sales-prospect-list

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Build a Targeted Prospect List

构建精准的潜在客户列表

Help the user build a targeted B2B prospect list — from ICP definition through filter strategy, list sizing, segmentation, and quality checks. This skill is platform-agnostic but references Apollo.io as the primary tool. The same frameworks apply to ZoomInfo, LinkedIn Sales Navigator, Clay, Lusha, or any data provider.

帮助用户构建精准的B2B潜在客户列表——从理想客户画像（ICP）定义，到筛选策略、列表规模确定、细分以及质量检查。本Skill不绑定特定平台，但以Apollo.io作为主要参考工具。相同的框架也适用于ZoomInfo、LinkedIn Sales Navigator、Clay、Lusha或任何数据提供商。

Step 1 — Gather context

步骤1 — 收集背景信息

Ask the user:

Who are you trying to reach?
- A) I have a clear ICP — I know the titles, industries, and company sizes
- B) I have a rough idea — need help refining it
- C) Starting from scratch — help me define who to target
- D) I have an existing list that needs filtering/cleaning
What's the campaign goal?
- A) Cold outbound — net-new pipeline generation
- B) ABM — targeted outreach to named accounts
- C) Event follow-up — building a list around a conference/webinar
- D) Expansion — finding new contacts at existing customer accounts
- E) Competitive displacement — targeting competitors' customers
- F) Other — describe it
What data provider(s) do you have access to?
- A) Apollo.io
- B) LinkedIn Sales Navigator
- C) ZoomInfo
- D) Clay
- E) Lusha / Cognism / other
- F) Just a CRM with existing data
- G) None yet — recommend one
How many prospects do you need? (rough target — 100? 500? 5,000?)

If the user's request already provides most of this context, skip directly to the relevant step. Lead with your best-effort answer using reasonable assumptions (stated explicitly), then ask only the most critical 1-2 clarifying questions at the end — don't gate your response behind gathering complete context.

询问用户：

你希望触达哪些人群？
- A) 我有明确的ICP——清楚目标职位、行业和公司规模
- B) 我有大致想法——需要帮助细化
- C) 从零开始——帮我定义目标人群
- D) 我已有列表——需要筛选/清理
营销活动的目标是什么？
- A) 冷触达——挖掘全新销售管道
- B) ABM（基于客户的营销）——针对指定客户的精准触达
- C) 活动跟进——围绕会议/网络研讨会构建列表
- D) 客户拓展——在现有客户中寻找新联系人
- E) 竞品客户转化——针对竞品的客户进行触达
- F) 其他——请描述
你可以使用哪些数据提供商？
- A) Apollo.io
- B) LinkedIn Sales Navigator
- C) ZoomInfo
- D) Clay
- E) Lusha / Cognism / 其他
- F) 仅拥有包含现有数据的CRM
- G) 还没有——请推荐一个
你需要多少个潜在客户？（大致目标——100？500？5000？）

如果用户的请求已提供大部分此类背景信息，可直接跳至相关步骤。 先基于合理假设给出初步回答（需明确说明假设），最后仅询问1-2个最关键的澄清问题——不要要求用户提供完整背景后才给出回应。

Step 2 — Define the ICP

步骤2 — 定义ICP

Build a structured ICP using four dimensions:

从四个维度构建结构化的ICP：

Firmographic filters (company-level)

公司层面的企业属性筛选器

Filter	Questions to ask	Example values
Industry	What industries are your best customers in?	SaaS, Financial Services, Healthcare
Company size	How many employees? What revenue range?	50-500 employees, $10M-$100M revenue
Geography	Where are your target companies?	US, UK, DACH region
Funding stage	Do you target by funding?	Series A-C, bootstrapped, public
Technology	What tech stack signals a good fit?	Uses Salesforce, runs Kubernetes, has React in stack

