Outbound Sales Playbook: Building a Predictable Cold-Sourcing Engine

Most B2B SaaS founders try outbound twice. The first attempt: founder sends 200 cold emails personally; books 15 demos; closes 3; calls it "outbound works." The second attempt (after hiring an SDR or outsourced firm): SDR sends 20,000 cold emails over 3 months; books 6 demos; closes 0; founder concludes "outbound doesn't work." Neither conclusion is right. The real lesson: outbound at scale is a SYSTEM — research, list-building, sequence design, deliverability, qualification, handoff to AE, measurement — and it requires investment that founders consistently underestimate.

A real outbound function is multi-channel (email + LinkedIn + sometimes phone), tightly integrated with marketing intent signals, ICP-disciplined, scaled via SDRs (or AI agents in 2026), measured at every step, and ruthless about dropping accounts that don't fit. Done well, outbound contributes 30-60% of pipeline at growth-stage B2B and is the most-controllable lever in the GTM stack. Done badly, it burns reputation, destroys deliverability, and produces a steady stream of false-positive demos that waste AE time.

This is distinct from Cold Outreach (the messaging discipline), Account-Based Marketing (a strategic motion that incorporates outbound), and Demand Generation Playbook (broader system across all channels). This article is about OUTBOUND specifically — building the cold-sourcing engine end-to-end.

What Done Looks Like

A working outbound sales function produces:

Predictable, measurable pipeline contribution (30-60% of new pipeline at growth stage)
ICP-fit accounts targeted (not random)
Multi-channel cadences executed at scale
Email deliverability protected (sender reputation > 95)
Demo-set rate that supports unit economics (e.g., 1-3% of cold contacts → demo)
Demo-to-opportunity conversion 50%+ (handoff quality good)
AE feedback loop functional (which leads convert; which don't)
Per-rep + per-segment KPIs trackable
CAC at acceptable level (LTV/CAC > 3; payback < 18 months)
Tech stack tight (no random tool sprawl)
Compliance discipline (CAN-SPAM, GDPR, privacy)
Continuous experimentation cadence (new sequences, new offers)

This pairs with Cold Outreach (messaging), Sales Engagement Platforms (Reference), Sales Intelligence & Prospect Data (Reference), Account-Based Marketing, Demand Generation Playbook, Channel Selection, Sales Playbook, Sales Discovery Call Playbook, Sales Forecasting & Pipeline Management, Sales Operations Playbook, First Sales Hire, Sales Territory Design, Sales Compensation Plans, Email Deliverability (VibeWeek), and Ideal Customer Profile.

When NOT to Run Outbound

Outbound isn't right for every business. Don't invest if:

Your ACV is < $5K and motion is self-serve. Outbound CAC > LTV. Use marketing channels instead.

You don't have a clear ICP. Outbound to "the world" wastes effort + damages reputation. Get Ideal Customer Profile right first.

You haven't done founder-led outbound. Founder should send first 200-500 cold emails personally before delegating. Otherwise you can't write the playbook.

Your product isn't yet demoing well. Outbound surfaces demos; if demos don't close, outbound multiplies waste.

Your runway is under 9 months. Outbound takes 3-6 months to ramp. Use channels with shorter feedback loops.

Your buyer is highly resistant to cold (e.g., enterprise security buyers). Some buyer profiles strictly inbound. Different motion required.

You don't have email deliverability hygiene. Outbound at scale without proper SPF/DKIM/DMARC + warm-up = spam folder forever.

If any apply: fix the precondition.

When Outbound Works

Best fit:

B2B with ACV $10K-$500K. Sweet spot for outbound CAC.
Buyer is a known role / persona. "VP Sales at SaaS company 50-500 employees" is targetable.
Product solves an obvious + recognizable pain. "Speed up sales emails" is searchable + sellable cold.
Sales cycle 30-180 days. Not too fast (no need for outbound) or too slow (cold contact loses momentum).
You have multi-channel signal sources. ZoomInfo + LinkedIn + intent data > just spray.

