<objective>
Comprehensive B2B sales skill for Epiphan Video BDRs covering cold outreach to video/streaming verticals (Higher Ed, Government, Corporate AV, Courts/Legal, Healthcare, K-12, Houses of Worship), lead scoring for Pearl-2/Mini/Nano/Nexus/EC20/Connect, revenue operations (pipeline metrics, forecasting, LTV:CAC), and sales execution (MEDDIC qualification, SPIN discovery, demo delivery for video capture/streaming workflows, objection handling for video capture/streaming/NDI/SRT).
</objective>
<quick_start>
**Tool order for any outreach/list mode:**
1. `execute_sql` / `hubspot_search_contacts` (Epiphan AI) — candidate pull.
2. `qualify_lead` (Epiphan AI) per contact — Gate Q, mandatory before tiering, scoring, enrichment or any sequence enrollment. Its verdict is the dedupe / Golden Rules / ATL-BTL source of truth.
3. `search_product_catalog` + Epiphan Knowledge `search` — Gate S before any Pearl/EC20/Connect/competitor claim in copy or objection responses.
4. `get_writing_style` once, `check_my_copy` per string — Gate B.
5. Gmail `create_draft` from `tkipper@epiphan.com` — Gate D. Apollo/Nooks sequence enrollment only after Tim approves the drafts.
Lead scoring: Hot: 70+ | Warm: 40-69 | Nurture: <40
Pipeline coverage: 3-4x quota (SMB), 4-5x (Enterprise)
LTV:CAC ratio: Target >3:1
MEDDIC: Metrics, Economic Buyer, Decision Criteria, Decision Process, Identify Pain, Champion
Cold email: Under 100 words, problem-forward (video capture, streaming, hybrid/remote), clear single CTA
Tim's 2026 Quota (full ramp since June): 24 deals/mo, $714K pipeline/mo; H2 revenue targets rise monthly ($475K Jul → $600K Dec) — read the current month's row from the comp plan table in CLAUDE.md. Stretch = 126%+ (1.5x accelerator). 50+ daily dials, 8-12% connect rate, <60 min speed to lead
</quick_start>
<house_rules>
Customer-facing skill. Before output, run the gates in order (details: ~/.claude/skills/epiphan-house-rules-skill/reference/gates.md):
S: search_product_catalog (Epiphan AI) or Epiphan Knowledge search before any product claim. B: check_my_copy on every customer-facing string.
Q: qualify_lead right after the candidate pull, before tiering/enrichment/drafts. D: outbound text becomes a Gmail draft, never a send.
Identifiers (owner IDs, portal, emails, seats) come from reference/identifiers.md in that skill — do not restate them here.
On any gate failure follow reference/self-healing.md (retry, degrade with a flag, alert Tim, halt) and append a line to ~/.claude/skill-runs/<skill-dir>.jsonl.
</house_rules>
<output_contracts>
| Mode | Trigger examples | Deliverable |
|------|------------------|-------------|
| Lead scoring / tiering | "score these leads", "tier this list" | Table: contact, `qualify_lead` verdict, tier (Gold/Silver/Bronze), ATL/GRAY/BTL, score, next action. No contact appears without a Gate Q verdict. |
| Outreach | "outreach plan", "write the first touch" | Gmail `create_draft` per contact (Gate B-checked, Gate S-verified); draft ids returned. Apollo/Nooks enrollment is a separate, confirm-first step after Tim reviews the drafts. |
| Pipeline review / forecast | "pipeline coverage", "forecast", "weekly review" | Markdown review (template below) from `hubspot_search_deals` / `execute_sql`; numbers never estimated when the CRM is reachable. |
| Discovery / demo prep | "MEDDIC checklist", "call prep" | Inline checklist + MEDDIC gaps; HubSpot note via `hubspot_create_contact_note` on request. Full demo arc → `demo-execution-playbook-skill`. |
| Objection response | "they said too expensive" | LAER response; any product/competitor fact inside it is Gate S-verified or marked `[spec unverified]`. |
</output_contracts>
<success_criteria>
Sales process is successful when:
- Leads scored and tiered (Gold/Silver/Bronze) before outreach, and every one carries a `qualify_lead` verdict (Gate Q) — no inline re-implementation of Golden Rules or ATL/BTL
- Product/competitor claims in copy and objection responses verified via Gate S; copy passed `check_my_copy`; outbound staged as Gmail drafts
- Pipeline coverage at 3-4x quota minimum
- LTV:CAC ratio >3:1
- MEDDIC criteria documented for each qualified opportunity
- Demo follows RECAP → AGENDA → SHOW VALUE → SUMMARIZE → NEXT STEPS
- Objections handled with LAER (Listen, Acknowledge, Explore, Respond)
- Outcome sidecar + run-log line written
</success_criteria>
<core_content>
Comprehensive B2B sales skill: outreach, revenue operations, and demo execution.
