assuming a product defect.

- What hypotheses could explain the spike, such as targeting errors, inappropriate creative, frequency increases, localization failures, campaign launches, eligibility bugs, or policy mismatches.

- What evidence would be needed from ads delivery logs, campaign metadata, user journeys, support transcripts, moderation systems, and marketplace metrics.

- How to assess user and business impact, including complaint severity, churn risk, trip/order completion, ad revenue, advertiser impact, and safety or regulatory exposure.

- How to determine immediate mitigations, escalation paths, owner responsibilities, and communication needs across Ads, Safety, Support, Legal, Local Ops, and Engineering.

- How to prevent recurrence through monitoring, alerting, policy checks, QA processes, advertiser controls, and family-specific guardrails.

The goal is to demonstrate a structured RCA approach that separates signal from noise, narrows the blast radius, identifies the most likely root causes with evidence, and balances fast user-protection actions with longer-term fixes appropriate for a global Uber Ads ecosystem.