Skip to content
Travel and booking · Australia

4,000+ API tests guarding every booking

API regression across staging and production, payment flows and load tests for a booking platform.

4,000+ API test cases automated
  • 4,000+ API test cases automated
  • 600+ bugs reported
  • 95% of API endpoints covered
  • Platformsweb, API
  • Engagement14 months · 3 QA engineers
Challenge

A booking platform with a large API surface was tested by hand through the UI. API changes broke partner integrations without warning, and nobody knew how the system would hold under peak traffic.

What we did

We automated the API suite with Postman and Newman and ran it on both staging and production. We covered payment and refund flows with Stripe and PayPal, geolocation search and third-party integrations, and load-tested search and booking.

Result

4,000+ automated API tests now run across both environments and 600+ bugs were reported along the way. Breaking API changes are caught before partners see them.

Before and after
Before Test SquadAfter
API testing By hand, through the UI 4,000+ automated tests
Environments Staging and production behave differently Same suite runs on both
Coverage Unknown 95% of endpoints
Peak load Never measured Verified at 1,500 concurrent users
Timeline
  1. Weeks 1–4

    API inventory, test design, first collections

  2. Weeks 5–20

    Automation of core booking and payment APIs

  3. Weeks 21–40

    Full coverage, production smoke runs, reporting in Allure

  4. Weeks 41–60

    Load testing, monitoring dashboards, ongoing regression

Tools

  • Postman
  • Newman
  • Allure
  • k6
  • GitLab CI
  • Swagger
  • Datadog
  • Rollbar
  • Grafana

Testing types

APIcontract (OpenAPI)integrationregressionsmoke on productionpayment testing (Stripe, PayPal)geolocationdata integrityload testingtest automation

Want results like these?

Book a short call about your product. We can start with a free QA audit or a trial.

Book a call →