Skip to research
SAVED EVIDENCEResearch review

easyJet · CX evidence review

All 15 fixed first replies returned the same cancellation-reason menu. The menu did not answer the three refund scenarios, but its opening wording does not contradict policy; the button journey was not tested.

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

THE SAVED OBSERVATIONS

What the evidence shows.

easyJet

easyJet website chat

15 / 15Fixed first replies returned a menuAcross three different cases asked five times each, every first reply was the cancellation-reason menu.

All 15 fixed first replies returned the same cancellation-reason menu. The menu did not answer the three refund scenarios, but its opening wording does not contradict policy; the button journey was not tested.

Coverage observation

The first answer was a reason menu

The fixed questions did not receive the requested refund answer in their first reply.

Needs review

The button journey remains open

No menu option was pressed. This does not establish a wrong policy answer or failed end-to-end cancellation journey.

METHOD & SCOPE

How this study was run.

20 virtual-user chats × 2 messages and 3 fixed questions × 5 tries produced 55 recorded replies. Two trial-run messages are outside that analysis. The v4 report records 57 messages in 36 new chats, on 10 October, 21:24–22:00 UTC.

These are hypothetical policy conversations, not completed customer transactions.

Cancel within 24 hours: Standard LGW–BCN ticket bought directly by card 10 hours ago; departure in 10 days; voluntary change of mind. Expected against saved policy: Explain the 24-hour option and applicable cancellation fee, distinguish government taxes, and state the permitted refund route. Observed context: All five first replies were identical. Whether a typed question should be answered directly or routed through the menu needs easyJet’s confirmation.

Cancel after 24 hours: The same Standard booking, bought 30 hours ago, with departure in 10 days. No disruption or special circumstances. Expected against saved policy: Separate the non-refundable airfare from any government-tax claim and explain the agreed claim window and refund route. Observed context: The help page and Terms differ in tax-refund wording. Neither the menu’s eventual outcome nor which wording governs was tested or confirmed.

Flight leaves in 90 minutes: Standard ticket bought directly by card 10 hours ago; its only flight departs in 90 minutes. Expected against saved policy: Explain the two-hour departure cutoff rather than offering the 24-hour cancellation route without that condition; distinguish any government-tax claim. Observed context: No menu button was pressed. The first reply’s lack of a cutoff answer is observed; a wrong cancellation decision is not established.

LIMITATIONS

What this does not establish.

  • Typed questions only: no menu buttons were clicked, so later answers and completion of the cancellation journey are unknown.
  • A reason menu is not a wrong-policy statement. Direct-answer expectations and repeated confirmation steps require easyJet’s decision.
  • The 55 analysed replies were menus: 53 identical reason menus and two booking-time confirmation menus. Counts are observations, not customer failure rates.
  • The fee table’s charging unit was unspecified; no refund amount was calculated. Tax-refund wording differs between the saved help page and Terms.
  • One website and evening; virtual users rather than real customers; no login, booking, personal data or human agent.
  • New visible chats do not establish independent backend sessions. easyJet did not approve or confirm this study.

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