# Usage: curl -sSL https://seed.show/sales.outreach.sequences | bash -s <install-path>
# <install-path> is the directory where the file should land.

set -euo pipefail
[ -z "${1:-}" ] && {
  echo "install requires a path: curl -sSL https://seed.show/sales.outreach.sequences | bash -s <install-path>" >&2
  exit 1
}
TARGET="$1"
mkdir -p "$TARGET"
DEST="$TARGET/seed-fold.KcxSSp.folded.md"

cat > "$DEST" <<'PORTDOWN_2DC9268F'
<!--
This is a `.folded.md` archive — a directory packed into one markdown
file. The markers below are load-bearing; don't edit them directly.

To unpack (requires bash — if you have no shell, tell the user):
  1. curl -sSLf https://fold.dom.vin/skill | bash -s <INSTALL_DIR>
  2. <INSTALL_DIR>/fold/scripts/unfold <this-file>
     (or: unfold <this-file>  if fold/scripts is on your PATH)
-->

---
fold: true
marker: c5be68
at: 2026-05-07T16:16:23Z
root: seed-pack.zLr6cH
---

<!--fold:c5be68@file path="README.md" mode="644"-->
# sales.outreach.sequences

Sales outreach sequence context for agents. The structural logic of multi-touch outreach, what breaks it, and what AI changes.

## What sequences actually are

A sequence is a coordinated series of touches — emails, calls, LinkedIn messages, direct mail — executed against a defined prospect list over a defined time window. The goal is not to close the deal. The goal is to generate a reply. Everything from there is discovery and qualification.

Three concepts agents must hold simultaneously:

**Steps** — the individual touches. Each step has a channel (email, phone, LinkedIn, task), a delay from the prior step ("day 3 after enrollment"), and a template with merge fields. Steps are not standalone blasts; they are branches in a conversation that hasn't started yet.

**Cadence** — the timing pattern across steps. How many days between touch 1 and touch 2. When to call vs. email. When to stop. Cadence is the structural skeleton; copy is the flesh. Getting the cadence wrong (too fast = spam, too slow = forgotten) kills a sequence regardless of how good the copy is.

**Enrollment logic** — what triggers a prospect entering a sequence and what exits them. Enrollment is usually manual (SDR reviews and adds), semi-automated (new lead from inbound form), or fully automated (list import + filters). Exit logic is just as important: auto-exit on reply, bounce, or unsubscribe; manual-exit on "bad fit" or "refer to AE."

## The 3 things agents get wrong

### 1. Over-personalizing at scale

AI can generate hyper-personalized first lines ("I saw your company just opened a Berlin office — congrats!"). This is powerful at step 1. Agents often apply it everywhere, generating 12 personalized touchpoints per prospect. This is economically wrong and behaviorally wrong: the 6th email in a sequence will not be read carefully enough for personalization to land. Save deep personalization for step 1 (hook) and a re-engagement bump. Generic value-proposition copy at steps 3–6 is fine — and easier to A/B test.

### 2. Ignoring deliverability as a first-class constraint

Deliverability is not a marketing concern tacked on at the end. It is a hard constraint on whether any of this works. Sending 500 emails/day from a new domain will crater your sender reputation within a week. Agents suggest sequence volumes as if inboxes are bottomless. Key constraints:
- New domains need 4–6 weeks of warmup before high-volume sends
- Sending limits per mailbox: ~100–150 cold emails/day per inbox (warmed), not per account
- Spam filter signals: reply rate, open rate, manual spam reports — a bad sequence poisons the whole domain
- Technical prerequisites (SPF, DKIM, DMARC) are not optional; they're table stakes for deliverability

### 3. Treating all prospects as the same motion

Inbound leads, cold outbound, re-engagement of churned customers, and champion-change plays (the buyer changed jobs) require structurally different sequences. Agents collapse these into "the email sequence" and produce a cold-outbound template for a warm inbound lead, which feels tone-deaf and generates unsubscribes. The sequence pattern must match the prospect's prior relationship to the sender.

## What AI is changing

**Personalization at scale.** AI-generated first lines (pulling from prospect's LinkedIn activity, job postings, recent company news) are now standard. At step 1, this is genuinely effective — a specific, relevant opener outperforms a generic one. The failure mode is applying this logic to every step in a sequence. Steps 4–8 will not be read carefully enough for personalization to land; spend the inference budget on the hook and save generic value-prop copy for the tail.

