Back to Details

prospect-research-to-cadence

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese
<objective> Automate the full prospect research → outreach pipeline. Takes a company name, domain, or contact and produces enriched firmographics, personalized multi-touch email sequences, call scripts with MEDDIC hooks, and optionally loads contacts into Apollo sequences — all in one workflow. Draft + approve mode: research and drafts are generated, then paused for Tim's approval before loading sequences. </objective>
<quick_start> Single prospect: "research prospect Baylor University" → enriches org + decision-makers → drafts personalized outreach → presents for approval → loads to Apollo sequence
Batch mode: "build cadence for these 5 companies: [list]" → runs pipeline for each → presents batch draft → loads approved contacts
Trigger phrases:
  • "research prospect [company/domain]"
  • "build cadence for [company]"
  • "outreach for [company]"
  • "research-to-cadence [company]"
  • "enrich and sequence [list]"
  • "new prospect batch" </quick_start>
<success_criteria>
  • Company enriched with firmographics, tech stack, news, and Epiphan use case angle
  • 3-5 ATL decision-maker contacts identified per approved title patterns (VP/Director/Dean/Chief/Provost = economic buyers; Managers flagged as GRAY for manual review)
  • Personalized 3-touch email sequence drafted per contact
  • Call script with MEDDIC discovery hooks generated
  • Draft presented for Tim's approval BEFORE any sequence loading
  • On approval: contacts created in Apollo + added to specified sequence
  • Golden Rules exclusions enforced (no customers, no channel, no product-usage contacts) </success_criteria>
<workflow>
<objective> 自动化完整的潜在客户研究→外展流程。输入公司名称、域名或联系人信息，即可生成补全后的企业画像、个性化多触点邮件序列、带有MEDDIC钩子的通话脚本，还可选择将联系人导入Apollo序列——全部在一个工作流中完成。草稿+审批模式：生成研究内容和草稿后，暂停等待Tim审批，之后再导入序列。 </objective>
<quick_start> 单个潜在客户： "research prospect Baylor University" → 补全组织信息+决策人信息 → 生成个性化外展草稿 → 提交审批 → 导入Apollo序列
批量模式： "build cadence for these 5 companies: [列表]" → 为每个公司运行流程 → 提交批量草稿 → 导入获批的联系人
触发短语：
  • "research prospect [公司/域名]"
  • "build cadence for [公司]"
  • "outreach for [公司]"
  • "research-to-cadence [公司]"
  • "enrich and sequence [列表]"
  • "new prospect batch" </quick_start>
<success_criteria>
  • 公司信息已补全，包含企业画像、技术栈、新闻资讯以及Epiphan的应用场景切入点
  • 每个获批的职位类型下识别出3-5位ATL决策人联系人（副总裁/总监/院长/首席/教务长=经济买家；经理标记为GRAY需手动审核）
  • 为每位联系人生成个性化3触点邮件序列草稿
  • 生成带有MEDDIC发现钩子的通话脚本
  • 草稿需提交给Tim审批，之后才能导入序列
  • 审批通过后：在Apollo中创建联系人并添加到指定序列
  • 执行黄金规则排除（排除现有客户、渠道合作伙伴、已使用产品的联系人） </success_criteria>
<workflow>

Pipeline Stages

流程阶段

INPUT                    RESEARCH                 DRAFT                    APPROVE              LOAD
─────────────────────────────────────────────────────────────────────────────────────────────────────
Company/Domain    →  Apollo Org Enrich     →  Email Sequences    →  Tim Reviews     →  Apollo Sequence
                  →  Apollo People Search  →  Call Scripts       →  Approves/Edits  →  Contact Create
                  →  Epiphan CRM Check     →  ICP Score          →                  →  Add to Cadence
                  →  Web Research          →  Pain Hypothesis    →                  →
INPUT                    RESEARCH                 DRAFT                    APPROVE              LOAD
─────────────────────────────────────────────────────────────────────────────────────────────────────
Company/Domain    →  Apollo Org Enrich     →  Email Sequences    →  Tim Reviews     →  Apollo Sequence
                  →  Apollo People Search  →  Call Scripts       →  Approves/Edits  →  Contact Create
                  →  Epiphan CRM Check     →  ICP Score          →                  →  Add to Cadence
                  →  Web Research          →  Pain Hypothesis    →                  →

