Sales · Outbound

Cold outbound

RecommendedFull motion: enrichment, deliverability, CRM

Apollo for the leads, Clay for the enrichment + per-row LLM, Smartlead to send at deliverability scale, Claude to write the copy, HubSpot to catch the meetings.

· 5 months ago
SALESINTERMEDIATEIntermediateFrom $108/mo
The stack
Apollo
Apollo
Lead source

275M contacts. Filter by ICP, export to Clay.

Free · $49/seat/mo Basic · $79/seat/mo Professional (annual billing)Alts: Clay
Clay
Clay
Enrichment + per-row LLM

Trigger-based personalization at scale. The competitive moat of modern outbound.

$185/mo Launch · $495/mo Growth
Claude
Copy

Tighter, less salesy hooks than the alternatives. 4-line emails out-perform 8-line.

Free · $20/mo Pro · $100/mo Max 5x · $200/mo Ultra · Sonnet API $2/$10 per M tokens (input/output)Alts: ChatGPT
Smartlead
Smartlead
Send infra

Unlimited inboxes + warmup. The reason your sequences land in inbox not spam.

$39/mo Basic · $94/mo Pro

Free tier covers small teams. Sequences + meetings on the same record.

Free · $20/mo/seat Starter
Real monthly cost
small
$108/mo
Solo, 500 sends/mo
  • apollo$49
  • clay$0 trial
  • claude$20
  • smartlead$39
  • hubspotFree
medium
$342/mo
2 SDRs, 5k sends/mo
  • apollo$79
  • clay$149
  • claude$20
  • smartlead$94
  • hubspotFree
heavy
$1,148/mo
5 SDRs, 30k sends/mo
  • apollo$390
  • clay$349
  • claude$80 API
  • smartlead$179
  • hubspot$150 Starter
Recent changes to this stack
featureOCT 29

Usage and Cost Analytics API launched in beta to track token consumption and costs

Anthropic released the Usage and Cost Analytics API in beta, enabling organizations to monitor token usage and costs across Claude API consumption with filtering and grouping by dimensions like model, workspace, and API key. The API provides /v1/organizations/analytics/usage and /v1/organizations/analytics/cost endpoints. Legacy usage reporting endpoints remain supported.

featureOCT 23

Claude Tag API released in beta for Claude Enterprise organizations on Slack

The Claude Tag API is now available in beta for Claude Enterprise organizations, providing endpoints to read and configure Claude Tag settings on Slack channels and workspaces with Access bundle support. The API requires Admin API keys with read:claude_tag or write:claude_tag scope. This feature is exclusive to Claude Enterprise.

Workflow
  1. 1
    Define ICPApollo

    Apollo: filter to your tightest ICP. Export 1k contacts.

  2. 2
    EnrichClay

    Clay: pull tech stack, headcount changes, recent funding. One LLM call per row to write the hook.

    Prompt · Per-row personalization (Clay LLM column)
    You are writing a 1-line opener for a cold email. ONE specific reference, not two.
    
    Company: {{company_name}}
    Role of recipient: {{title}}
    Recent trigger (pick the strongest of these): {{recent_funding}} | {{headcount_change}} | {{tech_change}} | {{news_event}}
    
    Output ONE sentence (max 22 words) that:
    - references exactly one trigger by name
    - lands a small, accurate observation about what it means for them
    - does not flatter, does not say "I noticed", does not say "congrats"
    - ends in a period, not a question
    
    Output only the sentence.
  3. 3
    Write copyClaude

    Claude writes the 4-line email per persona. Test 3 hooks per ICP.

    Prompt · 4-line cold email by persona
    Write a 4-line cold email.
    
    Persona: {{title}} at {{company_size}} {{vertical}} companies.
    Pain we solve: {{1-line pain}}.
    Proof: {{1 specific outcome with a number}}.
    Soft CTA: {{e.g. "open to a 12-min look next week?"}}.
    
    Constraints:
    - Subject line: 3 to 5 words, lowercase, no clickbait. Output 3 options.
    - Body: exactly 4 lines (line 1 hook, line 2 pain, line 3 proof, line 4 CTA).
    - Total under 80 words.
    - No "I hope this finds you well", no "I'd love to", no "quick question".
    - One CTA only. No P.S.
    
    Output: subject options, then the email body.
    Prompt · 3 hook variants for A/B test
    Generate 3 distinct hooks for the email above. Each hook should test a different angle:
    1. Trigger-based (recent change at their company)
    2. Peer-based (a similar company already does this)
    3. Status-quo cost (what doing nothing costs them)
    
    Each hook is 1 line under 18 words. Output as a numbered list, no commentary.
  4. 4

    Smartlead, 30 sends/inbox/day, 6 inboxes per persona. Warmup always on.

  5. 5
    CatchHubSpot

    HubSpot sequences hand off to AE. Meetings booked = stop the sequence.

What it produced
B2B SaaS, $25k ACV

1.4% reply rate, 0.6% meeting rate at 4k sends/month. 5 closed-won in Q3 from outbound alone, $130k ACV pipeline.

Common pitfalls
Inbox burn

Send from inboxes you don't care about. One sequence's bad luck kills your real domain.

Over-personalized = creepy

Reference one trigger, never two. Two reads as stalking.

No follow-up = no replies

60% of replies come from email 3 or 4. Sequence ≥ 5.

Other ways to do Cold outbound
Weekly changelog

What changed in AI tools and the stacks that ship. One email a week.

Spotted something off?
Wrong price, dead link, stale tool — anything. We review every fix.
Suggest a fix to this stack