The response addresses the specific review
AI uses the review content, approved knowledge and brand standards. It does not simply change the name in a ready-made template.
Praise for the service and a comment about the room
The guest praises reception and breakfast but notes that the room was too warm.
Response ready for editing
The suggestion thanks the guest for the points raised, addresses the comment and avoids claims the hotel cannot confirm.
- content aligned with brand calibration
- visible sources and risk level
- publication only after a staff decision
A ready-to-use suggestion in the user’s workspace
The suggestion combines the review, brand standards and the property’s knowledge. A difficult case does not receive a routine suggestion; it is sent to the assistant and escalation process.
Creates a response
Prepares a specific version tailored to the content, language and tone of the message.
Shows the basis
Highlights the sources, rules applied and quality-check result.
Recognises the boundary
Blocks a routine suggestion and passes the case on when high risk is detected.
How is a suggestion created beside a review?
Open the review
The extension recognises the current review in the portal or reputation queue.
Gather the context
AI uses the content, language, calibration, hotel knowledge and risk classification.
Check the suggestion
The response is checked for facts, tone, promises and escalation.
User decision
The staff member edits the content and approves its publication themselves.
The suggestion retains its sources and check result
The user can see which parts of the review, facts and rules informed the response.
What does AI use?
- review content and language
- brand response calibration
- approved hotel knowledge
- priority, quality control and risk level
What does the team receive?
- draft response
- alternative content versions
- sources, rules and risk level
A suggestion does not mean automatic publication
At this stage, the staff member can change any part or reject the suggestion entirely. High-risk cases always go to a human.
The staff member edits, approves or rejects the response and publishes it in the portal themselves.
The module does not publish independently, respond to crisis cases, grant compensation or bypass the risk-filter block.
A contextual response ready beside the review
The staff member starts with a version tailored to the specific review.
The tone, facts and boundaries come from approved calibration.
The user can see the sources and always decides whether to publish.
The extension brings knowledge and control into the portal
The Central Reputation Panel provides the queue, AI in the browser shows the suggestion, and calibration and controls maintain the standard.
Related process elements
Frequently asked questions
Is the response published automatically?
No. A staff member edits and approves every publication.
Can AI prepare several versions?
Yes. It can show alternative versions based on the same calibration.
What happens to a high-risk case?
The routine suggestion is blocked and the case is sent to crisis assistance and a human.
Can the user see the response sources?
Yes. The module shows the facts and rules used, along with the check result.