Stage 1: Research (Tools Used)

阶段1：研究（使用工具）

1a. Company Intelligence

1a. 企业情报

Use these MCP tools in parallel:
ToolPurposeKey Data
apollo_organizations_enrich
FirmographicsIndustry, size, revenue, tech stack
apollo_organizations_job_postings
Hiring signalsAV/IT roles = buying signal
hubspot_search_companies
Existing relationship checkDeals, lifecycle stage
crm_search_customers
Epiphan CRM customer matchDevice count, orders, channel flag
analytics_search_by_email
Device registration lookupRegistered devices by contact email
Web searchRecent newsFunding, expansion, leadership changes
并行使用以下MCP工具：
工具用途关键数据
apollo_organizations_enrich
企业画像补全行业、规模、收入、技术栈
apollo_organizations_job_postings
招聘信号AV/IT岗位=采购信号
hubspot_search_companies
现有合作关系核查交易记录、生命周期阶段
crm_search_customers
Epiphan CRM客户匹配设备数量、订单、渠道标记
analytics_search_by_email
设备注册查询按联系人邮箱统计已注册设备
Web搜索近期新闻融资、扩张、领导层变动

1a-bis. Clay Enrichment (Waterfall)

1a-bis. Clay信息补全（瀑布式 fallback）

For companies/contacts where Apollo data is incomplete (missing phone, tech stack, or funding data):
ToolPurposeKey Data
find-and-enrich-contacts-at-company
(Clay)		Deep contact enrichmentPhone waterfall, social profiles, employment history
add-contact-data-points
(Clay)		Append missing data pointsPhone verification, org chart depth
find-and-enrich-company
(Clay)		Company intelligenceTechnographics, hiring signals, funding
Clay acts as a waterfall fallback — it aggregates 50+ data providers. Use it when Apollo returns incomplete results, especially for phone numbers.
对于Apollo数据不完整的公司/联系人（缺少电话、技术栈或融资数据）：
工具用途关键数据
find-and-enrich-contacts-at-company
(Clay)		深度联系人信息补全电话瀑布式获取、社交资料、就业经历
add-contact-data-points
(Clay)		补充缺失数据点电话验证、组织架构层级
find-and-enrich-company
(Clay)		企业情报补全技术画像、招聘信号、融资信息
Clay作为瀑布式备选工具——它聚合了50+数据提供商。当Apollo返回结果不完整时使用，尤其是获取电话号码时。

1b. Golden Rules Filter

1b. 黄金规则过滤

STOP and skip if ANY of these are true:
  • lifecyclestage = 'customer'
    in HubSpot
  • first_conversion
    contains 'Pearl', 'setup', 'Connect', 'signup'
  • Company 
    device_count >= 1
    in Epiphan CRM
  • engagement_overview
    contains product usage
  • is_channel = true
  • hubspot_owner_id
    IN ('82625923', '423155215', '190030668') — AEs: Lex Evans, Ron Epstein, Phillip Sandler → 90-Day Stale Exception: If 
    hs_lastmodifieddate
    > 90 days ago → SURFACE as 
    STALE AE LEAD
    with ATL/BTL tier + deal value. Tim reviews to re-engage, push to demo, or help close. → Ron Epstein: all leads. Lex & Phil: NA only. <90 days → still exclude.
OPERATING PRINCIPLE: Tim is NET-NEW first, but can re-engage STALE AE leads (90+ days inactive) to recover pipeline value.
若满足以下任一条件，立即停止并跳过：
  • HubSpot中
    lifecyclestage = 'customer'
  • first_conversion
    包含'Pearl'、'setup'、'Connect'、'signup'
  • Epiphan CRM中公司
    device_count >= 1
  • engagement_overview
    包含产品使用记录
  • is_channel = true
  • hubspot_owner_id
    属于('82625923', '423155215', '190030668') —— 客户经理：Lex Evans、Ron Epstein、Phillip Sandler → 90天 stale 例外规则： 若
    hs_lastmodifieddate
    > 90天前 → 标记为
    STALE AE LEAD
    ，附带ATL/BTL层级+交易价值。由Tim审核以重新触达、推进演示或协助成交。 → Ron Epstein：所有线索。Lex & Phil：仅北美地区。<90天的线索仍需排除。
操作原则： Tim优先处理全新线索，但可重新触达STALE AE线索（90+天未活跃）以恢复管线价值。