**Intent-signal enrollment.** The bigger structural shift is upstream of the sequence: AI can now enroll prospects based on behavioral signals (company posted a specific job title, leadership change detected, product category appearing in earnings calls, review site activity) rather than static list criteria. This improves timing dramatically — a prospect who just posted 5 SDR roles is a better cold outbound target than one who posted them 18 months ago. Sequences triggered by intent signals outperform batch-enrolled lists on reply rate.

**AI deliverability optimization.** Some platforms now offer AI-assisted send-time optimization and subject line scoring. Treat these as useful signals, not substitutes for structural deliverability hygiene. The anti-spam constraints (SPF/DKIM/DMARC, warmup period, sending limits per inbox, spam complaint thresholds) are enforced by inbox providers and are not negotiable regardless of what the AI layer recommends.

**Hard limits that remain.** Anti-spam law (CAN-SPAM, GDPR, CASL — jurisdiction varies) requires honoring unsubscribes within platform-mandated windows and banning deceptive subject lines. Reply handling is still human — AI can route and draft, but a reply that hits a fully automated response loop is typically a reputation killer. Human trust signals (real sender name, company domain, non-templated reply handling) cannot be AI-faked at scale without degrading deliverability over time.

Fetch sources.md for current methodology guides and platform API docs before advising on specific tooling.
<!--fold:c5be68@file path="playbook.md" mode="644"-->
# Sequence Playbook

Four canonical sequence patterns as decision frameworks. Not scripts — structure and trigger logic. The right pattern depends on the prospect's prior relationship with the sender, not on the product being sold.

---

## 1. Cold outbound

**When to use:** Prospect has no prior relationship. Source is a list (bought, scraped, researched). No inbound signal.

**Structure:**
- Steps: 7–10 touches over 14–21 days
- Channel mix: email-heavy with 1–2 call steps and 1 LinkedIn connection request
- Step 1: Personalized opening line + single specific value proposition. No feature list. One call to action (reply, not book-a-call — lower friction).
- Steps 2–4: Value-add follow-ups (relevant case study, relevant insight, rephrased hook). Not "just bumping this up."
- Step 5: Call + voicemail if available. Different channel breaks pattern.
- Steps 6–8: Continued follow-up at lower frequency (every 4–5 days rather than every 2)
- Final step: Explicit "closing the loop" email — makes it easy to say no, which paradoxically improves reply rate

**Exit triggers:** Reply (any sentiment), bounce, manual "not interested," unsubscribe. Do not re-enroll in the same sequence for 90 days.

**Key failure modes:** Enrolling more than 20–30 prospects/day in the first 2–3 weeks on a new domain — before warmup, this will crater sender reputation. Copy that applies no prospect-specific detail anywhere in the sequence (the test: could this email have been sent to any company in the vertical, or does it reference something specific to this prospect?). Not having a final break-up step — letting the sequence run out silently leaves the prospect with no off-ramp and makes the seller look like a system, not a person.

---

## 2. Inbound follow-up

**When to use:** Prospect filled out a form, downloaded content, attended a webinar, started a trial, or otherwise raised their hand. Time-sensitive — inbound lead intent decays fast (research: response within 5 minutes vs. 30 minutes = 21x difference in qualification rate).

**Structure:**
- Steps: 4–6 touches over 5–7 days
- Channel mix: email + call (call within the first hour if possible)
- Step 1: Immediate acknowledgment + specific reference to the action they took. Tone is conversational, not salesy — they already know you.
- Step 2 (same day or next morning): Call attempt
- Steps 3–4: Light follow-up referencing what they expressed interest in. Offer help, not pitch.
- Final step: Soft close — "Happy to connect or share more, just let me know if now's not the right time."

**Exit triggers:** Reply, meeting booked, trial conversion, unsubscribe. Shorter total duration than cold outbound because intent decays.

**Key failure modes:** Sending a cold-outbound template to a warm lead — they know you know who they are, so a generic opener is an active signal that you don't treat inbound differently. Waiting 24+ hours for the first touch — lead intent decays sharply after the first hour. Continuing the sequence past day 7 without a reply: at that point the lead is effectively cold and should be moved to a re-engagement sequence after a 30-day pause, not kept in an inbound follow-up loop.