筛选维度	需要询问的问题	示例值
行业	你的最佳客户来自哪些行业？	SaaS、金融服务、医疗健康
公司规模	员工数量？营收范围？	50-500名员工，1000万-1亿美元营收
地域	目标公司位于哪些地区？	美国、英国、DACH地区（德国、奥地利、瑞士）
融资阶段	是否按融资阶段筛选目标客户？	A-C轮融资、bootstrapped（自筹资金）、上市公司
技术栈	哪些技术栈信号表明客户匹配度高？	使用Salesforce、运行Kubernetes、技术栈包含React

Demographic filters (person-level)

个人层面的人口属性筛选器

Filter	Questions to ask	Example values
Job title	What titles do you sell to?	VP Engineering, Head of Sales, CTO
Seniority	What level of decision-maker?	Director+, VP+, C-suite
Department	Which function?	Engineering, Sales, Marketing, Finance
Job function	What do they actually do?	People management, IC, operations

筛选维度	需要询问的问题	示例值
职位头衔	你的产品面向哪些职位的人群销售？	工程副总裁、销售主管、CTO
职级	目标决策者的级别？	总监及以上、副总裁及以上、高管层
部门	目标职能部门？	工程、销售、营销、财务
工作职责	他们的实际工作内容？	人员管理、独立贡献者（IC）、运营

Behavioral/signal filters

行为/信号筛选器

Filter	Questions to ask	Example values
Hiring signals	Are they hiring for roles your product replaces/supports?	Posting for SDRs, data engineers
Funding	Recent funding that creates budget?	Raised Series B in last 6 months
Job changes	New leaders who bring new tools?	Started VP role in last 90 days
Intent	Researching topics related to your product?	Intent topics in your category
Tech install	Recently adopted complementary/competitive tech?	Added Snowflake, removed competitor

筛选维度	需要询问的问题	示例值
招聘信号	他们是否在招聘你的产品可替代/支持的岗位？	发布SDR、数据工程师招聘信息
融资信号	近期融资是否带来了预算空间？	过去6个月内完成B轮融资
职位变动信号	新任领导层是否会引入新工具？	过去90天内担任副总裁职位
意向信号	是否在研究与你的产品相关的主题？	所在品类的意向主题
技术部署信号	近期是否采用了互补/竞争技术？	新增Snowflake、移除竞品工具

Negative filters (who to exclude)

排除筛选器（需要排除的对象）

Existing customers
Companies in active deals
Do-not-contact list
Companies too small or too large
Industries you don't serve
Geographies you can't support

现有客户
正在跟进的潜在客户
请勿联系名单
规模过小或过大的公司
不服务的行业
无法覆盖的地域

Step 3 — Build the list

步骤3 — 构建列表

Filter strategy by persona

按角色设计筛选策略

For each persona in the ICP, design the filter combination:

Primary persona (highest priority):

Filters: [specific combination]
Expected list size: [estimate]
Quality check: Does every record on this list make sense as a prospect?

Secondary persona (if applicable):

Filters: [adjusted combination]
Expected list size: [estimate]

针对ICP中的每个角色，设计筛选组合：

核心角色（最高优先级）：

筛选条件：[具体组合]
预期列表规模：[预估数量]
质量检查：列表中的每个记录是否都符合潜在客户标准？

次要角色（如适用）：

筛选条件：[调整后的组合]
预期列表规模：[预估数量]

List sizing guidance

列表规模指导

Campaign type	Recommended list size	Why
Hyper-targeted ABM	25-100	Quality over quantity — deep personalization per account
Focused outbound	200-500	Manageable for Level 3+ personalization
Scaled outbound	500-2,000	Balanced scale with Level 2 personalization
High-volume outbound	2,000-10,000	Volume play — Level 1-2 personalization, A/B testing at scale

The right size depends on your personalization capacity. A list of 5,000 is useless if your team can only personalize 50 emails per week.