1c. Contact Discovery

1c. 联系人发掘

Use 
apollo_mixed_people_api_search
with filters:
  • Titles (ATL-only): VP, Director, Dean, CIO, CTO, CFO, Provost, Superintendent, Court Administrator, City Manager, Senior Pastor
  • Gray Zone Titles (flag for review): Manager (AV/Facilities/IT) — include ONLY if Apollo confirms reports to Director+ AND has delegated budget >$25K
  • Exclude (NEVER ATL): Warehouse Manager, Network Manager, Systems Administrator, AV Technician, Graphic Design Instructor, Program Administrator, Web Designer, Classroom Support, Lab Coordinator, Maintenance, Building Engineer, Multimedia Services Manager, Video Production Specialist, Streaming Crew
  • Seniority: director, vp, c_suite
  • Departments: IT, engineering, operations, education
  • Limit: 3-5 contacts per company
For each contact, run 
apollo_people_match
to get verified email + phone.
使用
apollo_mixed_people_api_search
并应用以下筛选条件：
  • 职位（仅ATL）： VP、Director、Dean、CIO、CTO、CFO、Provost、Superintendent、Court Administrator、City Manager、Senior Pastor
  • 灰色区域职位（标记需审核）： Manager（AV/设施/IT）—— 仅当Apollo确认其汇报给Director及以上且拥有>25K美元的授权预算时才纳入
  • 排除（绝不属于ATL）： Warehouse Manager、Network Manager、Systems Administrator、AV Technician、Graphic Design Instructor、Program Administrator、Web Designer、Classroom Support、Lab Coordinator、Maintenance、Building Engineer、Multimedia Services Manager、Video Production Specialist、Streaming Crew
  • 职级： director、vp、c_suite
  • 部门： IT、工程、运营、教育
  • 限制： 每家公司3-5位联系人
为每位联系人运行
apollo_people_match
以获取验证后的邮箱+电话。

ATL/BTL Classification (applied to each discovered contact)

ATL/BTL分类（应用于每位发掘的联系人）

Classify each contact using approved title patterns (CLAUDE.md § ATL/BTL Classification v1.0):
  • ATL: Chief, VP, Director, Dean, Provost, Superintendent, Court Administrator, City Manager, Senior Pastor, Executive Pastor
  • BTL: Technician, Specialist, Coordinator, Support, Administrator (Systems/Network/Database), Engineer (AV/Network/Systems), Operator, Instructor/Professor/Faculty, Designer, Assistant, Clerk (non-Court Admin), Volunteer, Intern, Student, Resident, Help Desk
  • NEVER ATL: Warehouse Manager, Network Manager, Systems Administrator, AV Technician, Graphic Design Instructor, Program Administrator, Web Designer, Classroom Support, Lab Coordinator, Maintenance, Building Engineer, Multimedia Services Manager, Video Production Specialist, Streaming Crew
  • GRAY: Manager (AV/Facilities/IT) — ATL only if reports to Director+ AND has delegated budget >$25K. Flag for Tim's manual review before sequence load
Tag each contact: 
atl_btl_tier = 'ATL' | 'BTL' | 'GRAY'
Only ATL contacts are auto-approved for sequence loading. GRAY contacts require Tim's explicit approval. BTL contacts are deprioritized (included in batch but called last).
For contacts where 
apollo_people_match
returns no phone number, fall back to Clay's 
find-and-enrich-contacts-at-company
for phone waterfall.
使用获批的职位模式对每位联系人进行分类（参考CLAUDE.md § ATL/BTL Classification v1.0）：
  • ATL： Chief、VP、Director、Dean、Provost、Superintendent、Court Administrator、City Manager、Senior Pastor、Executive Pastor
  • BTL： Technician、Specialist、Coordinator、Support、Administrator（系统/网络/数据库）、Engineer（AV/网络/系统）、Operator、Instructor/Professor/Faculty、Designer、Assistant、Clerk（非Court Admin）、Volunteer、Intern、Student、Resident、Help Desk
  • 绝不属于ATL： Warehouse Manager、Network Manager、Systems Administrator、AV Technician、Graphic Design Instructor、Program Administrator、Web Designer、Classroom Support、Lab Coordinator、Maintenance、Building Engineer、Multimedia Services Manager、Video Production Specialist、Streaming Crew
  • GRAY： Manager（AV/设施/IT）—— 仅当汇报给Director及以上且拥有>25K美元授权预算时才属于ATL。在导入序列前需标记为Tim手动审核项
为每位联系人添加标签：
atl_btl_tier = 'ATL' | 'BTL' | 'GRAY'
仅ATL联系人可自动获批导入序列。GRAY联系人需Tim明确审批。BTL联系人优先级较低（纳入批次但最后联系）。
对于
apollo_people_match
未返回电话号码的联系人，使用Clay的
find-and-enrich-contacts-at-company
进行瀑布式电话获取。