The Outbound Funnel

Standard outbound funnel:

1. ACCOUNTS IDENTIFIED (TAM)
   - ICP-fit companies enumerated
   - Source: sales intelligence (ZoomInfo, Apollo) + manual research
   - Typical: 2,000-20,000 target accounts

2. CONTACTS SELECTED PER ACCOUNT
   - 1-5 contacts per account (multi-threading)
   - Personas: economic buyer, champion, end-user
   - Source: sales intelligence

3. SEQUENCED (cadence active)
   - Each contact in a sequence (5-10 touches over 14-21 days)
   - Sequence per persona / per ICP cluster
   - Multi-channel: email + LinkedIn + sometimes phone

4. TOUCHED (sequence touches delivered)
   - Email opened (~30-50% rate; declining post-Apple Mail Privacy)
   - Email replied (~3-8% positive reply rate good)
   - LinkedIn engagement (5-15% acceptance + reply)

5. POSITIVE REPLIED (engaged)
   - Counter-question or interest
   - Even objections count if they engage

6. MEETINGS BOOKED
   - 1-3% of touched contacts → meeting
   - Often via Calendly link or SDR-AE handoff

7. SQO (Sales Qualified Opportunity)
   - 50-70% of meetings → SQO (depends on AE qualification)
   - Often filtered out: bad-fit, low-intent, wrong-stage

8. CLOSED WON
   - 20-30% of SQO → closed won (depends on stage)

Reverse-engineer from revenue target:

Revenue goal: $5M new ARR
Average ACV: $20K
Closed won needed: 250
Win rate: 25%
SQOs needed: 1,000
SQO rate from meetings: 60%
Meetings needed: ~1,667
Meeting rate from touched: 1.5%
Touched contacts: 111,000
Average touches per contact: 6
Total touch-events: 666,000

Equivalent capacity:
- 1 SDR: ~10,000 touches/month → 1 SDR/year for 120K touches
- That's ~6 SDRs needed for the volume
- Or: SDR + AI augmentation; or SDR + outsourced

Output: math that ties revenue → SDR headcount.

Step 1: Define ICP + Account List

Tighter ICP = better outbound.

ICP definition:
- Industry / vertical
- Size (employees, revenue)
- Geography
- Tech stack (e.g., "uses Salesforce")
- Stage / funding signal
- Job titles / personas
- Buying triggers (e.g., recently raised; rapid hiring; new VP)

Account selection:

Source data via [Sales Intelligence Tools (Reference)](../../VibeReference/content/marketing-and-seo/sales-intelligence-prospect-data.md):
- ZoomInfo: broadest catalog
- Apollo.io: SMB-friendly
- Cognism: EU-strong
- Clearbit / HubSpot Breeze: HubSpot-aligned
- LinkedIn Sales Navigator: titles + filter by intent
- 6sense / Bombora: intent data layer
- RB2B: visitor de-anon

Account scoring:
- Tier 1 (top 100-500): perfect-fit, high-intent → priority
- Tier 2 (500-2K): good-fit → standard cadence
- Tier 3 (2K-10K+): broader fit → automated touches

Refresh:
- Quarterly account list refresh
- Track: which segments converted; which didn't
- Drop low-converting segments

Tools to manage:
- CRM (Salesforce / HubSpot) as system of record
- Sales engagement platform for cadences
- Analytics for measurement

Step 2: Build Multi-Channel Cadences

A cadence = sequence of touches across channels over time.

Sample cadence (12-touch over 21 days):

Day 1: Email 1 (intro; specific pain; soft CTA)
Day 2: LinkedIn connect request (with personalized note)
Day 4: Email 2 (different angle; case study)
Day 6: LinkedIn message (if connected) or 2nd connect attempt
Day 9: Email 3 (offer specific value; deeper insight)
Day 11: Phone call (if titles + ACV justify)
Day 14: Email 4 (tactical; ROI math)
Day 16: LinkedIn engagement (like / comment on their post)
Day 19: Email 5 ("Bumping this; should I close the loop?")
Day 21: Email 6 (final break-up; "Closing your file")

Channel mix:
- Email: 40-60% of touches
- LinkedIn: 30-40%
- Phone: 0-20% (varies; phone-heavy for high-ACV)
- Direct mail / video: rare, high-leverage in specific cases

