Overview
Agents that can deflect customer questions, route issues intelligently, or gather internal context for your team.
Last updated
Was this helpful?
Agents that can deflect customer questions, route issues intelligently, or gather internal context for your team.
Last updated
Was this helpful?
AI Agents can automatically answer customer questions and carry out a support workflow, including escalating a question to a team member.
BuildTestDeployYou can think about AI Agents as an extension of your team, users who can be assigned to issues. They can be used for many things, including:
Intelligent routing of issues between teams
Handling of questions based on knowledge base or documentation content
Initial information collection from customer
Gathering of context internally for your team
You can use AI Agents to attempt to automate the pieces of your workflow you want to - expect it to be an iterative process to identify the best workflows to automate, and to build your skills in instructing the agent.
If you're not sure whether a workflow or use case is a good fit for an AI Agent, reach out to the Pylon team for more guidance.
For AI Agents that will be interfacing with the customer directly, we recommend making it obvious through the name or avatar the it is an AI
Examples of common name formats are your company mascot/avatar or a name like "Ashley the AI" or "Ashley Bot".
Start small: Identify common questions or repeated internal workflows and rollout agents incrementally, don't attempt to automate everything at once.
For a customer-facing agent, examples are deflecting questions for low-value accounts first
Aim to use agents to create leverage for your team: Let your team do what they do best and use the AI Agents to augment.
Example: Have a team that works only Pacific hours but have international customers? Consider using an AI Agent to take a first pass of questions that come in off your normal hours
Example: Does someone on your team spent many hours triaging issues to the right people? Consider using the AI Agent to push issues to the right people, but not to respond to customers.
Example: Do you get a certain class of questions where the user never provides the information you need? Have the AI Agent collect information from the customer but leave it to your team to really solve the issue.
Create different AI Agents for different use cases: Agents (like humans) do best when their task is scoped and they are given access to only the set of knowledge and tools they need.
If your AI agent is not confident in a task, it will not attempt it.
Your AI agent will only attempt to work on issues that is has been assigned to. We provide you full control in determining what issues should have participation from your AI Agent.
If your AI Agent is customer facing and attempting to answer a user question, it will always cite sources directly to the customer, so that the base knowledge can be accessed directly.