Phone Verification Priority

电话验证优先级

ALWAYS attempt phone verification. Tim needs 50+ daily dials. Every contact loaded into a sequence MUST have a verified phone number. Priority order: Apollo phone → Clay waterfall → skip if no phone found (flag for manual research).
始终尝试电话验证。 Tim每日需要拨打50+通电话。导入序列的每位联系人必须拥有已验证的电话号码。优先级顺序：Apollo提供的电话 → Clay瀑布式获取 → 若无则跳过（标记为手动研究项）。

1d. ICP Scoring

1d. ICP评分

Score the prospect using Tim's ICP verticals:
VerticalScore
Higher Ed90
Courts/Legal85
Government80
Corporate AV80
Healthcare75
Houses of Worship70
K-1265
Boost +5 for: hiring AV roles, recent facility expansion, CMS mentioned in tech stack Boost +10 for: Extron/Matrox replacement signal (competitors exited market)
使用Tim的ICP垂直领域对潜在客户进行评分：
垂直领域分数
高等教育90
法院/法律85
政府80
企业AV80
医疗健康75
宗教场所70
K-12教育65
额外加分项：
  • 招聘AV岗位、近期设施扩张、技术栈中提及CMS：+5分
  • 存在Extron/Matrox替换信号（竞争对手退出市场）：+10分

Stage 2: Draft Outreach

阶段2：生成外展草稿

Email Sequence (3-touch)

邮件序列（3触点）

Load 
reference/email-templates.md
for templates. Customize with:
  • Touch 1 (Day 0): Pain-forward, reference specific news/signal, single CTA (15-min call)
  • Touch 2 (Day 3): Value prop with peer social proof (similar vertical), ask different question
  • Touch 3 (Day 7): Breakup email with resource offer (case study or guide)
Rules:
  • Under 100 words per email
  • No "I hope this finds you well"
  • Lead with THEIR problem, not Epiphan features
  • Reference a specific trigger (job posting, news, tech stack gap)
  • CTA = specific time ask ("15 minutes this Thursday?")
加载
reference/email-templates.md
中的模板。自定义规则：
  • 触点1（第0天）： 以痛点为切入点，参考特定新闻/信号，单一行动号召（15分钟通话）
  • 触点2（第3天）： 提供价值主张并附带同行社交证明（同垂直领域），提出不同问题
  • 触点3（第7天）： 收尾邮件，附带资源福利（案例研究或指南）
规则：
  • 每封邮件不超过100词
  • 禁止使用“祝您安好”之类的客套话
  • 以客户的问题为开头，而非Epiphan的产品功能
  • 参考特定触发信号（招聘信息、新闻、技术栈缺口）
  • 行动号召需明确时间（“本周四抽15分钟聊聊？”）