营销活动类型	推荐列表规模	原因
高度精准的ABM	25-100	质量优先——针对每个客户进行深度个性化触达
聚焦式冷触达	200-500	便于开展3级及以上的个性化触达
规模化冷触达	500-2000	平衡规模与2级个性化触达
高容量冷触达	2000-10000	以量取胜——1-2级个性化触达，规模化A/B测试

合适的规模取决于你的个性化能力。如果团队每周只能个性化50封邮件，那么5000人的列表毫无用处。

In Apollo.io

在Apollo.io中操作

Use Apollo's People Search with these filter categories:

Job Titles: Use "is any of" for title matching — Apollo supports exact and fuzzy matching
Seniority: Filter by management level (C-suite, VP, Director, Manager)
Company: Employee count ranges, revenue ranges, industry tags
Location: Country, state, city, or metro area
Technologies: Filter by tech stack (Apollo tracks 10,000+ technologies)
Signals: Job changes, funding events, hiring activity

Save the search as a dynamic list — new matches will appear automatically.

使用Apollo的人员搜索功能，结合以下筛选类别：

职位头衔：使用“匹配任意”进行头衔匹配——Apollo支持精确和模糊匹配
职级：按管理层级别筛选（高管层、副总裁、总监、经理）
公司：员工数量范围、营收范围、行业标签
地域：国家、州、城市或都会区
技术栈：按技术栈筛选（Apollo追踪10000+种技术）
信号：职位变动、融资事件、招聘活动

将搜索保存为动态列表——符合条件的新联系人会自动加入。

Exporting for Mailshake

导出至Mailshake

Export CSV with columns mapping to Mailshake recipient fields
Required:
```
email
```
(only mandatory column)
Recommended:
```
first
```
,
```
last
```
,
```
company
```
,
```
title
```
→ merge fields {{first}}, {{last}}, {{company}}, {{title}}
Custom fields: additional CSV columns become custom merge fields
Pre-send: verify emails before import — Mailshake does not verify on upload. See
```
/sales-deliverability
```

导出CSV文件，列需映射到Mailshake的收件人字段
必填项：
```
email
```
（唯一必填列）
推荐项：
```
first
```
,
```
last
```
,
```
company
```
,
```
title
```
→ 合并字段{{first}}, {{last}}, {{company}}, {{title}}
自定义字段：额外的CSV列将成为自定义合并字段
发送前：导入前验证邮箱——Mailshake不会在上传时验证邮箱。请查看/sales-deliverability

Importing into Lemlist

导入至Lemlist

CSV import: Upload CSV with
```
email
```
as the only required column. Recommended fields:
```
firstName
```
,
```
lastName
```
,
```
companyName
```
, plus any custom fields for personalization variables.
API import:
```
POST /api/campaigns/{id}/leads
```
with lead objects containing email and custom variables. See
```
/sales-lemlist
```
→
```
references/lemlist-api-reference.md
```
. Rate limit: 20 requests per 2 seconds.
People Database: Alternative to external list building — Lemlist's built-in 600M+ contact database with smart filtering (title, seniority, industry, company size, intent signals). Leads are pre-verified. Pay per lead via credits.
CRM import: Import leads directly from connected HubSpot, Salesforce, or Pipedrive via the CRM integration.
Pre-send: Enable Lemwarm on all email accounts and warm up 3-5 weeks before launching sequences. Verify emails before import to keep bounce rate <3%.

CSV导入：上传CSV文件，
```
email
```
为唯一必填列。推荐字段：
```
firstName
```
,
```
lastName
```
,
```
companyName
```
，以及用于个性化变量的自定义字段。
API导入：使用
```
POST /api/campaigns/{id}/leads
```
接口，传入包含邮箱和自定义变量的线索对象。请查看/sales-lemlist →
```
references/lemlist-api-reference.md
```
。速率限制：每2秒20次请求。
人员数据库：替代外部列表构建的方案——Lemlist内置6亿+联系人数据库，支持智能筛选（头衔、职级、行业、公司规模、意向信号）。线索已预先验证，通过积分按条付费。
CRM导入：通过CRM集成直接从关联的HubSpot、Salesforce或Pipedrive导入线索。
发送前：为所有邮箱账户启用Lemwarm，并在启动序列前预热3-5周。导入前验证邮箱，将 bounce 率控制在3%以下。