Personalization tiers:
- Tier 1 (top 100): 1:1 personalized; founder/AE writes
- Tier 2 (top 1K): templated + 1-2 sentence personalization
- Tier 3 (broader): AI-personalized at scale (but careful — risk of formulaic)

Cadence design principles:
- Open with curiosity, not pitch
- Each touch has different angle (don't repeat same message)
- Provide value (insight, data, resource) before asking
- Be specific (use their name, company, industry, recent news)
- Honest break-up (no false urgency at the end)

A/B testing:
- Subject line A/B
- Send time A/B
- Sequence length (5 vs 10 touches)
- Channel mix
- Persona differentiation

Sequence builder:
- Sales engagement platform (Outreach, Salesloft, Apollo)
- Templates per persona
- Iteration rapid; don't wait

Step 3: Email Deliverability Hygiene

This is the #1 silent killer of outbound. Without good deliverability, your emails go to spam.

Foundational setup:

1. Domain authentication
- SPF record published
- DKIM signing enabled (per sender domain)
- DMARC policy: quarantine or reject
- Verify all configured correctly via mxtoolbox / DMARC report

2. Sender domain segregation
- Separate cold-outbound from transactional
- Use cold-outbound subdomain (e.g., outreach.yourco.com)
- Don't risk transactional reputation with cold

3. Warm-up
- New domain: 4-8 weeks of warm-up before scaling
- Use [Mailwarm / Lemwarm / Smartlead Warmup] for automated warm-up
- Gradually increase volume

4. Volume limits
- Per-mailbox: 30-50 cold emails/day max (Google Workspace abuse threshold)
- Per-day across team: scaled to mailbox count
- Don't blast 1000+ from one mailbox

5. Multiple sending mailboxes
- Spread volume across 5-20 mailboxes
- Each mailbox 30-50/day
- Total team: 1500-10,000/day capacity

6. Quality content
- Avoid spam-trigger words ("Free", "Discount", "Act now")
- Don't include images (drives spam scoring)
- Plain text > heavy HTML for cold
- Custom signatures with no images / minimal links

7. List hygiene
- Verify emails before sending (Hunter / NeverBounce)
- Bounce-rate < 5% (10%+ flags spam)
- Remove hard-bounces immediately

8. Monitoring
- Sender reputation tools (Postmaster Tools, Sender Score)
- Watch open rates (declining = deliverability issue)
- Watch reply rates
- Spam-complaint rate <0.1%

Sales-engagement platforms handle most of this; you still need to do the foundational setup.

Step 4: Personalization at Scale

Outbound at scale tension: personalize OR scale, hard to both.

The 2026 stack:

A. Tier 1 (top 100): manual 1:1
- Founder / AE writes personally
- 30-60 minutes per email
- Highest reply rate (15-30%)
- Highest CAC; only for biggest accounts

B. Tier 2 (top 1K): templated + light personalization
- SDR adds 1-2 sentences specific to prospect
- Reference recent news, role, company specifics
- 5-10 minutes per email
- Reply rate 8-15%

C. Tier 3 (broader): AI-personalized
- Tools: Clay, Smartlead, Apollo AI personalization
- AI scrapes LinkedIn / company website / news
- Generates first-line personalization
- Cost: $0.10-1.00 per personalized email
- Reply rate 3-8% (declining over time as everyone uses AI)

D. Tier 4 (broadest): fully templated
- No personalization beyond [first_name]
- Cheap; high volume
- Reply rate 1-3%
- Risk of being filtered as spam

Best practice:
- Don't use AI for top-50 accounts (sound generic)
- Do use AI for tier 3 / 4 to scale
- Always have a human review step for tier 2

AI personalization tools:
- Clay (orchestration + scrape + AI; popular 2024-2026)
- Smartlead AI
- Apollo AI Sequences
- Outreach AI

Step 5: SDR Org Structure + Compensation

SDR (Sales Development Rep) is the standard outbound role.