Call Script

通话脚本

Generate MEDDIC-structured call script:
OPENER (15 sec): Reference email + trigger signal
PAIN PROBE (60 sec): "What's your current setup for [use case]?"
METRICS: "How are you measuring [quality/uptime/cost]?"
ECONOMIC BUYER: "Who else would need to sign off?"
DECISION CRITERIA: "What would you evaluate?"
CHAMPION TEST: "If this made sense, would you be open to a quick demo?"
生成MEDDIC结构的通话脚本：
开场（15秒）：提及邮件+触发信号
痛点探查（60秒）：“您当前[应用场景]的设置是怎样的？”
指标：“您如何衡量[质量/正常运行时间/成本]？”
经济买家：“还有哪些人需要签字批准？”
决策标准：“您会评估哪些方面？”
拥护者测试：“如果方案可行，您是否愿意安排一场快速演示？”

Deliverable Format

交付格式

Present draft as structured output:
═══════════════════════════════════════════
PROSPECT RESEARCH: [Company Name]
ICP Score: [XX/100] | Vertical: [vertical]
═══════════════════════════════════════════

COMPANY INTEL:
- Industry: [...]
- Size: [...] employees | Revenue: [...]
- Tech stack: [...]
- Trigger signal: [...]
- Epiphan angle: [...]

CONTACTS (ready for sequence):
┌──────────────────────────────────────────┐
│ 1. [Name] — [Title] [ATL/GRAY/BTL]       │
│    Email: [...] | Phone: [...]           │
│    LinkedIn: [...]                        │
│    Personalization hook: [...]            │
└──────────────────────────────────────────┘

EMAIL SEQUENCE:
Touch 1: [subject] → [body]
Touch 2: [subject] → [body]
Touch 3: [subject] → [body]

CALL SCRIPT:
[MEDDIC-structured script]

═══════════════════════════════════════════
APPROVE? (y/edit/skip)
═══════════════════════════════════════════
以结构化输出呈现草稿：
═══════════════════════════════════════════
潜在客户研究：[公司名称]
ICP评分：[XX/100] | 垂直领域：[vertical]
═══════════════════════════════════════════

企业情报：
- 行业：[...]
- 规模：[...]名员工 | 收入：[...]
- 技术栈：[...]
- 触发信号：[...]
- Epiphan切入点：[...]

联系人（可导入序列）：
┌──────────────────────────────────────────┐
│ 1. [姓名] — [职位] [ATL/GRAY/BTL]       │
│    邮箱：[...] | 电话：[...]           │
│    LinkedIn：[...]                        │
│    个性化切入点：[...]            │
└──────────────────────────────────────────┘

邮件序列：
触点1：[主题] → [正文]
触点2：[主题] → [正文]
触点3：[主题] → [正文]

通话脚本：
[MEDDIC结构脚本]

═══════════════════════════════════════════
是否批准？(y/编辑/跳过)
═══════════════════════════════════════════

Stage 3: Approve

阶段3：审批

  • Present draft to Tim using AskUserQuestion tool
  • Options: "Approve & Load", "Edit first", "Skip this prospect"
  • If edit: Tim provides changes, re-draft, re-present
  • 使用AskUserQuestion工具将草稿提交给Tim
  • 选项：“批准并导入”、“先编辑”、“跳过该潜在客户”
  • 若选择编辑：Tim提供修改意见，重新生成草稿并再次提交

Stage 4: Load to Apollo Sequence

阶段4：导入Apollo序列

On approval:
  1. Create contacts in Apollo via 
    apollo_contacts_create
    :
    • first_name, last_name, email, organization_name, title
  2. Find target sequence via 
    apollo_emailer_campaigns_search
    :
    • Search by sequence name Tim specifies (or default active sequence)
  3. Add to sequence via 
    apollo_emailer_campaigns_add_contact_ids
    :
    • Pass contact IDs + sequence ID
    • Confirm enrollment count
  4. Optional: Create Gmail draft via 
    gmail_create_draft
    :
    • For high-ICP prospects (score 80+), also create a direct Gmail draft
</workflow> <dependencies>
审批通过后：
  1. 创建联系人：通过
    apollo_contacts_create
    在Apollo中创建联系人：
    • first_name, last_name, email, organization_name, title
  2. 查找目标序列：通过
    apollo_emailer_campaigns_search
    查找目标序列：
    • 根据Tim指定的序列名称搜索（或使用默认活跃序列）
  3. 添加到序列：通过
    apollo_emailer_campaigns_add_contact_ids
    将联系人添加到序列：
    • 传入联系人ID+序列ID
    • 确认注册数量
  4. 可选：创建Gmail草稿：通过
    gmail_create_draft
    创建Gmail草稿：
    • 对于高ICP得分的潜在客户（得分80+），额外创建直接的Gmail草稿
</workflow> <dependencies>

