Guest message
Received Β· Jun 12, 2026, 9:54 p.m.
hello? Is this matter being looked into?
Expedia Β· Message
Received Β· Jun 12, 2026, 9:54 p.m.
hello? Is this matter being looked into?
Sent Β· Jun 12, 2026, 9:55 p.m.
Thank you for reaching out. We are looking into the matter regarding the charge from Pool A Paris. A member of our team will be in touch with you soon to discuss this further.
Source: 1
Score Β· 47.23%Use this table as a fast lookup for any incoming guest message. Find the request type, confirm whether the LLM handles it or escalates, and identify the specific trigger that activates the escalation.
| Request / Scenario | LLM Response | Escalation Trigger |
|---|---|---|
| Early Check-In | Share $60 + tax fee, explain availability | Guest confirms they want to pay β then escalate |
| Late Check-Out | Share fee schedule, explain availability option | Guest confirms fee + desired time β then escalate |
| Connecting Rooms | Explain $20/room/night fee, request names & conf #s | Guest confirms fee + provides all names & conf #s β then escalate |
| Room Upgrade / Extend Stay (incl. adding a night or a later checkout) | Ask for best contact phone number; do not route to the booking channel | Guest provides phone number β then escalate |
| Fee / Charge Confirmation (any) | Thank guest, acknowledge the accepted fee; do not apply the charge or guarantee the add-on | Guest accepts/authorizes any fee or surcharge β always escalate (CONFIRM-FEE) |
| Name Addition / Alt Check-In | Collect name, explain ID/card requirements | Primary name change, billing auth, identity mismatch |
| Stay Modification / Cancel (shorten, cancel, date shift, room change, primary-name change β not adding nights or an additional guest name) | Explain 3rd-party policy, direct to booking channel | 3rd party cannot process; complex booking changes |
| Invoice / Folio / Receipt (incl. itemized) | Send the self-service folio response (emailed after checkout; check spam; else caesars.com folio-request link) | Billing dispute / correction / refund; business-invoice format; incidentals hold past timeframe |
| Room Issue / Housekeeping | Apologize, gather details, direct to front desk first | Front desk unresponsive; safety impact; comp requested |
| Maintenance Request | Acknowledge, direct to front desk / Laurel text | Front desk unresponsive; repair still needed |
| Noise / Safety Complaint | Acknowledge, direct to Security / front desk | Security/front desk intervention needed; unresolved |
| Billing Dispute / Refund | Acknowledge, collect details | Any correction, dispute, or refund β always escalate |
| General Complaint / Comp | Empathize, document thoroughly | Compensation requested; multiple failures; emotional guest |
| Post-Stay Review β Negative | Acknowledge publicly with empathy template | Safety, discrimination, viral-risk β escalate within 12h |
| Post-Stay Review β Positive | Respond with standard thank-you template | No escalation required |
Source: 2
Score Β· 46.49%RAG Context: Section 2 general escalation rules for Paris Las Vegas routine guest-service requests β Name Additions & Guest Authorization, Stay Modifications & Cancellations, Invoice/Folio/Receipt Requests, Room Issues & Housekeeping, Maintenance Requests, Noise/Neighbor/Safety Complaints, Billing Issues & Charge Disputes, and General Complaints & Compensation Requests. For each: trigger phrases and a clear split between what the LLM responds to directly versus what it escalates to a human Concierge. Safety, medical, injury, theft, and billing-dispute messages always escalate. Part of the master framework in
paris-las-vegas-resort-casino-escalations.md.
These rules apply to routine guest-service requests that arrive via message. The LLM should acknowledge professionally, provide any information it can, and escalate when the situation requires human action or decision-making.
Trigger Phrases
This response was escalated. Was that decision correct?
Not yet assessed
Set by the preferred response below β saving a correction marks it incorrect, clearing the correction marks it correct.
Not yet assessed
Optionally provide a better response. Saving non-empty text marks the LLM response incorrect; clearing it marks the LLM response correct.