Automation agent
Playbook Executor
Given a fired signal, drives the matching playbook one step at a time.
What does Playbook Executor do?
Playbook Executor is the AI agent that actually runs your playbooks once a signal fires. Where the Workflow Builder is where you design a sequence - trigger, conditions, actions - Playbook Executor is what steps through it on a live account: sending the next outreach, waiting for a response, checking the condition for the next branch, or marking the sequence resolved. It's the difference between a playbook that exists in a document and one that actually happens.
What signals does it watch?
Playbook Executor doesn't watch for the original risk or opportunity signal itself - that's the job of agents like Champion Loss Detector or Renewal Risk Assessor. It watches the state of an in-progress playbook run instead.
- A new signal has fired that matches a workflow with a seeded or custom playbook attached
- A wait period inside a playbook has elapsed, and the next step is due
- A condition inside the playbook needs re-checking - has the account's health improved, did the champion respond
- A playbook step has no further action defined, which hands off to Escalation Handler rather than silently stopping
What actions does it take?
What it actually does at each step depends entirely on what that step is - this agent executes the workflow you built rather than deciding on its own what should happen.
- Sends the next outreach email in the sequence, drafted from your templates and account context
- Advances the account to the next step once a wait period or condition is satisfied
- Marks the sequence resolved when the exit condition is met - for example, the champion responds, or health recovers
- Any email it sends appears in the account's Customer 360 timeline, so the full sequence is auditable after the fact
Which workflows use Playbook Executor?
Playbook Executor runs inside every seeded workflow that includes a sequenced outreach step - it's the engine behind Champion Loss, Onboarding Stall and Renewal Risk in particular, and behind any custom workflow you build with a playbook attached.
What does this look like in practice?
Champion Loss Detector flags an account and the seeded Champion Loss workflow fires. Playbook Executor picks it up: sends a first check-in email to the new contact identified on the account, waits three days for a response, and - when none arrives - sends a second, shorter follow-up. A reply comes in on day five. Playbook Executor reads the response as engagement, marks the sequence resolved, and logs the full exchange to the account timeline for whoever opens it next.
Frequently asked questions about Playbook Executor
- Does Playbook Executor decide what the playbook does, or just run it?
- Just run it. You (or the ten seeded workflows) define the trigger, conditions and actions in the Workflow Builder; Playbook Executor is the agent that steps through that sequence on real accounts.
- What happens if a step in the playbook has no handler?
- It doesn't fail silently - the canvas itself refuses to publish a workflow with a step the engine can't execute, and if a live run somehow stalls, Escalation Handler raises an internal task rather than letting it loop.
- Can I see what it sent and when?
- Yes - every email Playbook Executor sends appears in that account's Customer 360 timeline, threaded like any other interaction, not hidden in a separate automation log.
- Does it ever send without approval?
- Only if the workflow's tier for that segment is set to fully autonomous and all four of the email send gates are satisfied - otherwise the message waits as a draft for a human to approve.
See Playbook Executor on your own accounts.
Book a 30-minute demo, or connect your stack and try it free for a week.