Quick Reference
| Domain |
Key Components |
Reference File |
| Outreach |
Cold email, sequences, domain warming, lead scoring |
reference/outreach.md |
| Revenue Ops |
Pipeline metrics, forecasting, dashboards, attribution |
reference/revenue-ops.md |
| Discovery |
MEDDIC, SPIN, demo flow, objection handling |
reference/discovery.md |
Part 1: Sales Outreach
The GTM Pipeline (Epiphan Video)
Candidate pull → qualify_lead (Gate Q) → Lead Scoring/Tier → Enrichment → Gmail drafts (Gate B+S+D) → Sequence on approval → CRM Routing
↓ ↓ ↓ ↓ ↓ ↓ ↓
Epiphan AI SQL Epiphan AI this skill Apollo/Clay Gmail create_draft Apollo/Nooks Epiphan AI HubSpot
Gate Q is not optional: nothing is scored, enriched, drafted or enrolled until qualify_lead has returned a verdict for it. If qualify_lead is unavailable, halt the list and alert Tim (see <self_healing>); do not fall back to a memory-based Golden Rules query.
Lead Tiering (Epiphan Video)
| Tier |
Criteria |
Priority |
Action |
Exclude |
| GOLD |
Demo request, pricing inquiry, Pearl family inquiry, general contact us |
Immediate |
Personalized drafts → Apollo sequence on approval |
qualify_lead verdict = exclude (customer, channel, AE-owned <90d, suppressed) |
| SILVER |
Content download (whitepapers), free guide download, Facebook Lead Ads from video/streaming content |
Week 1 |
Standard drafts → Apollo sequence on approval |
qualify_lead verdict = exclude (product-only engager: setup/signup conversions, product usage) |
| BRONZE |
Webinar/lecture symposium attendance, newsletter signup |
Nurture |
Drip campaign + nurture flow |
qualify_lead verdict = exclude |
STALE AE LEAD verdicts (AE-owned, >90 days inactive) are surfaced to Tim with tier and deal value, never auto-sequenced.
ATL/BTL + Golden Rules Gate (Pre-Sequence)
Tier (ATL / GRAY / BTL / NEVER) and exclusion (customer, channel partner, product-only engager, AE-owned with the 90-day stale exception, suppressed) come from the qualify_lead verdict — Gate Q in ~/.claude/skills/epiphan-house-rules-skill/reference/gates.md. AE owner IDs live in that skill's reference/identifiers.md; the canonical title keyword tables live in CLAUDE.md § ATL / BTL Decision-Maker Classification. Do not restate either here. GRAY (Manager with delegated budget >$25K) still requires Tim's review before sequencing.
Lead Scoring (0-100) — Epiphan Video
scoring_factors = {
'icp_fit': 0-30, # vertical ICP scores — declared ONCE in CLAUDE.md § ATL/BTL (Higher Ed 90 … K-12 65); read from there
'intent_signals': 0-25, # Pearl inquiry, CMS integration (Kaltura, Panopto, YuJa), streaming protocol (RTMP/S, SRT, HLS, NDI)
'engagement': 0-20, # Email opens, clicks, replies, website visits to Pearl product pages
'timing': 0-15, # Budget cycle (edu = spring/summer, govt = fiscal year), event season (conferences)
'budget_signals': 0-10 # Company size, multi-campus/location, existing AV investment
}
# Thresholds: Hot: 70+ | Warm: 40-69 | Nurture: <40
6-Agent Architecture (Epiphan MCP Tools)
| Agent |
Role |
MCP Tool |
Output |
| RESEARCHER |
Company intel, tech stack (Extron SMP EOL?, Blackmagic, Crestron, vMix, Teradek presence) |
identify_company + sales_brief (Epiphan AI) first; Clay MCP for gaps |
Enriched company data + competitor intel |
| QUALIFIER |
Dedupe, Golden Rules, ATL/BTL tier, ICP fit |
qualify_lead (Epiphan AI) — Gate Q |
Verdict + tier + 0-100 score |
| ENRICHER |
Contact discovery, ATL-first, for contacts qualify_lead cleared |
hubspot_find_contacts_by_role (Epiphan AI), then Apollo MCP |
Verified emails, org chart, buying signals |
| WRITER |
Personalized copy (Pearl pain points, CMS integrations) — Gate S facts, Gate B voice |
search_product_catalog, check_my_copy, Gmail create_draft |
Drafts for Tim's review; Apollo/Nooks sequence after approval |
| ANALYZER |
Reply intent (demo request vs nurture vs disqualify) |
Gmail MCP + Epiphan CRM MCP |
Route to next action |
| ROUTER |
Orchestration (add to sequence, update HubSpot, schedule follow-up) |
Calendar MCP + Epiphan CRM MCP |
Next-best-action |
Cold Email Principles (Epiphan Video)
- Short and specific - Under 100 words
- Problem-forward - Lead with their pain (hybrid lecture capture, board meeting recording, multi-location streaming, NDI/RTMP integration, CMS workflow)
- Clear CTA - One ask (usually 10-min Pearl product demo or technical fit call)
- Personalization - Company name + vertical signal (e.g., "I noticed you're in Higher Ed using Panopto...")
