Skip to content
← Back to work
Case Study / Revenue Management System Testing

Revenue Management System Testing

Led user acceptance testing for a revenue-management system, defined test scenarios and worked with analysts and the vendor to resolve data-mapping and forecasting defects before launch.

UAT leadershipData validationVendor coordination

The work

At Greyhound, I led user acceptance testing for a revenue-management system, defined test scenarios and worked with analysts and the vendor to resolve data-mapping and forecasting defects before launch.

Testing covered whether the new system had the right source data, whether that data was mapped correctly, and whether the forecasting behavior matched what the business expected.

CompanyGreyhound
WorkRevenue-management system user acceptance testing
ResponsibilityTest scenarios, data validation and defect investigation
CollaborationAnalysts and the system vendor

What I led

  • Defined test scenarios and expected behavior for the user acceptance testing workstream.
  • Led testing with analysts and coordinated investigation of discrepancies.
  • Validated source-data completeness and source-to-target mappings.
  • Isolated forecasting issues and worked with the vendor on corrections before launch.

Check the data before relying on the forecast

During source-data validation, I found that required records had not been loaded and some fields had been mapped incorrectly. I identified the missing and mismapped data and drove the corrections before go-live.

These checks addressed the inputs to the system, so a forecast could be assessed against complete, correctly mapped data.

Reconstructed test matrixIllustrative checks created for this case study from the work described here.
Source completeness
Expected behaviorRequired source records are available in the new system.
Example issueMissing records from the source-data load.
Field mappings
Expected behaviorSource values populate the intended target fields.
Example issueFields mapped incorrectly between source and target.
Booking-curve grouping
Expected behaviorHistorical bookings are grouped into the intended days-before-departure buckets.
Example issueIncorrect bucket assignment affecting the demand forecast.

Investigate the forecasting behavior

Demand forecasts for casino-destination markets looked wrong during testing. I traced the issue to a code defect affecting how historical bookings were assigned to days-before-departure buckets when building the booking curve.

I isolated the root cause and reported the details to the vendor. The vendor corrected the forecasting code before launch.

Outcome

The missing and mismapped source data was corrected before go-live, and the vendor fixed the booking-curve defect before launch.