Skip to research
mphorapulse
SAVED EVIDENCEResearch review

Jetstar Airways · CX evidence review

Five original-Flex-Plus questions received the same unrelated menu. Four repeats of a later-upgrade question reached payment deadlines. These are routing and coverage gaps, rather than proven wrong refund decisions.

Fieldwork
10 October 2026 (UTC)
Study scope
Three carrier-specific fixed questions × five attempts, alongside adaptive virtual-user conversations.

THE SAVED OBSERVATIONS

What the evidence shows.

Jetstar Airways

Ask Jess

5 / 5Original Flex Plus reached unrelated menusAll five identical questions got the same unrelated menu; its buttons were not pressed.

Five original-Flex-Plus questions received the same unrelated menu. Four repeats of a later-upgrade question reached payment deadlines. These are routing and coverage gaps, rather than proven wrong refund decisions.

Coverage observation

Flex Plus questions reached unrelated topics

The menu offered disruption/change, in-flight payments and insurance-claim confirmation instead of the requested voucher/deadline answer.

Coverage observation

A selected wrong-topic answer recurred

Four of five later-upgrade repeats answered about payment deadlines. The wording was chosen after a known failure.

Needs review

Relevant adaptive buttons were often present

Twenty-four of twenty-five adaptive topic menus offered a relevant button. Typed follow-ups were used instead.

METHOD & SCOPE

How this study was run.

20 virtual users × 2 replies plus 3 fixed questions × 5 tries produced 55 analysed replies. The original report records 65 typed messages in 39 chats, including additional captures. Tested 10 October, 20:26–21:23 UTC (11 October AEDT).

These are hypothetical policy conversations, not completed customer transactions.

Original Starter fare: JQ-operated MEL–SYD one-way, bought direct by card 10 hours ago; departure in 10 days. Original Starter, no bundle or FareCredit, voluntary cancellation without disruption or special remedy. Expected against saved policy: Answer whether this stated booking receives cash, a voucher or neither, while preserving the legal and other applicable exceptions. Observed context: All five replies used consistent general cancellation wording that does not conflict with policy. Whether the first answer should resolve the stated fare is a design decision.

Flex Plus bought at booking: The same domestic route, bought 10 hours ago with Flex Plus at the original booking; cancel 10 days before departure, before airport check-in. No special remedy or FareCredit. Expected against saved policy: State the voucher option and request deadline for original Flex Plus, with relevant conditions. Observed context: The identical menu offered flight disruption/change, in-flight payment methods and insurance-claim confirmation. None answered this voluntary-cancellation question; buttons were not pressed.

Starter later changed to Flex Plus: Original Starter with no FareCredit or bundle, already changed to Flex Plus; direct JQ MEL–SYD booking, ten days before departure. The question asks cancellation entitlement and deadline. Expected against saved policy: Apply the original fare history and distinguish cancellation/request deadlines from deadlines for paying for a booking. Observed context: This question was selected after the same wording failed in central_control_02. It shows recurrence, not prevalence. No personal details were entered into the handoff form.

LIMITATIONS

What this does not establish.

  • No analysed reply resolved cash, voucher or neither for the described booking; some nevertheless supplied correct general policy information.
  • Typed follow-ups were used instead of menu buttons. In the adaptive chats, 24 of 25 topic menus offered a relevant button.
  • The five repeated deadline questions were selected after one observed failure; their outcomes are not an estimate of customer prevalence.
  • The extra menu-button chat is outside the 55 analysed replies and was not repeated. Its missing later-upgrade exception needs policy-owner confirmation.
  • Long hypothetical wording and input refusals limit interpretation. The actual input-handling cause is unknown.
  • Virtual users, one public website and one day; no login or personal data. Jetstar did not confirm expected answers; visible new chats do not prove backend independence.

FOLLOW THE SOURCE

Study material.

Explore the available sources behind these findings.

YOUR CUSTOMERS. THEIR CONTEXT.

What would they experience?

Explore how your AI handles people’s goals, situations, and unfolding conversations with VIVID.

PULSE / GET IN TOUCH