B2B Cold Email Sequence Writer — AI Multi-Touch Cadence

Stop writing one email at a time. Give it a list of target companies and it writes the whole cold-email sequence for each — subject lines, bodies, timing, and a breakup email — personalized and grounded in a real fact about that company. Drafts only. You review and send from your own tool.

A single opener gets you one shot. Deals come from the follow-up. This Actor turns target company domains + what you sell + who you are into a complete multi-touch outbound cadence per company: each email personalized, spaced out over days, each with its own angle, ending in a polite last-attempt email. It never sends anything — it hands you ready-to-paste drafts.

🔗 Chain it after your scraper

Already building prospect lists with Google Maps Scraper (compass/crawler-google-places), Google Maps Email Extractor (lukaskrivka/google-maps-with-contact-details) or LinkedIn Profile Search Scraper (harvestapi/linkedin-profile-search)? Pipe the dataset straight in — paste its dataset ID (via the Integrations tab) and get a full, personalized multi-touch cold-email cadence per company, ready to paste into your sending tool. Works with any Maps, LinkedIn or company scraper's output.

Quick start

1. Fill in the input — set the fields in the input tab (prefilled with a working example).
2. Click Start — the Actor runs and writes one record per item to the dataset.
3. Get your data — export to JSON, CSV or Excel, or pull it via the API.

---

The flow

domains + what you sell          enrich each company            AI writes an N-touch cadence          finished drafts
  + sender identity          →     (real firmographics,      →    (intro · proof · case study ·   →    per company: subjects,
                                    ICP fit — optional)            breakup) grounded in a fact          bodies, timing, plain text

1. You give it target company domains, a one-line description of your offer, and your sender name + company.
2. It (optionally) enriches each company via fuze/ai-company-enrichment so personalization is grounded in a real, verifiable firmographic fact and an ICP-fit read — not a guess.
3. AI writes the full cadence for each company: a personalized intro, varied middle touches (social proof, a relevant use case / mini case study, a light follow-up), and a polite breakup email, with sensible day-by-day timing.
4. You get drafts — one complete sequence per company, as structured touches and as clean copy-paste plain text. Nothing is sent.

---

What you get per company

For every company, the Actor returns:

1. touches — the whole cadence as an array (one object per email), each with:
   • step — 1-based position in the sequence.
   • dayOffset — cumulative days from day 0 (e.g. 0, 3, 7, 12), so you know exactly when to send each.
   • channel — always "email".
   • subject — a short, specific, non-spammy subject line.
   • body — a 50–125 word email with one clear call to action, no hype.
   • purpose — the touch's job (e.g. "intro + value", "social proof", "case study", "breakup").
2. The last touch is always a polite breakup / last-attempt email that closes the loop gracefully.
3. personalizationBasis — the single real fact the cadence is anchored on (or "generic — no enrichment data" when none was available).
4. icpFitNote — one sentence on why they're a fit (from the enrichment ICP read), or null.
5. sequencePlain — the entire cadence as clean plain text, ready to copy into your sequencing tool, with no markdown fluff.
6. enrichmentUsed — whether real firmographics actually informed this sequence.

The model only uses facts it's given — it will not invent specifics about a company.

---

Three ways people use it

• SDRs & outbound sales teams — generate a whole multi-touch campaign for a target account list in one run: personalized sales follow-up emails with built-in timing, ready to load into Instantly, Smartlead, Lemlist, Clay, Apollo or HubSpot. Personalization at scale, without writing each cold email sequence by hand.
• Founders doing their own outbound — paste the ten accounts you actually want, get a complete outbound cadence for each, grounded in something real about them, and spend your time replying instead of drafting.
• Agencies running cold email for clients — spin up a fresh, on-brand cadence per client offer across hundreds of prospects. Consistent structure, varied angles, and a clean breakup email every time — drafts only, so you (and the client) stay in control of what actually goes out.

---

Input

Field	Type	Default	Notes
domains	string[]	["stripe.com","notion.so"]	Target company domains, URLs or emails (one per line).
offering	string	— (required)	What you sell — drives every touch.
senderName	string	— (required)	Your name, used to sign the emails.
senderCompany	string	— (required)	The company you're reaching out on behalf of.
numberOfTouches	integer	4	Emails in the cadence (min 2, max 7). Last one is always the breakup.
tone	string	friendly, concise, no hype	Email tone.
outputLanguage	string	en	Language for all subjects and bodies.
useEnrichment	boolean	true	Ground personalization in real firmographics + ICP fit.
maxResults	integer	50	Max companies (one charge each). Max 1000.
maxConcurrency	integer	4	Companies written in parallel. Max 15.
model	string	gpt-4o-mini	OpenAI model.