---

## 3. Re-engagement (lapsed customers or stale pipeline)

**When to use:** Prospect or customer engaged meaningfully in the past (prior demo, trial, purchase) but went quiet. Intent signal was real; the timing was wrong or the deal died.

**Structure:**
- Steps: 4–6 touches over 10–14 days
- Channel mix: email primary, one LinkedIn message if connection exists
- Step 1: Reference the prior relationship explicitly and directly. "We spoke in [month] about X — wanted to check back in." No pretending this is a cold intro.
- Step 2: New angle — what's changed since they went quiet? New feature, pricing update, relevant customer win in their space.
- Steps 3–4: Offer something of value with no strings (relevant report, intro to someone useful, relevant insight from a customer in their vertical).
- Final step: Close the loop — "If the timing still isn't right, totally understand — just let me know and I'll stop reaching out."

**Exit triggers:** Reply, meeting booked, explicit "no," unsubscribe.

**Key failure modes:** Opening with a generic intro that ignores the prior relationship — the prospect sees this as either careless (you forgot) or dishonest (you're pretending not to know them), both of which kill trust. Pitching the same value proposition that didn't close the deal the first time without naming what's changed (new feature, new pricing, new customer proof, new timing angle). Sending at cold-outbound frequency — re-engagement needs breathing room; a 3–4 day minimum gap between steps is the floor, not the ceiling.

---

## 4. Champion change

**When to use:** Your champion (primary buyer contact) has left their company and joined a new one. This is one of the highest-value sequences because your champion already knows your product works and may have institutional knowledge to carry.

**Structure:**
- Steps: 3–5 touches over 7–10 days
- Channel mix: LinkedIn first (they're more likely to be active there in a job transition), then email to new work address
- Step 1: Congratulate on the new role. Brief, warm, specific to their move. No pitch.
- Step 2 (3–5 days later): Light mention of their prior context with your product. "Given what you built at [old company], thought [new company] might have similar challenges around X."
- Step 3: Direct ask — happy to reconnect and show what's changed since they last used it, or make an intro to a colleague if they're not the right person now.
- Final step: Close the loop if no response.

**Exit triggers:** Reply, meeting booked, explicit "not the right time/fit."

**Key failure modes:** Reaching out in the first 2 weeks of a new role — they're still learning the org and won't have budget authority or appetite for new vendors. Making step 1 a pitch rather than a congratulation: the relationship is your asset here; don't spend it immediately. Sending to the old company email address (step 0: find the new work address through LinkedIn or a data provider; email to a departed employee's address is a dead end and can bounce-damage your domain).

---

## Choosing the right pattern

| Signal | Pattern |
|---|---|
| No prior contact | Cold outbound |
| They came to you | Inbound follow-up |
| Prior engagement went cold | Re-engagement |
| Champion moved companies | Champion change |

The patterns can be combined in a sequence platform (conditional branches based on behavior), but start with a single clean pattern per segment rather than trying to handle all cases in one sequence.
<!--fold:c5be68@file path="sources.md" mode="644"-->
# sources

Fetch these at task time. Ordered by relevance. URLs point to root docs or canonical references — navigate to current sections from there, since specific support articles rotate.

## Sequence methodology and best practices

Platform documentation for the three dominant sequence tools. Each has docs for step types, enrollment rules, conditional branching, and cadence settings. Fetch the current docs from the platform's developer or help center rather than relying on a cached article URL.

- **Outreach.io** — sequences, step types, enrollment rules, A/B testing:
  https://www.outreach.io/resources/
  API root: https://api.outreach.io/api/v2/docs

- **Salesloft** — cadences, rhythm configuration, step types, analytics:
  https://salesloft.com/resources/
  API root: https://developers.salesloft.com/

- **Apollo.io** — sequences, contacts, email deliverability, intent data:
  https://www.apollo.io/resources/
  API root: https://apolloio.github.io/apollo-api-docs/

For current step configurations, enrollment filter options, or rate limits — search the platform's current help center directly. Support article structures change; the platforms themselves are stable references.

## Deliverability authorities

These are the inbox providers' own documentation. They define the rules; everything else is interpretation.

