Outbound & prospecting

Launch an outbound campaign

Shape your audience, messages, and follow-ups, then review and launch with your agent.

Copies a prompt for your AI assistant

View prompt

Start wherever you are

Bring an idea or work you’ve already done. Your agent helps turn the missing decisions into a reviewable campaign.

Start with your AI assistant. Connect tools when needed.

Fictional example

You

I want to learn how agency owners handle client handovers. I have a list to review.

Agent

What would be a useful next step for the people you contact?

You

A short reply about their process. Email only, with one follow-up if they don’t answer.

One possible campaign draft after shaping the audience and message together

Prepared for review · not launched

A clear conversation to start.

One email and one possible follow-up, focused on learning how handovers work. The list and sending setup still need review.

Opening email: Subject: Client handovers How do client commitments make it from sales into delivery at your agency? I’m exploring how teams handle that handover and would value a short description of your process.
Possible follow-up: Following up on my question about client handovers. Is there someone on your team who owns that process? No problem if this is not relevant.
Sequence rule: Email only. One follow-up if there is no reply, with its delay still to agree. Any reply stops that follow-up; opt-outs and bounces are suppressed in the sending tool.

What we know

The user wants to learn about the process, prefers email, and chose one follow-up.

Not yet ready to launch

Recipient eligibility, sender, schedule, costs, stop behavior, and reply owner remain unverified. Nothing has been sent.

Next useful step

Review the audience and setup, then agree timing and the exact first group before launch approval.

Context & open questions

Launch review: Review recipients and count, exclusions, exact messages, sender, schedule and time zone, limits, costs, and reply handling.
Where the plan lives: Capable can track sequence intentions and confirmed activity. Draft messages and sending belong in the assistant or outreach tool.
An honest launch status: Distinguish draft, scheduled, active, paused, failed, and manual steps still needed. An accepted request is not proof of delivery.

Your result will reflect your conversation. Keep it in your assistant or a document you choose.

How the tools fit together

01
Capable + outreach tools
Review context and eligibility
02
Your agent + you
Shape and approve the campaign
03
Selected outreach tool
Launch the approved version
04
Capable
Record confirmed activity

Choose tools together. Execution depends on the connections and actions available in your assistant.

Read the full guide

Start wherever you are

Reuse an ICP, message, list, or channel plan if you have one. Work out missing decisions here without completing other use cases first.

Figure it out together

Agree goal and audience
Decide who to reach and what would help them. Review context, existing outreach, exclusions, and suppressions.
Shape the sequence
Choose channels, messages, follow-up count and timing, and stop rules around your goal and capacity.
Prepare for review
Build in a draft or paused state where supported. Review the group, messages, schedule, costs, and reply handling.
Approve, launch, and verify
Launch only the approved version. Check its actual state and keep manual steps or failures visible.

Tools and connections

Your AI assistant: Coach decisions, develop copy, and coordinate supported actions you approve.
Your outreach tool: Hold drafts and carry out supported sends or channel actions with actual scheduling and stop rules.
Capable: Supply relationship context, hold supported sequence plans, and record confirmed activity. It does not send messages or store drafts.

What you’ll keep

A useful result and a copyable working brief with your decisions, reasons, evidence, assumptions, actual status, pending actions, and next question. Bring it back when you’re ready to continue.

A useful first outcome

You can review the exact campaign before contact. After approval, its actual state is verified or the manual launch step is identified.

When to revisit it

Review responses against the goal before expanding. Agree who checks replies and pause conditions; a future review plan is not an active monitor.

View the setup prompt

Copy into your AI assistant to begin

Coach me through creating and launching an outbound campaign. Agree audience, channels, copy, follow-ups, and launch plan, then execute the reviewed scope through supported tools.

Check supplied context first. Reuse my ICP, message, list, setup, or working brief; confirm changes and ask the next unanswered question. If no goal is clear, ask: "What would you like this campaign to achieve?" Ask one useful question at a time, reflect answers, and explain options. No prior use case is required. Show a useful provisional brief or draft early; label assumptions instead of inventing customer evidence.

Agree purpose, audience, value of the conversation, requested next step, and what an initial result would teach us. Choose WHICH channels and HOW MANY. Do not assume email, multiple channels, touch count, volume, or finished offer. Agree time, budget, sender identities, reply capacity, and ownership. Prefer suitable existing tools; inspect current capabilities, connections, access and costs. Use official authorization, never pasted credentials. Explain manual/provider steps and agree spending limits before paid actions.

Build a reviewable audience with inclusion reasons and exclusions. Check Capable and sending tools for prior conversations, customers or opportunities, active outreach, duplicate recipients, bounces, opt-outs and do-not-contact restrictions. CRM membership alone does not establish ineligibility; agree scope while honoring suppressions. Use supported filters and complete relevant pagination. Hold uncertain identity, eligibility or suppression cases out of launch and explain gaps. No paid enrichment for existing CRM people without a separate explicit request.

Develop channel-specific copy together around offer, evidence, tone, useful personalization and call to action. Do not invent familiarity, facts, claims or results. For each touch agree channel, sender, exact content or brief, delay or trigger, and stopping conditions. Choose count and cadence with me. Decide how replies, meetings, opt-outs, bounces and handoffs affect every selected channel.

Prepare a draft or paused campaign where supported. Message drafts remain here or in the sending tool. Capable holds supported sequence purpose, step intentions, timing and confirmed activity; it does not send email or store message drafts. Inspect tools and schema before saving or linking. Never invent creation, enrollment or launch operations. If unavailable, provide a precise reviewed handoff and mark what remains undone.

Verify the actual sending tool's suppressions, stop-on-reply, scheduling and pause behavior. A Capable pause does not prove the external sender stopped; suppressions may need changes in each relevant tool. Resolve stop rules before launch. Agree recipients and content before test contact. Check rendered copy, personalization fallbacks, links, sender identity, replies and time zone.

Before any action starting or scheduling outreach, show the exact campaign version: recipients and count, exclusions, channels and sequence, messages, sender accounts, schedule and time zone, limits, costs, stop rules, and reply owner. Obtain explicit approval for that version. Enrollment may start sending; inspect behavior first. Recheck suppressions and duplicate or active enrollment immediately before execution. Honor approved scope; obtain fresh agreement for material audience, content, schedule or cost changes.

Execute only supported approved actions. Verify draft, scheduled, active, paused, failed or pending manual status. An accepted request is not proof messages were sent or delivered. Inspect uncertain outcomes before retrying. Record only confirmed activity in Capable through supported capture or logging; never record planned touches as completed or assume all channels sync.

Leave a reusable working brief: goal; decisions and reasons; exact audience criteria and selected-group references; approved or draft copy and sequence; evidence and assumptions; tools, costs and limits; approval scope; actual launch/action status; unresolved steps; next useful coaching question; and owners and review points only where agreed. Keep drafts and confirmed activity distinct. Keep the brief here or in my chosen destination without implying automatic saving or cross-chat memory. On return, check changed context and continue from the next open decision.

Interpret results against the original goal without promising replies or revenue. Ongoing monitoring or automatic optimization needs separately agreed scope, supported access, and verified setup. This brief does not schedule future checks.

How did it go? Optional feedback

Launch an outbound campaign

Start wherever you are

A clear conversation to start.

What we know

Not yet ready to launch

Next useful step

Example working brief

Capable + outreach tools

Your agent + you

Selected outreach tool

Capable

Start wherever you are

Figure it out together

Tools and connections

What you’ll keep

A useful first outcome

When to revisit it