- Gates - Product facts from
search_product_catalog / Epiphan Knowledge search (Gate S); every body + subject through check_my_copy (Gate B); staged as Gmail create_draft (Gate D) — never sent, never enrolled before Tim reviews
Apollo Email Sequence Structure (Epiphan Video)
| Step |
Timing |
Purpose |
Example |
| 1 |
Day 0 |
Initial outreach - video capture pain point |
"Hybrid lectures + lecture capture best practice" |
| 2 |
Day 3 |
Follow-up - different angle (CMS integration or protocol) |
"How are you integrating Kaltura/Panopto?" |
| 3 |
Day 7 |
Value add - case study or Pearl comparison |
"Why Higher Ed departments prefer Pearl Mini over [competitor]" |
| 4 |
Day 10 |
Break-up - last chance + link to free trial/webinar |
"One last thing: 15-min free Pearl fit assessment" |
Part 2: Revenue Operations
Core Metrics
Pipeline Metrics
pipeline_coverage:
formula: "Pipeline Value / Quota"
healthy: "3-4x for SMB, 4-5x for Enterprise"
warning: "Below 3x"
pipeline_velocity:
formula: "(# Opps x Win Rate x Avg Deal) / Cycle Days"
use: "Predict monthly revenue"
weighted_pipeline:
formula: "Sum of (Deal Value x Stage Probability)"
Conversion Funnel
| Stage |
Formula |
Benchmark |
| Lead to MQL |
MQLs / Total Leads |
15-30% |
| MQL to SQL |
SQLs / MQLs |
30-50% |
| SQL to Opp |
Opportunities / SQLs |
50-70% |
| Opp to Win |
Closed Won / Opportunities |
20-30% |
| Overall |
Closed Won / Total Leads |
1-5% |
Unit Economics
| Metric |
Formula |
Healthy |
| CAC |
(Sales + Marketing) / New Customers |
Depends on ACV |
| LTV |
(ARPU x Gross Margin) / Churn Rate |
- |
| LTV:CAC |
LTV / CAC |
>3:1 |
| Payback |
CAC / (ARPU x Gross Margin) |
<12 months |
Pipeline Stages
| Stage |
Probability |
Entry Criteria |
| Lead |
5% |
Contact captured |
| MQL |
10% |
Meets ICP |
| SQL |
20% |
BANT confirmed |
| Discovery |
30% |
Meeting scheduled |
| Demo |
50% |
Demo completed |
| Proposal |
70% |
Proposal sent |
| Negotiation |
85% |
Terms discussed |
| Closed Won |
100% |
Contract signed |
Forecasting Methods
| Method |
Formula |
Best For |
| Pipeline-based |
Sum(Deal x Stage Probability) |
Simple, data-driven |
| Historical |
Historical conversion x Pipeline |
Past performance |
| Commit-based |
Rep commits + Manager adjustment |
Incorporates judgment |
Part 3: Demo & Discovery
Call Structure
| Stage |
Goal |
Duration |
| Opening |
Build rapport, set agenda |
2-3 min |
| Discovery |
Uncover pain, qualify |
15-20 min |
| Demo |
Show relevant value |
15-20 min |
| Close |
Agree next steps |
5 min |
SPIN Questioning
| Type |
Purpose |
Example |
| Situation |
Understand context |
"Walk me through your current process..." |
| Problem |
Surface pain |
"What challenges do you face with...?" |
| Implication |
Deepen pain |
"What happens when that goes wrong?" |
| Need-Payoff |
Envision solution |
"If you could fix that, what would change?" |
MEDDIC Qualification
| Letter |
Element |
Key Question |
| M |
Metrics |
What's the measurable impact? |
| E |
Economic Buyer |
Who controls budget? |
| D |
Decision Criteria |
How will they decide? |
| D |
Decision Process |
What are steps to buy? |
| I |
Identify Pain |
What's compelling reason to act? |
| C |
Champion |
Who's selling internally? |
Demo Best Practices
1. RECAP (2 min)
"Based on our discovery, you mentioned [pain 1], [pain 2]..."