When to hire:
- Founder personally validates outbound works (200-500 self-sent emails)
- $1-3M ARR with proven outbound conversion
- AE bandwidth limited
- Predictable demand for cold-sourcing

Profile:
- 0-3 years experience (entry-level role)
- Resilient (rejection-heavy)
- Coachable
- Some sales process discipline
- Domain familiarity (or fast learner)

Comp structure:
- Base: $50-75K (US; varies by region)
- Variable: $20-40K (tied to meetings booked + SQO conversion)
- Total OTE: $70-115K
- 70/30 split typical (70% base, 30% variable)
- Variable tied to:
  - Meetings booked (60% weight)
  - SQO conversion (30% weight)
  - Quality scoring by AE (10% weight)

Quota:
- 60-100 meetings booked per month per SDR (varies by ACV; lower for high-ACV)
- Plus quality threshold (60%+ converted to SQO)

Career path:
- SDR → Senior SDR → AE (typical 12-18 months)
- Some stay SDR; promote to SDR Manager

Org size:
- 1 SDR per 1-2 AEs is typical
- For $5M ARR: 2-3 SDRs + 2-4 AEs
- For $50M ARR: 20-40 SDRs + 30-60 AEs

Outsourced alternative:
- BDR-as-a-service (e.g. Belkins, CIENCE, Belkins, SalesRoads)
- Pros: ramp fast; lower fixed cost
- Cons: quality variable; hard to control culture; less coachable
- Use for early validation; transition to in-house once proven

AI SDR alternative (2026 emerging):
- 11x.ai, Artisan AI, regie.ai, others
- AI does prospecting + outreach + booking
- Cost: $5-15K/mo per AI SDR equivalent
- Quality varies; supplement not replace humans
- Best for: top-of-funnel volume; not quality-replacement

Step 6: Tech Stack

Sample outbound stack (typical 2026):

Sales engagement platform:
- Outreach.io (enterprise standard)
- Salesloft (Outreach competitor)
- Apollo.io (all-in-one; cheaper)
- Lemlist (modern; budget)
- HubSpot Sales (HubSpot-bundled)
- Smartlead (modern; affordable)

Sales intelligence:
- ZoomInfo (enterprise standard; expensive)
- Apollo.io (data + outreach combined)
- Cognism (EU-strong)
- Clearbit / HubSpot Breeze
- LinkedIn Sales Navigator

Email infrastructure:
- Google Workspace (most common; multiple mailboxes)
- Microsoft 365 (alternative; some platforms not as well integrated)
- Mailgun / dedicated SMTP for transactional only

Warm-up:
- Mailwarm
- Lemwarm
- Smartlead Warmup
- Apollo Warm-Up

CRM:
- Salesforce (enterprise B2B standard)
- HubSpot CRM (SMB / mid-market)

Calendar booking:
- Calendly
- Cal.com
- Chili Piper (round-robin to AE; advanced)
- HubSpot Meetings

Personalization:
- Clay (orchestration + scrape; popular)
- Smartlead AI
- Apollo AI

Tracking + analytics:
- CRM dashboards
- Outreach / Salesloft analytics
- Custom Looker / Mode dashboards

Data enrichment:
- Clearbit Reveal / RB2B (visitor de-anon)
- 6sense / Bombora (intent data)
- Hunter.io (email finder for ad-hoc)
- NeverBounce (verification)

Recording / coaching:
- Gong / Chorus (call recording + analysis)
- Coaching layer for AE + SDR coaching

Default stack for $1-10M ARR:
- Apollo.io (data + sequences + dialer)
- HubSpot CRM
- Calendly
- Clay (for advanced personalization)
- Smartlead Warmup

Step 7: Measurement + Iteration

Per-SDR metrics:
- Touches/day (volume)
- Meetings booked
- Meetings held (some no-show)
- Meeting → SQO conversion
- SQO → opportunity conversion
- Activity per stage
- Reply quality (positive / negative)

Per-cadence metrics:
- Open rate
- Reply rate
- Positive reply rate
- Meeting set rate
- Per-step performance

Per-segment metrics:
- Conversion by ICP segment
- Conversion by industry
- Conversion by company size
- Conversion by persona

Per-message metrics:
- Open rate by subject line
- Reply rate by message
- Spam complaint rate

Quarterly retrospective:
- Which sequences worked
- Which didn't (kill them)
- New experiments to run
- ICP refinement signals

A/B testing:
- One variable at a time
- Statistically significant sample (1K+ touches per variant)
- Iterate quickly