Importing into Yesware

导入至Yesware

Individual add: Add recipients one at a time within a campaign — enter name, email, and custom fields
Bulk import: Upload a CSV or import from Salesforce list views (Enterprise plan)
Prospector: Alternatively, use Yesware's built-in Prospector (100M+ B2B contacts) to find and add leads directly — search by title, industry, company size, and more
Salesforce import: Enterprise plan users can import leads/contacts directly from Salesforce list views into campaigns
Pre-send: Verify emails before import — Yesware does not verify on upload. Keep bounce rate <3%.
Recipient limits: Free plan = 10 recipients/month, Pro = 20/month, Premium+ = unlimited

单个添加：在活动中逐个添加收件人——输入姓名、邮箱和自定义字段
批量导入：上传CSV文件或从Salesforce列表视图导入（仅企业版）
Prospector工具：替代方案，使用Yesware内置的1亿+B2B联系人数据库直接查找并添加线索——按头衔、行业、公司规模等搜索
Salesforce导入：企业版用户可直接从Salesforce列表视图导入线索/联系人至活动
发送前：导入前验证邮箱——Yesware不会在上传时验证邮箱。将 bounce 率控制在3%以下。
收件人限制：免费版=每月10个收件人，专业版=每月20个，高级版及以上=无限制

Importing into Mixmax

导入至Mixmax

From Gmail: Mixmax lives in Gmail — any contact you email is automatically available
From CSV: Upload CSV to add recipients to sequences (map email, first name, last name, custom fields)
From Salesforce: Import Salesforce leads/contacts directly into sequences (Growth+CRM plan); use Salesforce list views or reports as source
From HubSpot: Import HubSpot contact lists into Mixmax sequences (Growth plan)
Via API:
```
POST /sequences/:id/recipients
```
to programmatically add recipients to a sequence
Via rules: Auto-enroll contacts based on Salesforce triggers (e.g., new lead created → add to sequence)
Limits: Recipient limits per sequence vary by plan; check Mixmax plan details

从Gmail导入：Mixmax集成于Gmail——你发送过邮件的所有联系人会自动可用
从CSV导入：上传CSV文件将收件人添加至序列（映射邮箱、名字、姓氏、自定义字段）
从Salesforce导入：直接从Salesforce导入线索/联系人至序列（仅Growth+CRM版）；使用Salesforce列表视图或报表作为数据源
从HubSpot导入：将HubSpot联系人列表导入Mixmax序列（仅Growth版）
通过API导入：使用
```
POST /sequences/:id/recipients
```
接口以编程方式将收件人添加至序列
通过规则导入：基于Salesforce触发器自动添加联系人（例如：新建线索 → 加入序列）
限制：每个序列的收件人限制因套餐而异，请查看Mixmax套餐详情

Importing into Smartlead

导入至Smartlead

CSV import: Upload CSV with
```
email
```
as the only required column. Recommended fields:
```
first_name
```
,
```
last_name
```
,
```
company
```
, plus any custom fields for merge variables.

API import:

POST /api/v1/campaigns/{id}/leads

with a

lead_list

array containing lead objects. See

/sales-smartlead

→

references/smartlead-api-reference.md

SmartProspect: Alternative to external list building — Smartlead's built-in verified lead database with intent signals. 3x email verification (syntax, domain, mailbox). Pay per verified lead via credits.
Pre-send: Verify emails before import to keep bounce rate <3%. SmartProspect leads are pre-verified; external lists are not.