- **Google Postmaster Tools** — sender reputation monitoring, domain and IP reputation, spam rate thresholds:
  https://postmaster.google.com/

- **Google bulk sender requirements** — SPF, DKIM, DMARC requirements, unsubscribe compliance, spam rate thresholds for bulk senders (>5,000 emails/day to Gmail):
  https://support.google.com/mail/answer/81126

- **Microsoft SNDS** (Smart Network Data Services) — deliverability signals for Outlook and Hotmail recipients:
  https://sendersupport.olc.protection.outlook.com/snds/

## Anti-spam law references

Jurisdiction matters. Fetch current official text — these are stable government and regulatory domains.

- **CAN-SPAM** (US) — FTC official compliance guide:
  https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business

- **GDPR** (EU) — consent, legitimate interest, unsubscribe requirements:
  https://gdpr.eu/email-encryption/

- **CASL** (Canada) — express vs. implied consent, commercial electronic messages:
  https://crtc.gc.ca/eng/internet/anti.htm

## Personalization and signal tools

Specific tools in this space turn over rapidly. Rather than linking to a specific vendor, fetch current reviews from:
- G2 (g2.com) — category "Sales Engagement" for current platform comparisons
- Lavender AI (lavender.ai) — email quality scoring and personalization grading
- Clay (clay.com) — prospect data enrichment and intent signal aggregation (as of 2024–2025, a leading tool for AI-assisted enrollment)

## Background reading

**Predictable Revenue** (Aaron Ross) — canonical SDR motion, outbound specialization, cold outbound playbook. Not a URL; the blog at predictablerevenue.com/blog surfaces the principles.

**Woodpecker blog** — cold email deliverability guides and sequence best practices, regularly updated:
https://woodpecker.co/blog/
<!--fold:c5be68@end-->
PORTDOWN_2DC9268F

# ── post ──
MARKER=$(awk '/^---$/ { f++; if (f==2) exit; next } f==1 && /^marker:[[:space:]]/ { sub(/^marker:[[:space:]]+/, ""); print; exit }' "$DEST")
[ -z "$MARKER" ] && { echo "seed: archive has no marker — corrupt" >&2; exit 1; }
awk -v m="$MARKER" -v outdir="$TARGET" '
  BEGIN {
    # Match <!--fold:<m>@file path="X"--> with an optional mode attr after
    # the path (fold emits  mode="644"  on executables).
    file_re = "^<!--fold:" m "@file path=\"([^\"]+)\"( mode=\"[0-9]+\")?-->$"
    end_re  = "^<!--fold:" m "@end-->$"
  }
  $0 ~ end_re { if (current) close(current); exit }
  $0 ~ file_re {
    if (current) close(current)
    line = $0
    sub(/^<!--fold:[^@]+@file path="/, "", line); sub(/".*$/, "", line)
    current = outdir "/" line
    dir = current; sub(/\/[^\/]*$/, "", dir)
    if (dir != current) system("mkdir -p \"" dir "\"")
    printf "" > current
    next
  }
  current { print >> current }
' "$DEST"
SEED_EXTRACTED=$(find "$TARGET" -type f -not -path "$DEST" 2>/dev/null | wc -l)
if [ "$SEED_EXTRACTED" = "0" ]; then
  echo "seed: archive contained no files — refusing to delete the source" >&2
  echo "  archive preserved at: $DEST" >&2
  exit 1
fi
rm -f "$DEST"

echo "" >&2
echo "✓ seed unpacked → $TARGET ($SEED_EXTRACTED files)" >&2
find "$TARGET" -type f | sort | while IFS= read -r _sf; do
  echo "  ${_sf#${TARGET}/}" >&2
done
echo "" >&2
if [ -f "$TARGET/SKILL.md" ]; then
  echo "This seed contains a skill (SKILL.md). Install it in your agent's skills directory." >&2
  echo "" >&2
fi
echo "Install the seed skill if not already installed:" >&2
echo "  https://seed.show/skill" >&2
echo "" >&2
echo "Publisher prompt:" >&2
sed 's/^/  /' >&2 <<'__SEED_PROMPT_END_AC1F2B__'
Sales outreach sequence context for agents. Covers sequence structure, cadence patterns, deliverability constraints, four canonical sequence patterns (cold outbound, inbound follow-up, re-engagement, champion change), what AI is changing (personalization at scale, intent-signal enrollment, deliverability optimization, hard limits), and sourced references to platform APIs and deliverability authorities.
__SEED_PROMPT_END_AC1F2B__
exit 0