Debug a spike in complaints from luxury travelers on Guest App

Problem Statement Description

Product context: Airbnb is a travel marketplace connecting guests and hosts; its products include stays, experiences, host tools, guest booking, trust and safety, payments, and reviews.

Airbnb has observed a sudden spike in complaints from luxury travelers using the Guest App. This segment typically has higher expectations around listing quality, accuracy, service responsiveness, check-in smoothness, trust, privacy, and issue resolution. The complaints may be coming from different points in the guest journey, such as search, booking, pre-arrival communication, check-in, stay experience, post-stay support, or refunds.

Your task is to approach this as a root-cause analysis problem. You should frame the anomaly clearly, determine whether the spike is real or caused by measurement changes, identify which luxury traveler cohorts or journey stages are most affected, and develop a structured set of hypotheses to investigate. Consider that Airbnb is a two-sided marketplace, so potential causes may involve guest app changes, host behavior, listing supply quality, pricing, support workflows, policy changes, seasonality, or external travel conditions.

The experience should consider:

- How to define the complaint spike, including baseline period, complaint rate denominator, severity, complaint categories, and affected channels

- How to segment the issue by geography, trip type, listing tier, price band, platform, app version, booking source, stay dates, and guest tenure

- How to check instrumentation, tagging, support intake changes, review classification, app release timing, and data pipeline reliability

- How to separate pre-booking, booking, pre-trip, in-stay, and post-stay complaint drivers

- Hypotheses related to listing quality, expectation mismatch, host responsiveness, check-in, amenities, cancellations, payments, refunds, and customer support

- Evidence needed to confirm or reject hypotheses, including product analytics, complaint text, support tickets, host-side metrics, reviews, NPS, and operational data

- Short-term mitigations to protect affected guests while investigation continues

- Longer-term prevention mechanisms, monitoring, ownership, and escalation paths

The goal is to demonstrate a clear, structured RCA approach that protects guest trust and marketplace quality without prematurely jumping to a solution. Your response should show how you would isolate the root cause, prioritize the most likely explanations, coordinate with relevant teams, and decide what action is needed based on evidence.