- My role
- Lead QA Engineer
- Industry
- Hospitality / Booking-as-a-Service
- Platform
- Web + Mobile web
- Year
- 2024
Overview
Zirow is a hospitality operations platform that lets guests discover, book, and pay for stays while giving hosts and admins the tools to manage inventory, pricing, availability and payouts. I owned quality end-to-end — from the first exploratory sessions on staging to launch readiness and post-release regression cycles.
What I owned
- ◆Guest booking flow: search → property detail → date/guest selection → checkout → confirmation
- ◆Payment paths across cards, transfer and wallet — including retries, failures and refunds
- ◆Host dashboard: listing creation, calendar sync, pricing rules and payout requests
- ◆Admin console: user moderation, dispute handling and financial reconciliation
- ◆Notifications: email, in-app and SMS across booking lifecycle events
Approach
Test strategy & risk mapping
Money-moving flows and multi-role state transitions were flagged as highest risk. I built a risk-based test plan that concentrated deep coverage on checkout, payout and booking-status transitions, and kept a lighter regression pass across cosmetic and content surfaces.
End-to-end scenario coverage
Wrote 200+ manual test cases mapped to real guest, host and admin journeys — happy paths, edge cases (timezone drift, overlapping bookings, expired cards, partial refunds) and negative paths (fraud attempts, race conditions on the same slot).
Cross-device & cross-browser
Verified booking and payment flows on iOS Safari, Android Chrome, and desktop Chrome / Firefox / Edge, with a focus on mobile because the majority of guest traffic converted on phones.
Defect triage & partnership with engineering
Logged reproducible bugs with clear steps, expected vs. actual, environment, and video/console evidence. Sat in triage with engineering and product to prioritise blockers, agree severity, and unblock releases without shipping regressions.
Outcomes
- ✓Zero critical payment defects at launch
- ✓Cut booking-flow bug escape rate significantly release-over-release
- ✓Established a reusable regression pack the team runs before every deploy
- ✓Documented recovery playbooks for failed-payment and disputed-booking edge cases