Create Smart Tagger

Smart Tagger provides a supper easy and powerful way to keep your user engagement organized - describe what you want and leave the rest to AI.

Smart taggers are very powerful tools that can be used for many purposes including the following:

• Serve as tools to classify, organize and analyze your messages.

• You can utilize smart tagger to filter out important messages and let Aissistant respond only to these queries. (option to "respond only to tagged")

• Tags can be employed to disregard certain messages, such as advertisements. (option of "no reply")

• Tags can be used to add additional context and/or instruction to further fine-tune the AI behavior under specific scenarios.

How to create a smart tagger?

To create a tag, you can either modify a template to suit your requirements or input the necessary details directly into the tag creation interface.

Basic

Direction

Direction indicates where the detection of tag will happen. It has two options: Inbound / Outbound. Inbound is that tag detection will happen on user's request. In that direction, you can use tag to categorize on what user wants. Outbound is that tag detection happens when AI generates the response. You can use the tag in that direction to detect what AI says, for instance, "if AI says I will transfer you to human" -> "human_transfer" tag.

Inbound is more commonly used

The tag detection, in both direction, include the conversation history. Therefore, when you describe the scenario for the tag, the scenario can be based on the past conversation vs. just the current user input

Scenario

Scenario is the trigger that describe the situation under which the corresponding tag should be created. When describing the scenario, please use clear and straightforward language. For instance, you might use descriptions like "user wants to make a return" or "user wants to cancel an order.”

Contextual Instruction, Context and Summary

Context

Context under the tag is the additional information to be provided to AI when the corresponding tag is identified. For instance, if the tag is related to a user wanting to make a return, you could incorporate your return policy in this section.

Contextual Instruction

Contextual instruction, under the tag, is the additional instruction to be provided to AI when the corresponding tag is identified. The information provided is employed to fine tune Aissistant's responses. For example, you can provide directives to prompt the request for an order number, the reason for a return, and so on.

Summary

The summary that AI to generate when the corresponding tag is detected. The generated summary will show in agent platform as a note or comment, to be consumed by human team. It could be a very useful tool to be used as communication between AI and human team.

Control (Post Action)

Suspend afterwards

This feature allows Aissistant to respond to the immediate message but prevents it from addressing subsequent messages. It's typically used to transition the conversation from Aissistant to the human team.

No respond

Aissistant will ignore the messages, it won’t respond or draft for the tag.

“Respond only to tagged”, when enabled, Aissistant will only try to respond to incoming messages that have at least one of the tags. It applies to both “Auto-Reply” and “Auto-Draft”.

This mode is best suited noisy channels like shared corporate email.

How does it work with Agent Platform?

When Aissistant creates a tag, the tag will be added to the Agent Platform.

Once tag is created, most of Agent Platform provides mechanism to carry on further actions including automation, notification, etc.