The Second Trip · Independent proposal for Airbnb
I’m Yun Sol. I created this short case to show how I would turn an Asia growth opportunity into a locally relevant, measurable launch.
My recommendation
Pilot an intent-led return program in India. When a first-time guest plans again, match the new trip—not the last holiday. Test relevance before paying for a return.
The problem to investigate: does a first holiday make Airbnb relevant for the next occasion? Airbnb identifies an Asia execution-capacity constraint;1 a repeat-trip gap is still a hypothesis.
My proposed intervention
Use her new dates, group size and stated budget to surface suitable, available homes through existing channels. Test the shortlist before adding an offer.
No new travel intent? No nudge. Poor first stay? Resolve it before promoting another.
The launch & the proof
India’s first-time bookers grew over 75% YoY in Q1 2026.2
Momentum worth investigating—not proof of weak repeat.
Validate the gap. Audit CRM and repeat cohorts; interview returners and non-returners. Pick one domestic short-break corridor with suitable supply.
Launch a controlled journey. Country owns the local plan; I coordinate CRM, merchandising, supply, policy and payments. Named owners, deadlines, holdout.
Resolve, don’t just report. I run one review of results and blocked commitments. Finance and Analytics validate the read; I drive the next decision.
Randomise eligible guests: current CRM, relevant shortlist, or the same shortlist plus a capped offer.
Scale with reliable lift and sound economics. Keep no-offer if discounts add nothing. Stop for harm or no worthwhile gain.
Before spending: check existing lifecycle coverage, cohort size, current repeat behaviour and guest interviews. India and the corridor are conditional choices. If the issue is first-stay quality or suitable supply, address that first. No new product or absent feature is assumed.
Launch readiness: country lead approves the local plan; central teams confirm channel, inventory, consent and tracking readiness. I keep the dependency log, escalate overdue critical blockers to the regional lead and hold launch when prerequisites fail. Refresh total price and availability at display time. Never infer budget or companions from the last trip.
AI for operations: draft a source-linked exception brief from approved metrics and commitments. Owners verify explanations; Finance and Analytics retain metric standards. Measure reporting time saved and corrections. No autonomous spend changes or guest profiling. No live AI runs here.
Experiment: randomise first-stay completers at renewed domestic short-break intent. Retain existing CRM in the control. A: current CRM; B: matched shortlist; C: B plus a capped, approved offer with a named funder—not assumed host funding. If volume is limited, stage A/B then B/C.
Decision standard: agree minimum worthwhile lift, sample size, contribution definition and harm limits with Finance and Analytics before launch. Include every assigned guest; track total account-level stays and contribution, not only promoted inventory. Account for shared supply and displacement where needed.
Wait for the evidence: day-90 reads cover delivery and safety, not proof of growth. Wait for mature 180-day cohorts; follow up beyond that for pull-forward and discount dependence. Healthy unassisted repeat, existing program coverage or no incremental gain changes the recommendation. No uplift forecast is claimed.
Regional launches, segmentation, monetisation, live operations and P&L ownership across SEA and selected APAC markets. I connect a local insight to a shipped program—and change course when the evidence says no.
1 / Known operating challenge: Airbnb’s role description identifies Asia’s execution-capacity constraint. Country teams own market plans; regional delivery connects them; Finance and Analytics own measurement standards.
2 / Acquisition signal: Q1 2026 shareholder letter, geographic performance reports India first-time booker growth above 75% YoY. Weak retention is not an established finding.
Build on what exists: Q2 results describe personalised recommendations and promotions. Summer Release includes group planning and AI tools. This proposes a focused execution test, not a new CRM or travel planner.
Limits: cohort size, product overlap, repeat behaviour, signal permissions, supply and economics remain unknown. No interviews or internal analysis have been conducted. Aditi, her trips and the illustration are fictional. No model, account, booking, customer data or tracking is connected. Career statements reflect my confirmed experience, not hospitality expertise. Public sources checked 16 October 2026.