I started by aligning on scope: Airbnb = marketplace for short- and long-term homestays (ignore Experiences for now). We’re investigating revenue decline on homestays only.

Metric sanity: Revenue = host payout + Airbnb fees recognized on a completed booking (prepaid or post-stay depending on policy). No changes to definition, data pipelines look healthy. Pattern: sudden decline over ~15 days (not seasonal), global, all demographics, but skewed to Android.

I ruled out quick externals: no regulatory shocks, no political unrest, no operational strikes. One macro signal exists—a recent global financial scare—which could soften demand. But given the sudden, 15-day cliff and Android skew, I treated macro as background, not the primary cause.

Then I walked the funnel on Android:

App open / sign-in: stable
Search (dates/guests/filters) → results: stable
Results (listing grid) → PDP click-through: suspect area (recent UI change)
PDP → checkout → payment: stable

We examined what shipped: a new Android UI that removed “price + discount” badges from the search results grid. PDP still shows full pricing and discounts; checkout unchanged.

Hypothesis fit: If price/discount visibility disappears on the results page, users lose the core decision signal that drives PDP clicks. Expected symptoms: lower CTR from results, more back-and-forth, higher abandonment, and fewer completed bookings—i.e., revenue down—particularly on Android (where the change launched). That aligns with the timing, device skew, and suddenness. The macro downturn likely adds some drag, but it’s not sufficient to explain the sharp drop.

RCA (succinct): Revenue ↓ because bookings ↓; bookings ↓ because results-to-PDP CTR fell; CTR fell due to Android UI regression removing price/discount cues on the results grid. Macro demand softness is a secondary amplifier.

Immediate actions (0–2 weeks):

Rollback / hotfix Android results grid to restore price + discount visibility (A/B guarded; canary cohorts first).
Guardrails: Ship a release gate that blocks rollout if results→PDP CTR, add-to-trip, or checkout starts fall >X% vs. control.
Comms: Lightweight in-app nudge banner on Android (“New price clarity on results”) to recover trust/intent.

Follow-ups (2–8 weeks):

Price salience design: Standardize total price per stay/night, discount badge, and fees preview on the card (test variants).
Ranking tweak: Temporarily boost listings with active discounts for affected cohorts to accelerate recovery.
Diagnostics: Dashboards that track per-step conversion (search → results → PDP → book) by device/OS/version; add UI telemetry (impression of price badge) to attribute impact.

Macro hedge (in parallel):

Targeted incentives in regions/segments showing the steepest elasticity (e.g., limited coupons for low-seasonality dates).
Messaging tests on affordability (flexible refund filters, pay-in-installments where legal).

How I’d measure success:

North Star: Bookings and revenue per Android DAU back to pre-drop baseline (then +X%).
Leading indicators: Results→PDP CTR ↑, PDP→checkout CVR ↑, share of bookings with discount badge viewed.
Quality checks: Refund rate, host cancellations (ensure no perverse effects), and help-center tickets about “confusing pricing” ↓.

One-liner summary I’d give in the room: This wasn’t a demand-gen or payments issue—it was a price salience regression on Android search results that choked the funnel at the results→PDP step. Roll back the change, ship gates to prevent repeats, restore and then improve price/discount visibility, and use light demand levers to offset broader macro softness.

Airbnb Interview Questions & Answers

Conduct a root cause analysis to determine why Airbnb's revenue has declined