Dashboards:
- Daily: SDR activity + meetings booked
- Weekly: cadence performance
- Monthly: conversion funnel
- Quarterly: ICP-fit + segment learnings

Common Failure Modes

Failure 1: Spray-and-pray
- Pattern: 50K cold emails to anyone with relevant title
- Reality: <1% conversion; reputation damage; spam folder
- Fix: tight ICP; quality > volume

Failure 2: No deliverability hygiene
- Pattern: emails go to spam; SDR thinks "outbound doesn't work"
- Fix: SPF/DKIM/DMARC; warm-up; volume limits; monitor sender reputation

Failure 3: Generic templates
- Pattern: "Hey [first_name], hope you're well..." (everyone sends this)
- Reality: 0.5% reply rate; brand erosion
- Fix: personalization beyond [first_name]; specific to their context

Failure 4: SDR-AE handoff broken
- Pattern: SDR books meeting; AE no-shows or comes unprepared
- Reality: customer trust hit; SDR demoralized
- Fix: tight handoff process; AE acknowledges meeting in <2hr; pre-call brief

Failure 5: AE qualifies meetings out aggressively
- Pattern: SDR books; AE rejects 70% as "not qualified"
- Reality: SDR comp dies; SDR quits
- Fix: clear MQL definition; SDR-AE 1:1 on disagreements; calibration

Failure 6: Compensation tied to volume only
- Pattern: SDR books junk meetings to hit quota
- Reality: AE wastes time; conversion poor
- Fix: comp tied to QUALITY (SQO conversion + AE quality score)

Failure 7: No founder review of cadences
- Pattern: SDRs writing on their own; founder doesn't review
- Reality: bad messaging; brand voice erosion
- Fix: founder reviews cadence library quarterly minimum

Failure 8: Buying ZoomInfo without strategy
- Pattern: $30K/yr on ZoomInfo; SDR makes random lists
- Reality: data without strategy = waste
- Fix: ICP-driven account selection; review which segments convert

Failure 9: Outbound + Marketing not aligned
- Pattern: marketing nurturing same accounts; SDR cold-emails too
- Reality: customer feels stalked
- Fix: ABM coordination; suppression lists; cadence respect for marketing engagement

Failure 10: Three-channel cadences without depth
- Pattern: 12-touch cadence with same generic message rephrased
- Fix: each touch a unique angle / value-add

Failure 11: Outsourced firm produces low-quality
- Pattern: BDR-as-a-service; high volume; low conversion
- Fix: rigorous QA; founder reviews messages; convert to in-house

Failure 12: Phone call low-touch
- Pattern: SDR doesn't call; relies on email only
- Reality: phone calls in cadence increase conversion 30-50%
- Fix: include calls; train SDRs

Failure 13: Compliance ignored (GDPR / CAN-SPAM)
- Pattern: cold email to EU contacts without legitimate interest
- Reality: legal risk; reputation damage
- Fix: compliance review per region; opt-out clear

Failure 14: AI SDR + no human oversight
- Pattern: 11x.ai sends thousands; nobody reviews
- Reality: hallucinated content; brand damage
- Fix: human review batch; AI for volume + human for quality

Failure 15: No experiment cadence
- Pattern: same sequences for 12 months
- Reality: market shifts; stale messages
- Fix: 1-2 new sequence experiments per quarter

What Done Looks Like (recap)

A working outbound function:

Mistakes to Avoid

Spraying without ICP discipline. Quality > volume always.
Skipping deliverability hygiene. Emails go to spam = nothing else matters.
Generic templates. Personalization beyond [first_name] is non-negotiable.
Volume-only SDR comp. Quality must factor in.
Skipping founder validation. Founder sends 200-500 emails before SDR hire.
AI SDR without human oversight. Volume × hallucination = damage.
Outsourced firm without QA. Low cost; low conversion; brand risk.
Marketing-Sales misalignment. Suppression lists + ABM coordination.
No cadence iteration. Markets shift; stale messages stop working.
Phone-skipping. Calls increase conversion materially.
Compliance gaps (especially GDPR). Legal risk + reputation.
No SDR career path. Promote to AE; don't trap them.
Forgetting handoff. SDR books meeting; AE must execute beautifully.
Tracking only volume. SQO conversion is the truer metric.