CSV导入：上传CSV文件，
```
email
```
为唯一必填列。推荐字段：
```
first_name
```
,
```
last_name
```
,
```
company
```
，以及用于合并变量的自定义字段。
API导入：使用
```
POST /api/v1/campaigns/{id}/leads
```
接口，传入包含
```
lead_list
```
数组的线索对象。请查看/sales-smartlead →
```
references/smartlead-api-reference.md
```
。
SmartProspect工具：替代外部列表构建的方案——Smartlead内置已验证的线索数据库，包含意向信号。提供3重邮箱验证（语法、域名、邮箱有效性），通过积分按已验证线索付费。
发送前：导入前验证邮箱，将 bounce 率控制在3%以下。SmartProspect线索已预先验证；外部列表未验证。

Importing into Woodpecker

导入至Woodpecker

From Lead Finder: Built-in B2B database with 1B+ leads — search by company, title, industry, location. Uses data credits (400 free, more from €28/2K credits). Export directly into campaigns.
From CSV: Upload CSV to add prospects to campaigns (map email, first name, last name, company, custom fields)
From Google Sheets: Native integration for ongoing prospect sync
From HubSpot/Pipedrive: Native CRM integrations sync contacts into Woodpecker
Via API:
```
POST /prospects
```
to create prospects, then add to campaigns via campaign endpoints
Via Zapier/Clay: Connect any lead source to Woodpecker prospect creation
Auto-verification: Woodpecker automatically validates prospect emails via Bouncer when added — invalid emails are flagged before sending
Limits: Pricing based on "contacted prospects" per month — every new person added to a campaign counts against limit

从Lead Finder导入：内置10亿+B2B线索数据库——按公司、头衔、行业、地域搜索。使用数据积分（免费400个，额外积分28欧元/2000个）。可直接导出至活动。
从CSV导入：上传CSV文件将潜在客户添加至活动（映射邮箱、名字、姓氏、公司、自定义字段）
从Google Sheets导入：原生集成支持持续的潜在客户同步
从HubSpot/Pipedrive导入：原生CRM集成可将联系人同步至Woodpecker
通过API导入：使用
```
POST /prospects
```
接口创建潜在客户，再通过活动接口添加至活动
通过Zapier/Clay导入：连接任意线索源至Woodpecker的潜在客户创建功能
自动验证：Woodpecker通过Bouncer自动验证添加的潜在客户邮箱——无效邮箱会在发送前被标记
限制：定价基于每月“已联系潜在客户”数量——每个添加至活动的新人均会占用额度

Importing into Reply.io

导入至Reply.io

From B2B database: Reply.io has a built-in database of 1B+ contacts — search by industry, company size, title, location, tech stack. Use data credits to reveal emails/phones.
From CSV: Upload CSV to add contacts to sequences (map email, first name, last name, company, custom fields)
From Chrome extension: Reply.io email finder extension finds emails while browsing LinkedIn or company websites
From Salesforce/HubSpot: Native 2-way sync imports CRM contacts directly into Reply.io
Via API:
```
POST /contacts
```
to create contacts, then add to sequences via Sequence Contacts endpoints
Via Zapier/Make: Connect any lead source to Reply.io contact creation
Limits: Unlimited contact storage on Multichannel plan; data credits required for B2B database reveals

从B2B数据库导入：Reply.io内置10亿+联系人数据库——按行业、公司规模、头衔、地域、技术栈搜索。使用数据积分解锁邮箱/电话信息。
从CSV导入：上传CSV文件将联系人添加至序列（映射邮箱、名字、姓氏、公司、自定义字段）
从Chrome扩展导入：Reply.io邮箱查找扩展可在浏览LinkedIn或公司网站时查找邮箱
从Salesforce/HubSpot导入：原生双向同步可直接从CRM导入联系人至Reply.io
通过API导入：使用
```
POST /contacts
```
接口创建联系人，再通过序列联系人接口添加至序列
通过Zapier/Make导入：连接任意线索源至Reply.io的联系人创建功能
限制：多渠道版支持无限联系人存储；B2B数据库解锁需消耗数据积分

Step 4 — Segment and prioritize

步骤4 — 细分与优先级排序

Split the list into tiers for sequencing:

