Design an experimentation dashboard for Business Travel

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 is exploring how to evaluate product experiments for its Business Travel experience, where guests may be booking work trips, combining stays with local experiences, coordinating with company policies, or comparing Airbnb against hotels and other travel platforms. You are asked to design an experimentation dashboard that helps product, data, operations, and marketplace teams understand whether Business Travel experiments are improving user outcomes without harming trust, host quality, or marketplace health.

The dashboard should support decisions across the experiment lifecycle: pre-launch readiness, live experiment monitoring, segment-level diagnosis, and post-experiment readout. It should be useful for experiments such as changes to business-travel search filters, checkout flows, policy-compliant listings, receipt/invoicing flows, work-trip labeling, or recommendations for experiences around a business stay.

Your task is not to pick a specific experiment to run, but to define what an effective experimentation dashboard should measure, how metrics should be structured, and how teams should interpret the data. Be clear about the users of the dashboard, the core metric definitions, denominators, cohorts, instrumentation needs, and guardrails required to make launch or rollback decisions.

The experience should consider:

- Who uses the dashboard, such as PMs, data scientists, business travel operations, trust and safety, host marketplace teams, and leadership.

- Primary experiment metrics for Business Travel, including how success is defined and what the denominator should be.

- Funnel coverage from search or trip intent through booking, stay completion, expenses/receipts, reviews, and repeat usage.

- Cohorts and segmentation, such as business travelers versus leisure travelers, company-managed versus self-managed trips, geography, trip length, host type, and new versus returning users.

- Guardrail metrics for cancellations, customer support contacts, disputes, safety issues, refund rates, host acceptance, listing quality, and marketplace balance.

- Instrumentation requirements, including event tracking, experiment assignment, exposure logging, attribution windows, and data quality checks.

- Decision usefulness, including confidence intervals, sample size visibility, experiment duration, alerts, and clear readouts for ship, iterate, or stop decisions.

The goal is to describe a metrics dashboard that enables Airbnb teams to evaluate Business Travel experiments confidently, balancing guest conversion and business-trip utility with trust, safety, host supply health, and long-term marketplace quality.