--- name: sales-cadence description: "Design multi-channel outbound cadences: touch count, spacing, channel mix, step content, A/B tests, and benchmarks. Use when reply rates are low, touchpoint count or spacing is unclear, sequence performance plateaus, or the user asks for a full outbound cadence. Not for one-off cold email copy: use cold-email. Not for deliverability, warmup, or spam problems: use sales-deliverability. Not for enriching the contact list: use sales-enrich." license: MIT metadata: author: TechTide AI (Alex Cinovoj) provenance: rewritten from patterns in sales-skills/sales (MIT) category: Sales & RevOps --- # Multi-Channel Outbound Cadence A cadence is a structured argument delivered across channels over weeks. Every touch adds a new angle or it gets cut. Email-only sequences are half a cadence. ## Workflow 1. **Gather context.** Campaign goal (cold, inbound follow-up, trigger-based, re-engagement, expansion), target persona, available channels (email, phone, LinkedIn, video), and constraints (duration, max touches per week, compliance, messaging to reuse). If the request already covers most of this, answer first with stated assumptions and ask only the 1-2 critical gaps at the end. If channels are unknown, ask; never assume a dialer or LinkedIn seat exists. 2. **Design the architecture** as a day-by-day table: day, step, channel, action, notes. - Mix channels: never 3+ consecutive touches on the same channel. - Spacing: days 1-7 every 1-2 days (engagement peaks early), days 8-14 every 2-3 days, day 15+ every 3-4 days. Minimum 2 business days between touches except same-day email plus LinkedIn on day 1. - Touch count by seniority: execs 8-12 touches over 3-4 weeks (spaced, restrained); directors and managers 12-16 over 3-4 weeks; individual contributors 10-14 over 2-3 weeks. - Place calls within 24 hours after an email so you can reference it. - LinkedIn early for the connection, later for social proof. Do not mirror the emails. - Final touch is always a breakup email. 3. **Write the content for every step.** No placeholders-only skeletons. - Emails: A/B subject variants, body with `{{tokens}}`, one specific CTA. Under 125 words cold, under 200 warm. - Calls: 10-second opener, one-sentence reason for calling, 2-3 questions, a meeting ask with a specific time, and a sub-30-second voicemail. - LinkedIn: connection note under 300 characters with zero selling; follow-up referencing a post, mutual, or company insight. - Video (steps 5+ if email is silent): 60-90 seconds, their name, one researched insight, soft CTA. - Personalization: level 3 minimum for cold (something specific about their company: news, stack, job posting). Level 4 (their own content) for top targets. Name, company, and title alone is level 1 and does not count. 4. **Plan A/B tests.** One variable at a time, 100+ sends per variant (200 preferred), judge on positive reply rate, wait at least 5 business days, roll winners in, start the next test. Always be testing; cadences decay. 5. **Set benchmarks and review cadence.** Read references/benchmarks.md when reporting targets or diagnosing underperformance. Weeks 1-2 baseline, weeks 3-4 first test results, monthly funnel review, quarterly message refresh. ## Good vs Bad **Bad step 4 email:** "Hi {{first_name}}, just checking in on my last email. Did you get a chance to review? Happy to jump on a call whenever." **Good step 4 email:** "{{first_name}}, one thing I didn't mention: teams your size usually hit {{pain}} right after {{trigger_event}}. solved it by {{one_line_how}}, and cut {{metric}}. Worth 15 minutes Thursday?" New angle, proof, specific ask. "Just checking in" is a deleted touch, not a written one. ## Verification Before delivering, audit the architecture table. Count consecutive same-channel touches: expect a maximum of 2. Check gaps: expect no interval under 2 business days after day 1. Check every email body for the phrase "checking in" or "following up on my last": expect zero hits; each touch must state a new angle. Any failure: rewrite the offending steps before presenting. After 2 weeks live, compare metrics to references/benchmarks.md. Expect open rate 40%+ and total replies 8%+. Open rate below 30% is a deliverability or subject problem: route to sales-deliverability before touching body copy. ## Footguns - **Email-heavy by default.** Generated cadences drift to all-email. Fix: if phone and LinkedIn are available, they appear in week 1, not as afterthoughts. - **Diagnosing reply problems as content problems.** Low opens are deliverability; low replies on good opens are message-market fit. Fix: read the metric pair before prescribing. - **Ignoring timezone on call steps.** Calls outside business hours are wasted touches. Fix: note best call windows (8-10am, 4-6pm prospect-local). - **Testing two variables at once.** Unreadable results. Fix: one variable, minimum sample, positive-reply metric. - **Compliance as an afterthought.** Cold outreach to regions with consent regimes needs a lawful basis and working opt-out. Fix: confirm target regions in step 1. ## Completion checklist - [ ] Context confirmed, channels verified with the user - [ ] Architecture table passes the mix/spacing audit - [ ] Full content written for every step, personalization level 3+ - [ ] A/B plan with one variable per test and sample thresholds - [ ] Benchmarks and review schedule included Any box unchecked: not done. Fix or say so.