Plan requirement
| Subscription | Suite Team or higher, Support Team or higher |
| Access | Admin |
An AI agent answering email. Higher risk than messaging, and worth doing only where the questions are genuinely repetitive.
Connect it
- In the AI agents workspace, click Dashboard in the sidebar.
- Hover over the agent, which will be showing Ready to activate, and click Activate.
- In Manage your channels, select the email channels, which include the API and web form, and click Save changes. To change that later, use Manage channels in the options menu.
- Scope which tickets it should answer, and decide what happens when it cannot. Both are below.
- Test by sending real messages from outside.
An agent created for email channels cannot later be activated on messaging, and only one AI agent can be active on a channel at a time. Both are decisions you make once.
Connecting it creates a trigger for you
Selecting an email or web form channel automatically creates a ticket trigger for that agent, called Ultimate.ai Multitouch Ticket Automation followed by the agent's ID, and it becomes active when you publish. It watches for public comments on tickets that are created or updated on that channel. Look for it under Triggers in Admin Center, or through Settings, then General, then Go to triggers in the AI agents workspace.
Two things about it matter here. It is configured by default not to respond to tickets submitted by agents, which is why a test sent from your own account produces nothing; you can remove that condition while testing. And in the customer's inbox the sender is the brand name and support address you chose in Admin Center, while in the Agent Workspace the reply comes from the system user unless you point it at a real user account, which then consumes a seat and shows that person's name and signature.
Why email is harder
No clarification. The customer wrote what they wrote and has gone. If the agent misreads it, the reply is wrong and arrives as your company's considered answer, in writing, in their inbox.
On messaging a misunderstanding costs one exchange. By email it costs the interaction.
Scope it narrowly
Use tags on users and organisations so the agent only answers the email you are confident about. Everything else goes to a person, untouched. An Advanced AI agent reads the fixed tag escalated_by_ultimate; for an Essentials one you add the condition to its trigger yourself.
Broad email automation is where deployments get switched off.
Dialogues do not carry across
An email AI agent cannot use dialogues. Anything you scripted for messaging does not come with it, and on email the flow comes from generative procedures instead. Worth planning for rather than discovering after you connect the channel.
Long emails, several questions
Customers write three questions in one message. Decide what should happen when only one is covered: answering one and ignoring two reads as not having read it.
Handing over is usually the right choice there.
Say it is automated
In the email itself, with a clear route to a person. An automated reply presenting itself as a colleague is the version people resent.
Watch replies to replies
When the customer answers the automated email, that continuation needs to reach a person unless you are confident it does not. This is where loops start.
Read the first week in full
Every email the agent sent. It is a finite number at the start and it is the only way to catch the answers that were confidently wrong.
Comments
0 comments
Article is closed for comments.