---
name: cold-email-sequence
description: >-
  Use when the user wants to write a cold outbound email sequence — the
  multi-touch cadence that opens conversations with target accounts. Also use
  when the user mentions cold email, outbound sequence, email cadence,
  prospecting emails, sales outreach, follow-up emails, break-up email, or
  "emails that get replies." Produces a 4–6 touch cold email sequence with
  subject lines, send timing, and one clear ask per email, grounded in the
  prospect's pain.
license: MIT
metadata:
  publisher: AAJ
  slug: cold-email-sequence
  category: Sales & Pipeline
  phase: Execute
  difficulty: Starter
  version: 1.0.0
  agents: [Claude Code, Cursor, OpenAI Codex, Windsurf, Cline]
  inputs: The ICP's sharpest pain, the differentiator and one proof point, the role being emailed, and the relevance hook.
  outputs: A 4–6 touch cold email sequence with subject lines, send timing, the angle per touch, and alternate first-touch subjects.
  related_aaj:
    - https://aajconsult.com/tools
  tags: [cold-email, outbound, sequence, prospecting, copywriting]
---

# Cold Email Sequence

Write a short outbound sequence that earns a reply by leading with the prospect's problem — not your product.

## When to use

When opening conversations with a target account list, when current outreach gets ignored, or when launching a new outbound motion.

## Before you start

1. **Read the brand/product context first.** Pull the ICP and its single sharpest pain, plus the differentiator and one piece of proof, from `.agents/product-marketing.md`. If none exists, ask the user.
2. **Gather inputs:** who's being emailed (the role) and the relevance hook that triggered the outreach (a signal, an event, a shared context). If the account list / ICP isn't set, run `target-account-list` and `brand-product-context` first.
3. **Confirm the objective:** opening a conversation. (For nurturing existing leads, that's a lifecycle sequence — a different intent.)

## Method

Anchor every email on one specific pain the role feels, and open in their world rather than with your company. One email, one idea, one low-friction ask. Keep each under ~90 words, plain text, written like a person — and vary the angle across touches instead of just bumping the thread.

## Workflow

1. **Anchor each email on one specific pain** the role feels.
2. **Give each one ask** — low-friction (a question, a small yes), not "book a 30-minute demo" on touch one.
3. **Keep it short** — 50–90 words, plain text, no jargon.
4. **Use proof sparingly** — one concrete outcome or number beats adjectives.
5. **Vary the angle across touches** (pain → proof → a different pain → break-up).
6. **End with a respectful break-up**, then present the full sequence.

## Reference

This skill bundles no data files. Pull the pain, differentiator, proof, and voice from `.agents/product-marketing.md`; write in the brand's voice.

## Present the result

- **4–6 emails**, each with: a subject line (short, lowercase, relevance or curiosity — no clickbait), the body, and the single ask.
- **Send timing** (e.g., day 1, 3, 6, 10, 14).
- The **pain/angle** each touch uses.
- **2–3 alternate first-touch subject lines** to test.

## Guardrails & common mistakes

- **Lead with them, not you.** If an email is about your company instead of the prospect's problem, rewrite it.
- **One ask per email.** Multiple asks (or a hard demo ask on touch one) kill reply rates.
- **Keep it under ~90 words.** Long cold emails don't get read.
- **Back every claim.** Make no outcome claim that isn't supported by real proof from context — invented numbers destroy trust and deliverability reputation.

## Related AAJ resources

- AAJ tools — https://aajconsult.com/tools (copy and outbound tools).

## Related skills

`target-account-list` · `discovery-call-framework` · `objection-handling` · `brand-product-context`

## Credits

Original AAJ skill. The Agent Skills format and the marketing-skills catalog by Corey Haines (`coreyhaines31/marketingskills`, MIT) were references for structure and coverage; this skill is independently written. See the repository README for the full reference list.