2. AGENDA (1 min)
"I'll show how we address each. Stop me anytime."
3. SHOW VALUE (15-20 min)
Pain -> Feature -> Benefit -> Proof (repeat for each pain)
4. SUMMARIZE (2 min)
"So you'd be able to [benefit 1], [benefit 2]..."
5. NEXT STEPS (5 min)
"What questions? What's our next step?"
Demo Rules
- Show, don't tell - Open the product, demonstrate
- Connect to pain - Every feature tied to their problem
- Pause for reactions - "How does that compare to current?"
Objection Handling (LAER)
L - Listen (fully, don't interrupt)
A - Acknowledge (validate the concern)
E - Explore (understand the root)
R - Respond (address specifically)
<methodology_integration>
## Methodology Integration
Sales-revenue integrates with the strategy skill cluster for first-principles selling:
Discovery Calls:
- Use JTBD Forces of Progress to identify what's pushing the prospect to switch
- Use Challenger's Teach-Tailor-Take Control to structure the conversation around a commercial insight
- Use NSTTD calibrated questions ("What does success look like?") instead of closed questions
Cold Outreach:
- Lead with a Challenger reframe (not product features)
- Open with an NSTTD accusation audit ("You probably get dozens of vendor emails...")
- Frame the CTA as a no-oriented question ("Would it be out of the question to...?")
- Under 100 words, technically precise
Objection Handling:
- Use NSTTD labeling to name the objection: "It seems like timing is a concern..."
- Apply Challenger's assertive redirect to steer back to value
- Use calibrated questions to make them solve YOUR problem: "How am I supposed to do that?"
Price Negotiations:
- Use the Ackerman model (65% → 85% → 95% → 100% with precise final number)
- Anchor with Challenger rational drowning (quantify the problem before discussing price)
See: challenger-sale-skill, never-split-the-difference-skill, jobs-to-be-done-skill
</methodology_integration>
Product and competitor facts in these responses (NDI/SRT, Kaltura integration, TCO vs Extron/Blackmagic, trial terms) are Gate S-verified via search_product_catalog + Epiphan Knowledge search before they are said or written; otherwise mark [spec unverified].
| Objection |
Response Framework (Epiphan) |
| "Too expensive" |
Acknowledge -> "Compared to Extron SMP or Blackmagic?" -> Show Total Cost of Ownership (Pearl-2 vs legacy + integration) |
| "We use [Blackmagic/Crestron/vMix]" |
Acknowledge -> "What pain points?" -> Differentiate (Pearl = portable, NDI + SRT, Kaltura native integration) |
| "Not ready / testing phase" |
Acknowledge -> "When's your pilot window?" -> Offer 30-day trial of Pearl Mini + Epiphan Connect |
| "Need IT approval" |
Acknowledge -> "What will IT ask?" -> Prepare tech brief on NDI/RTMP/SRT support, security |
| "Looking at free option" |
Acknowledge -> "OBS works, but what about reliability + support?" -> Show Pearl value (pro support, firmware updates, CMS integration) |
Call Prep Checklist (Epiphan Video)
### Research (10 min)
- [ ] Company website - video/streaming mentions, CMS (Kaltura/Panopto/YuJa/Echo360)
- [ ] LinkedIn - prospect AV/IT/Broadcast background
- [ ] Tech stack - BuiltWith (CDN, video), job postings (streaming engineer?)
- [ ] Competitor presence - Extron, Blackmagic, Crestron, vMix, Teradek
### Preparation (5 min)
- [ ] Hypothesis: Hybrid lectures? Multi-location streaming? CMS integration pain?
- [ ] 3 discovery questions ready (Pearl capability fit, current workflow, CMS integration)
- [ ] Pearl demo environment ready (focus on their vertical: K-12 classroom? University lecture hall? Government broadcast?)