Required MCP Tools

所需MCP工具

  • Apollo MCP: organizations_enrich, organizations_job_postings, mixed_people_api_search, people_match, contacts_create, emailer_campaigns_search, emailer_campaigns_add_contact_ids, email_accounts_index (resolve 
    send_email_from_email_account_id
    when loading contacts into sequences)
  • Clay MCP: find-and-enrich-contacts-at-company, find-and-enrich-company, add-contact-data-points (waterfall enrichment fallback for phone numbers and deep firmographics)
  • Epiphan CRM MCP: crm_search_customers, analytics_search_by_email, hubspot_search_companies, hubspot_search_contacts
  • Gmail MCP: gmail_create_draft (for high-ICP direct outreach)
  • Web Search: For news, trigger signals
  • Apollo MCP： organizations_enrich, organizations_job_postings, mixed_people_api_search, people_match, contacts_create, emailer_campaigns_search, emailer_campaigns_add_contact_ids, email_accounts_index（将联系人导入序列时解析
    send_email_from_email_account_id
    ）
  • Clay MCP： find-and-enrich-contacts-at-company, find-and-enrich-company, add-contact-data-points（用于电话号码和深度企业画像的瀑布式补全备选工具）
  • Epiphan CRM MCP： crm_search_customers, analytics_search_by_email, hubspot_search_companies, hubspot_search_contacts
  • Gmail MCP： gmail_create_draft（用于高ICP潜在客户的直接外展）
  • Web搜索： 用于获取新闻、触发信号

Sibling Skills Referenced

参考的关联技能

  • sales-revenue-skill
    — Email templates, lead scoring tiers, MEDDIC framework
  • hubspot-revops-skill
    — Golden Rules filter logic, HubSpot query patterns
  • research-skill
    — Competitive intelligence, firmographic research patterns
  • CLAUDE.md § ATL/BTL Classification v1.0
    — Approved title keyword patterns for ATL/BTL/Gray classification (2026-03-17)
</dependencies>
  • sales-revenue-skill
    —— 邮件模板、线索评分层级、MEDDIC框架
  • hubspot-revops-skill
    —— 黄金规则过滤逻辑、HubSpot查询模式
  • research-skill
    —— 竞争情报、企业画像研究模式
  • CLAUDE.md § ATL/BTL Classification v1.0
    —— 获批的ATL/BTL/Gray分类职位关键词模式（2026-03-17）
</dependencies>

Emit Outcome Sidecar

输出结果副文件

As the final step, write to 
~/.claude/skill-analytics/last-outcome-prospect-research-to-cadence.json
:
json
{"ts":"[UTC ISO8601]","skill":"prospect-research-to-cadence","version":"1.0.0","variant":"default",
 "status":"[success|partial|error]","runtime_ms":[estimated ms from start],
 "metrics":{"prospects_researched":[n],"emails_drafted":[n],"sequences_loaded":[n],"atl_count":[n],"icp_avg_score":[n]},
 "error":null,"session_id":"[YYYY-MM-DD]"}
Use status "partial" if some stages failed but results were produced. Use "error" only if no output was generated.
最后一步，写入
~/.claude/skill-analytics/last-outcome-prospect-research-to-cadence.json
：
json
{"ts":"[UTC ISO8601]","skill":"prospect-research-to-cadence","version":"1.0.0","variant":"default",
 "status":"[success|partial|error]","runtime_ms":[estimated ms from start],
 "metrics":{"prospects_researched":[n],"emails_drafted":[n],"sequences_loaded":[n],"atl_count":[n],"icp_avg_score":[n]},
 "error":null,"session_id":"[YYYY-MM-DD]"}
若部分阶段失败但仍生成结果，状态设为"partial"。仅当未生成任何输出时设为"error"。