Manage personas, competitor filters and communication guidelines centrally
In the new AI agent settings (Bot settings --> Livechat settings), you can define how the AI agent should communicate. The settings are made via the persona, a defined list of competitors and binding communication guidelines. The following explains what the individual setting options do. An extended control function can be activated as an option, which automatically blocks responses if defined guidelines are breached. This ensures standardised, brand-compliant communication with end customers.
Persona
The persona determines the tone of voice and role in which the chatbot communicates. Based on the defined characteristics, the AI agent adapts the choice of words, address and style to the desired effect. This configuration forms the basis for consistent, brand-appropriate communication.
Role
- The functional position of the AI agent is described in the Role field, e.g. customer service or sales.
Description
- In the Description field, specifications are defined for tonality, response length and language style (e.g. casual or formal).
Competitive protection
A list of competitors that the AI agent should refrain from naming or describing can be stored in the competition protection area. This blocking logic protects sensitive brand positioning and prevents unintended statements about defined market participants.
Company
- The Company field is used to specify the company perspective from which the AI agent responds.
Competitor
- In the Add competitor field, you can enter direct competitors whose naming is excluded. In a specific use case, the AI agent replaces the third-party brand name with the name of your own company. In this case, the feature replaces the name of the competitor Melibo with moinAI and responds to the customer request accordingly.
Additional compliance check
When the compliance check is activated, an additional check is carried out. It recognises rule violations and automatically blocks non-compliant responses.
Communication rules
The communication rules define stylistic guidelines that influence the tone of the answers without changing their content. The AI agent's responses are automatically adapted to the defined rules. The aim is to achieve a standardised, brand-appropriate use of language.
Add a communication rule
New rules, e.g. ‘use simple language’, can be entered in the free text field at the bottom right.
Click on the plus sign to apply the rule. A rubbish bin symbol then appears in place of the plus sign. Click on this symbol to remove an added communication rule.
Click on the disc icon at the top of the menu bar to save the new rule and take it into account when generating responses.