- [ ] Clear next step in mind (trial, technical assessment, peer reference call)
### Mindset
- [ ] Curiosity about their video workflow, not pitch mode
- [ ] Understand their AV/IT ecosystem first (NDI? RTMP? SRT comfort level?)
Weekly Pipeline Review Template
### Coverage Check
- Current pipeline: $___
- Quota this month: $___
- Coverage ratio: ___x (target: 3-4x)
### Stage Movement
| Stage | Start | End | Net |
|-------|-------|-----|-----|
| Discovery | | | |
| Demo | | | |
| Proposal | | | |
### Deals at Risk
| Deal | Amount | Days in Stage | Risk |
|------|--------|---------------|------|
### Action Items
- [ ] Stalled deals to address
- [ ] Proposals to follow up
- [ ] Deals to close this week
Integration Notes (Epiphan Video)
- Email Sequences: Apollo.io (MCP connected) — GOLD/SILVER tier, enrolled only after Gmail drafts are approved
- CRM: HubSpot (Epiphan CRM MCP connected) — sync deals, track device_count and engagement_overview
- Enrichment: Apollo MCP (contact discovery), Clay MCP (company intel + competitor tech stack)
- Discovery Tools: Gmail MCP (reply detection), Google Calendar MCP (schedule follow-ups)
- Channel Partners: excluded by the
qualify_lead verdict (is_channel = true); known names: AVI-SPL, Diversified, CTI, CCS Presentation Systems, Ford AV
- Qualification / ATL-BTL / identifiers: Gate Q +
reference/identifiers.md in epiphan-house-rules-skill; keyword tables and ICP scores in CLAUDE.md
- Related Skills: demo-execution-playbook-skill, meddic-call-prep-auto-skill, cold-email-sequence-generator-skill, sales-methodology-implementer-skill, challenger-sale-skill, never-split-the-difference-skill
- Products: current lineup per
search_product_catalog (Pearl family, EC20 PTZ, Epiphan Connect/Edge/Unify); never quote specs from this list
- CMS Integrations: Kaltura, Panopto, YuJa, Echo360, Opencast
- Streaming Protocols: RTMP/S, SRT, HLS, NDI, RTSP
Golden Rules — EXCLUDE These Leads
The exclusion set (customer, product-only engager, device_count ≥ 1, channel partner, AE-owned with the 90-day stale exception, suppressed) is evaluated by qualify_lead — Gate Q — immediately after the candidate pull and before any Apollo/Nooks enrollment. Owner IDs and the full rule text live in epiphan-house-rules-skill/reference/identifiers.md and CLAUDE.md § Golden Rules; this skill does not restate them.
Reference Files
reference/outreach.md - Email templates, Apollo sequences, Pearl messaging by verticalreference/revenue-ops.md - Metrics, dashboards, forecasting, pipeline healthreference/discovery.md - MEDDIC scorecard, Pearl demo scripts, objection library (video capture specific)reference/sequence-audit-removal-pattern.md - removing contacts from sequences safely
<self_healing>
Failure = no scores/drafts/review produced; partial = a stage degraded (name it). Ladder per external call (Epiphan AI, Apollo, Clay, Gmail, Calendar): retry once with backoff → degrade (Apollo down → Clay-only enrichment; Clay down → HubSpot data only; Brand MCP down → `[brand check skipped]`; catalog empty → `[spec unverified]`) → alert Tim via Slack DM → halt. **`qualify_lead` unavailable = halt** for any list/outreach mode; never substitute an inline Golden Rules query. Append one line per run to `~/.claude/skill-runs/sales-revenue-skill.jsonl`. Full contract: `~/.claude/skills/epiphan-house-rules-skill/reference/self-healing.md`.
</self_healing>
Emit Outcome Sidecar
As the final step, write to ~/.claude/skill-analytics/last-outcome-sales-revenue-skill.json:
{"ts":"[UTC ISO8601]","skill":"sales-revenue-skill","version":"<frontmatter version>","variant":"default",
"status":"[success|partial|error]","runtime_ms":[estimated ms from start],
"metrics":{"emails_generated":[n],"calls_scripted":[n],"leads_qualified":[n],"pipeline_value":[n]},
"error":null,"session_id":"[YYYY-MM-DD]"}
Use status "partial" if some stages failed but results were produced (name the stage in error). Use "error" only if no output was generated.
</core_content>