Create your AI Bot scenario
Table of Contents
Available for AlcmeonChat, WhatsApp, LINE channels
Alcmeon bots leverage generative AI to deliver responses that closely resemble human interactions. They give accurate information to customer queries reducing the need for advisor involvement.
To generate these responses, the AI bot analyses the conversation context and uses your knowledge base to provide accurate and relevant answers.
Create your AI Agents
In the section “Settings / Alcmeon AI” you can configure the AI Agent(s) that will power your bot scenario.
For more information, check out our detailed article: Creating your AI Agent .
Setting up your AI Bot scenario
Here's the typical step-by-step process for integrating AI-generated responses into your Alcmeon bot scenario. Each step will be explained in more detail below:
1. Retrieve the customer's question
The first part of this scenario branch involves to ask the client to type a question. To do this, you will need to use two specific steps to pass the customer's question to the AI.
Use the “Subbot” step, and click the “pencil” icon to edit it. Select the “internal” subbot called “Capture message content” and validate. Once added to your scenario, this step will have three possible outcomes:
- Message content captured : This means, the step succeeded, and the variable “message_content” now contains the user’s last message.
- Message format not supported : The user's last message contains an unsupported format, such as a Line sticker, image or video. We recommend linking this output to a step that asks the customer to type a new question.
- Exit failed : A technical error occurred. We recommend linking this output to a step that asks the customer to type a new question.
Connect the "Message content captured" output to the optional next step, or if you don't want to implement it, to the AI Connector step.
2. Introduce the AI assistant (optional)
This step is optional, but we recommend it. It allows you to inform the customer that an AI assistant will now provide a response, and that it may take a few seconds, which is often the case with generative AI.
Connect the output of this step to the IA Connector step.
3. Hand over to AI
This step sends the customer's question to your AI Agent. The IA will then generate an answer or indicate that it was unable to provide an answer. The latter may happen for several reasons that we will detail further.
Configure the AI Connector
.png)
Click the “pencil” icon on this step to open its settings panel. You will find 3 options:
Maximum number of characters per message : This setting defines the character limit your AI Agent should follow when generating a response. Note that the limit must not exceed the limit imposed by the channel such as WhatsApp for example. Also note that generative AIs occasionally exceeds the given limit. If this happens, Alcmeon will split the AI response into two messages.
String that prefixes all AI responses : Allows you to start all AI responses with the same string of characters. Our tip: simply use an emoji followed by a space to let your customer know they're talking to an AI bot.
AI Agent : Select the IA Agent responsible for generating respnses from the drop-down menu. This must be configured in the “Settings / Alcmeon AI” section.
Managing different outputs
Below, each exit includes a description, the bot event (1) triggered when the exit is taken, and our recommendation for connecting it to the next step.
Exit Name |
Description |
Bot event |
Recommended next step |
|---|---|---|---|
AI - response (no advisor needed) |
The AI returned a valid response which has been sent to the customer. | AI Event: AI computed an answer |
Connect to an “AI Satisfaction” step |
AI - no response (off topic) |
The message was irrelevant (subject outside the customer's scope, problematic messages, etc.). | AI Event: AI has no answer for you |
Transfer to an advisor |
AI - response (advisor needed) |
The AI bot returned a response but recommended involving an advisor for more detailed help. | AI Event: AI computed an answer but should hand over to an advisor now |
Transfer to an advisor |
AI - no response (advisor needed) |
The AI could not generate a sufficiently relevant response . | AI Event: AI computed an answer NOT sent, handover to an advisor |
Transfer to an advisor |
Invalid message format returned by AI |
The AI's response format was invalid (unsupported media, message too many long, etc.) | AI Event: Invalid format of message returned by AI |
Transfer to an advisor |
AI - no response (interrupted by user) |
The customer sent another message before the IA had time to reply. | AI Event: The user sent another message before the AI could respond |
Transfer to an advisor |
AI API Error |
A technical error occured in the response from the IA provider or the provider did not respond. | AI Event: AI API error |
Transfer to an advisor |
Exit failure |
A technical error occured in the connector itself. |
AI Event: Failure exit |
Transfer to an advisor |
(1) Bot events are internal tracking elements of Alcmeon, that help you analyze bot performance and user behavior in the Analytics / Volumes / Bot Events section.
4. Measure satisfaction
When the AI Connector step triggers the output “AI - response (no advisor needed),” it meant the AI's response has been sent to the customer. This is a good moment to ask for their feedback. To do so, connect the output to an “AI Satisfaction” step.
In this step you can configure:
- The text of the question asking the customer if they find the AI response relevant
- The text for the “Yes” response button
- The text for the “No” response button
This stage has three possible exits:
- A “Yes” button: Link this to a message box with a menu to say goodbye to the customer, and optionally offer a button to return to the start of the scenario.
- A “No” button: Link this to the scenario branch that transfers the customer’s message to an advisor.
- A “User input” output: If the customer doesn’t click either of the two buttons and instead types a new message, it likely continues their previous question. We recommend linking this output to the “Capture message content” sub bot step earlier in the scenario. This will let Alcmeon inform the AI that the new question is a continuation, providing the AI with the necessary context to respond appropriately.
5. Handling AI non-response cases
Here is an example that illustrates our recommendations from the table above. Feel free to use it as inspiration to build your own scenario!