Tier	Criteria	Volume	Approach
Tier 1 — Hot	Matches ICP + has buying signal (intent, job change, funding)	10-20% of list	Highest personalization, multi-channel, fastest follow-up
Tier 2 — Warm	Matches ICP, no signal but strong fit	30-40% of list	Standard personalization, proven sequence
Tier 3 — Broad	Partial ICP match, worth testing	40-50% of list	Template-based, A/B test messaging

将列表分为不同层级用于触达序列：

层级	标准	占比	触达方式
1级——高优先级	匹配ICP + 存在购买信号（意向、职位变动、融资）	列表的10-20%	最高程度个性化、多渠道触达、最快跟进
2级——中优先级	匹配ICP、无信号但匹配度高	列表的30-40%	标准个性化、成熟触达序列
3级——广覆盖	部分匹配ICP、值得测试	列表的40-50%	模板化内容、规模化A/B测试

Lead qualification scoring

线索评分模型

Use a weighted scoring model to rank prospects numerically, not just by tier. This gives you a concrete starting point that you can tune based on conversion data.

Criterion	Weight	What to score	Example: 80+ = Tier 1
Title/role match	30%	How closely does the prospect's title match your buyer persona?	VP Engineering = 30, Director = 20, Manager = 10
Company size fit	25%	Does the company fall within your ICP's employee/revenue range?	Sweet spot = 25, adjacent = 15, edge = 5
Industry match	20%	Is the company in a target industry?	Primary industry = 20, adjacent = 10, other = 0
Role recency	15%	How recently did they start this role? (New leaders buy)	<90 days = 15, 90-365 days = 10, >1 year = 5
Buying signals	10%	Intent data, hiring signals, funding, tech changes	Active signal = 10, weak signal = 5, none = 0

Total: 100 points max. Tier 1 = 80+, Tier 2 = 50-79, Tier 3 = 30-49, Skip = <30.

Tune these weights for your motion: outbound-heavy teams should increase company size and industry weight; signal-rich teams (PLG, intent-driven) should increase buying signals weight. Review and recalibrate quarterly using actual reply/meeting rates per score band.

使用加权评分模型对潜在客户进行数字排名，而非仅按层级划分。这能为你提供一个具体的起点，后续可根据转化数据调整。

评分项	权重	评分标准	示例：80分以上=1级
头衔/角色匹配度	30%	潜在客户的头衔与你的买家角色匹配度如何？	工程副总裁=30分，总监=20分，经理=10分
公司规模匹配度	25%	公司是否符合ICP的员工/营收范围？	核心区间=25分，相邻区间=15分，边缘区间=5分
行业匹配度	20%	公司是否属于目标行业？	核心行业=20分，相邻行业=10分，其他=0分
职位任职时长	15%	他们担任该职位的时长？（新任领导层更易采购）	少于90天=15分，90-365天=10分，超过1年=5分
购买信号	10%	意向数据、招聘信号、融资、技术变动	明确信号=10分，弱信号=5分，无信号=0分

总分：最高100分。1级=80分以上，2级=50-79分，3级=30-49分，排除=30分以下。

根据你的业务模式调整权重：以 outbound 为主的团队可提高公司规模和行业的权重；信号丰富的团队（PLG、意向驱动型）可提高购买信号的权重。每季度根据各分数段的实际回复/会议率回顾并重新校准。

Personalization angles

个性化方向

Capture personalization angles during list building — not later at outreach time. This bridges the gap between list building and sequence writing and makes the handoff to

/sales-cadence

much smoother.

Tier	What to capture	Why it matters
Tier 1 — Hot	Recent company news, specific tech stack observations, mutual connections, recent job change context, funding details	These prospects get the most personalized outreach — you need concrete hooks
Tier 2 — Warm	Industry pain points, company growth stage, hiring patterns, competitive tool usage	Personalization at the segment level — "companies like yours" not "your specific situation"
Tier 3 — Broad	Industry vertical, company size band, persona-level pain points	Template-level personalization — enough to feel relevant, not enough to be bespoke

For each Tier 1 prospect, note at least 2 personalization angles during list building. For Tier 2, note 1 per prospect or 1 per segment. Tier 3 gets segment-level angles only.

