Skip to main content
Every BDR agent has a detailed activity feed that shows a chronological log of every action it has taken. This is your window into what the agent is doing and how it’s performing — and it works the same for both Template and Personalised agents.

​
The agent detail page

Go to Discovery → AI Agents and click an agent’s name. The detail page shows:

​
Stats overview

At the top, you’ll see key performance metrics:
  • Emails sent — total outreach emails dispatched
  • Budget spent — how much of the monthly budget has been used (with remaining amount)
  • Companies processed — how many companies the agent has worked on

​
Agent status

The current state of the agent: If the agent is active but idle, you’ll see the idle reason — this tells you why it’s waiting (e.g. “outside operating hours”, “no companies to process”, “budget limit reached”).
An agent in simulation mode (no connected email account) shows the same feed, but “Send Email” entries record what it would have sent rather than a real send.

​
The activity feed

Below the stats, the activity feed shows every action in reverse chronological order. Each entry includes:
  • Timestamp — when the action happened
  • Action type — what the agent did (see below)
  • Company — which company the action was for
  • Status — success, failed, or pending
  • AI reasoning — the agent’s explanation of why it took that action

​
Action types

​
AI reasoning

For each action, the agent records its reasoning — why it chose this action, why it picked this contact, or why it’s escalating. This transparency lets you understand the agent’s decision-making and adjust your instructions or playbook if needed.

​
The outreach lifecycle

Each company the agent works on moves through a series of outreach statuses. These drive which companies the agent picks up next and how they appear on the Lead Sheet.

​
Following one company’s journey

For any company, you can see the progression:
  1. Contact selection — the agent evaluates contacts, prioritising economic buyers and senior decision-makers
  2. Outreach — it sends an initial email
  3. Follow-up — if there’s no reply, it follows up on your cadence
  4. Response handling — when a prospect replies, the agent classifies the intent and decides whether to escalate

​
Response classification

When a prospect replies, the agent works out their intent:

​
Escalations, drafts and takeovers

When a prospect replies, the agent either escalates the company, drafts a reply for your review, or steps aside if you’ve taken over. These all surface in the activity feed and on the Lead Sheet. See Managing replies and taking over for the full detail on handling escalations, reviewing and sending drafts (as the agent or as yourself), taking over a conversation, and putting a company on a Not-Interested hold.

​
Blocked emails

If a safety check ever blocks an outbound email (for example, it spotted a problem with the content), the feed records it with a View Full Email button. This opens the exact email the agent tried to send and the reason it was held back, so you have a complete audit trail. The agent automatically has another go at composing a corrected email.

​
Controls

From the agent detail page, you can:
  • Pause — stop the agent sending new outreach (it still handles replies)
  • Resume — restart a paused agent
  • Edit — change instructions, templates or playbook, budget, or other settings
  • Delete — permanently remove the agent and all its activity history
Review escalations and draft replies regularly — these are the moments where human judgement makes the biggest difference. The agent handles the volume; you handle the nuance.