Minimal input example

{
  "domains": ["stripe.com", "notion.so"],
  "offering": "We sell a SOC2 compliance automation platform for SaaS companies.",
  "senderName": "Fredrik Hansson",
  "senderCompany": "Fuze",
  "numberOfTouches": 4
}

---

Output example

{
  "domain": "stripe.com",
  "companyName": "Stripe",
  "personalizationBasis": "Stripe sells payments infrastructure to internet businesses, so audited security posture is core to its customers.",
  "icpFitNote": "A large payments platform whose customers demand strong, continuously-audited compliance — a natural fit for compliance automation.",
  "enrichmentUsed": true,
  "touches": [
    {
      "step": 1,
      "dayOffset": 0,
      "channel": "email",
      "subject": "SOC2 evidence, minus the screenshots",
      "body": "Hi — Stripe's customers lean hard on your security posture, which usually means a lot of manual evidence-gathering at audit time. We automate SOC2 evidence collection and control monitoring so teams stop screenshotting. Worth a quick look to see if it saves your team a week each cycle?\n\nBest,\nFredrik Hansson, Fuze",
      "purpose": "intro + value"
    },
    {
      "step": 2,
      "dayOffset": 3,
      "channel": "email",
      "subject": "How teams your size cut audit prep",
      "body": "Following up — teams running continuous compliance usually cut audit prep from weeks to days once evidence is collected automatically. Happy to share how a similar platform team set it up. Open to a 15-minute walkthrough next week?\n\nFredrik",
      "purpose": "social proof"
    },
    {
      "step": 3,
      "dayOffset": 7,
      "channel": "email",
      "subject": "A quick before/after",
      "body": "One more useful data point: a SaaS team we work with went from manual quarterly evidence pulls to always-on monitoring, and passed their renewal with zero last-minute scramble. If that's a pain for your team, I can send the short write-up. Want it?\n\nFredrik",
      "purpose": "case study"
    },
    {
      "step": 4,
      "dayOffset": 12,
      "channel": "email",
      "subject": "Should I close the loop?",
      "body": "I don't want to crowd your inbox, so this is my last note. If automating SOC2 evidence isn't a priority right now, no problem at all — just let me know and I'll step back. If timing changes down the line, I'm easy to reach.\n\nThanks for reading,\nFredrik Hansson, Fuze",
      "purpose": "breakup"
    }
  ],
  "sequencePlain": "Day 0 — Subject: SOC2 evidence, minus the screenshots\nHi — Stripe's customers lean hard on your security posture... \n\nDay 3 — Subject: How teams your size cut audit prep\nFollowing up — teams running continuous compliance... \n\nDay 7 — Subject: A quick before/after\nOne more useful data point... \n\nDay 12 — Subject: Should I close the loop?\nI don't want to crowd your inbox, so this is my last note..."
}

---

Pay-per-event pricing

This Actor uses pay-per-event billing with a single event:

Event	Price	Charged when
sequence-generated	$0.20	Once per company turned into a complete personalized multi-touch cold-email sequence. Covers the optional enrichment and the AI drafting of every touch.

The Actor logs its OpenAI token usage and estimated AI cost at the end of every run, so the event price always stays safely above the model cost.

---

Setup

The AI layer calls the OpenAI API. Set the environment variable / secret OPENAI_API_KEY on the Actor. Without it, the run stops with a clear error.

---

Compliance

This Actor generates drafts only — it never sends any email. You review every sequence and send it from your own outreach tool, so you remain in control of consent, sending volume, local anti-spam law (GDPR / CAN-SPAM / PECR) and deliverability. Treat the output as a first draft to edit, not a send button.

---

Use it from an AI agent (Apify MCP server)

This Actor is available through the Apify MCP server, so AI agents (Claude, ChatGPT, or anything that speaks MCP) can call it as a tool — hand it domains, offering and your sender identity, and read back a complete cold-email cadence per company — with no custom integration code.

---

Run it locally

npm install
# put your input in storage/key_value_stores/default/INPUT.json
export OPENAI_API_KEY=sk-...   # required — powers the cadence writing
apify run -p

---

Related actors

• AI Cold Email Personalizer — the single-opener sibling: one personalized hook, subject and draft per prospect.
• AI Company Enrichment — turn domains into structured firmographics, contacts and ICP fit (the enrichment layer this Actor uses).
• B2B Intent Radar — surface accounts showing buying intent before you reach out.
• Email Finder — find the addresses to send your finished sequence to.