在列表构建阶段就捕捉个性化方向——而非在触达阶段再处理。这能弥合列表构建与序列撰写之间的差距，使向/sales-cadence的交接更顺畅。

层级	需要捕捉的信息	重要性
1级——高优先级	近期公司新闻、具体技术栈观察、共同联系人、近期职位变动背景、融资细节	这些潜在客户需要最高程度的个性化触达——你需要具体的切入点
2级——中优先级	行业痛点、公司增长阶段、招聘模式、竞品工具使用情况	细分层面的个性化——“与你类似的公司”而非“你的具体情况”
3级——广覆盖	行业垂直领域、公司规模区间、角色层面的痛点	模板级个性化——足够贴合但无需定制

对于每个1级潜在客户，在列表构建阶段至少记录2个个性化方向。2级潜在客户每个记录或每个细分记录1个。3级仅记录细分层面的方向。

Segmentation variables

细分变量

Beyond tiers, consider segmenting by:

Industry — messaging should reflect industry-specific pain points
Persona — a VP of Engineering and a CTO need different messaging
Company size — SMB and enterprise have different buying processes
Geography — timezone, language, regional references

Each segment should have tailored messaging in the sequence. One generic sequence for the whole list will underperform.

除层级外，还可考虑按以下维度细分：

行业——触达信息应反映行业特定痛点
角色——工程副总裁和CTO需要不同的触达信息
公司规模——中小企业和企业客户的采购流程不同
地域——时区、语言、地域参考

每个细分群体在序列中都应有定制化的触达信息。针对整个列表使用通用序列的效果会很差。

Step 5 — Quality checks

步骤5 — 质量检查

Before launching outreach, validate the list:

在启动触达前，验证列表：

Data quality checklist

数据质量检查清单

List health metrics

列表健康指标

Metric	Target	If below...
% with verified email	>80%	Enrich before sequencing — see `/sales-enrich`
% with phone number	>40%	Acceptable for email-first sequences; enrich for phone-heavy cadences
% matching all ICP criteria	>70%	Tighten filters — too many partial matches dilute effectiveness
Bounce rate (after first send)	<3%	List quality issue — use verification tools before sending

指标	目标值	若未达标...
已验证邮箱占比	>80%	触达前补充信息——请查看/sales-enrich
有电话号码占比	>40%	对于以邮箱为主的序列可接受；对于以电话为主的序列需补充信息
完全匹配ICP标准占比	>70%	收紧筛选条件——过多部分匹配会降低效果
首次发送后的 bounce 率	<3%	列表质量问题——发送前使用验证工具

Gotchas

注意事项

Don't build a list without defining the ICP first. Claude tends to jump straight to filter recommendations. A list without a clear ICP is just a random collection of people. Always start with "who are your best customers?" and work backward.
Don't optimize for list size over quality. Claude defaults to loosening filters to "get more results." A list of 200 perfect-fit prospects outperforms a list of 5,000 partial matches. Push back when the user asks for more volume at the expense of targeting.
Don't ignore negative filters. Excluding existing customers, active deals, and do-not-contact lists is as important as the inclusion criteria. Claude often forgets to ask about exclusions.
Don't assume job titles are standardized. "VP of Engineering" and "Vice President, Engineering" and "VP Engineering" are all the same person. "Head of Growth" could be marketing or sales. Always recommend title fuzzy matching or use Apollo's seniority filters alongside title keywords.
Don't skip the segmentation step. A flat, unsegmented list leads to generic messaging. Every list should be split into at least 2-3 segments based on persona, industry, or signal strength.

