USE CASE · VOICE
Build a voice agent that follows up by email
e2a gives voice agents a real email identity so they can turn calls into follow-ups, receive replies, preserve conversation context, and route sensitive commitments through human approval.
THE WORKFLOW
What the agent does
- 01A voice agent completes a call and records the outcome.
- 02The application creates a concise follow-up from the call context.
- 03e2a sends the follow-up from the voice agent's verified address.
- 04The recipient replies by email and e2a delivers the message to the agent.
- 05The agent answers, escalates, or schedules the next action.
- 06Sensitive commitments can wait for human approval before sending.
WHY E2A
- ✓A persistent email identity for the voice agent
- ✓Outbound HTTP API and threaded replies
- ✓Inbound WebSocket, webhook, REST, or MCP delivery
- ✓conversation_id correlation with call or CRM state
- ✓Human approval for commitments and sensitive follow-ups
START BUILDING
The integration stays small
const followUp = await voiceAgent.summarizeCall(call);
await client.messages.send(agentEmail, {
to: [{ email: call.recipient }],
subject: followUp.subject,
text: followUp.text,
conversation_id: call.id,
});
// Later: receive and reply to the email conversationThe agent owns the business logic. e2a handles the email identity, delivery, authentication evidence, and reply routing.
KEEP GOING
Connect the pieces
FAQ
What can a voice agent do with email?
A voice agent can send a call summary, confirm next steps, receive a reply, answer follow-up questions, and continue the interaction over a persistent email conversation.
Can the email reply reconnect to the original call?
Yes. The application can bind e2a's conversation_id to the call, CRM record, or voice-session identifier so later email turns reconnect to the original workflow.
Can a human review a voice agent's follow-up?
Yes. e2a's human-in-the-loop approval gate can hold follow-ups for review before they are delivered to the recipient.
Does the voice agent need a public webhook?
No. The email side of the workflow can use WebSocket delivery, REST polling, or MCP when the voice application runs locally or behind a firewall.
Build your voice follow-up agent
Open-source, free to start, and ready for your first agent workflow.
Get started →