不要在未定义ICP的情况下构建列表。 Claude倾向于直接给出筛选建议。没有明确ICP的列表只是随机的人员集合。始终从“你的最佳客户是谁？”开始，反向推导。
不要为了列表规模牺牲质量。 Claude默认会放宽筛选条件以“获取更多结果”。200个完全匹配的潜在客户列表的效果优于5000个部分匹配的列表。当用户要求以牺牲精准度为代价增加数量时，应提出反对。
不要忽略排除筛选器。 排除现有客户、正在跟进的潜在客户和请勿联系名单与包含筛选条件同样重要。Claude经常忘记询问排除条件。
不要假设职位头衔是标准化的。 “工程副总裁”、“Vice President, Engineering”和“VP Engineering”都是指同一类人。“增长主管”可能属于营销或销售部门。始终建议使用头衔模糊匹配，或结合Apollo的职级筛选与头衔关键词。
不要跳过细分步骤。 扁平、未细分的列表会导致通用化的触达信息。每个列表至少应按角色、行业或信号强度分为2-3个细分群体。

Related skills

Examples

示例

Example 1: Building a cold outbound list

示例1：构建冷触达列表

User says: "Build a list of VP Engineering at Series B SaaS companies with 100-500 employees in the US" Skill does:

Confirms ICP dimensions (title, funding stage, size, geography)
Recommends Apollo search filters with exact settings
Suggests list size of 300-500, segmented by sub-industry (DevTools, MarTech, FinTech)
Provides quality checks before launching sequences Result: Targeted, segmented list ready for personalized outreach

用户需求：“构建美国地区、员工规模100-500人、B轮融资的SaaS公司中工程副总裁的列表” Skill操作：

确认ICP维度（头衔、融资阶段、规模、地域）
推荐Apollo搜索筛选的具体设置
建议列表规模为300-500，按细分行业（DevTools、MarTech、FinTech）细分
提供启动序列前的质量检查步骤结果：已完成细分的精准列表，可直接用于个性化触达

Example 2: ABM target account list

示例2：ABM目标客户列表

User says: "I need to find the buying committee at 20 target accounts in healthcare" Skill does:

Clarifies which 20 accounts and what roles are in the buying committee
Recommends person-level filters per account (CEO, CTO, VP Ops, CISO for each)
Suggests mapping 3-5 contacts per account for multi-threading
Provides an account-level prioritization framework Result: Multi-threaded contact list across 20 accounts, organized by role and engagement priority

用户需求：“我需要在20家医疗健康行业的目标客户中找到采购决策委员会成员” Skill操作：

明确20家具体客户以及采购决策委员会的角色
推荐针对每个客户的个人层面筛选条件（每家客户的CEO、CTO、运营副总裁、CISO）
建议每个客户匹配3-5个联系人，实现多线程触达
提供客户层面的优先级排序框架结果：覆盖20家客户的多线程联系人列表，按角色和参与优先级组织

Troubleshooting

故障排除

Too few results

结果过少

Cause: Filters are too restrictive Solution: Loosen one filter at a time. Start with geography (expand to adjacent markets), then company size (widen the range), then titles (add synonyms). Use Apollo's "similar titles" suggestions.

原因：筛选条件过于严格 解决方案：逐个放宽筛选条件。先从地域开始（扩展至相邻市场），然后是公司规模（扩大范围），再是头衔（添加同义词）。使用Apollo的“相似头衔”建议。

Too many results

结果过多

Cause: Filters are too broad Solution: Add a seniority filter (Director+ if you sell to leadership), narrow the industry (sub-industry instead of top-level), or add a technology filter to find companies using complementary tools.

原因：筛选条件过于宽泛 解决方案：添加职级筛选条件（若面向领导层销售则选择总监及以上）、缩小行业范围（使用细分行业而非顶级行业），或添加技术栈筛选条件以找到使用互补工具的公司。

List has low email coverage

列表邮箱覆盖率低

Cause: Many records don't have verified emails in the data provider Solution: Use

/sales-enrich

with waterfall enrichment (try multiple providers). For C-suite and VP-level contacts, direct phone outreach may be more effective than email anyway.

原因：数据提供商中许多记录没有已验证的邮箱 解决方案：使用/sales-enrich进行瀑布式补充（尝试多个提供商）。对于高管层和副总裁级别的联系人，直接电话触达可能比邮箱更有效。