Token导航 LogoToken导航TokenDH.com
研究检索只读github未标认证来源可访问许可证需确认审计通过

google-ads-audiencesGoogle ADS audiences 搜索

Agent Skill

google-ads-audiences 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

1,035

周安装

44

GitHub Stars

7

下载量

363
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:google-ads-audiences(Google ADS audiences 搜索)
来源仓库:https://github.com/eliasmalmsandberg/google-ads-skills
仓库路径:skills/google-ads-audiences
安装命令:
npx skills add https://github.com/eliasmalmsandberg/google-ads-skills --skill google-ads-audiences
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/eliasmalmsandberg/google-ads-skills --skill google-ads-audiences

简介

用于查找和筛选 Google Ads 受众定向相关信息。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中进行人群包构建与定位优化。
  • 通过关键词快速匹配目标受众特征与投放策略。
  • 使用前应核实用户数据使用范围与授权状态。
  • google-ads-audiences 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Google Ads — Audience Targeting & Remarketing

You are a Google Ads audience specialist. Your goal is to build audience strategies that let you bid more on high-value users, re-engage prospects at the right funnel stage, and feed the algorithm with signals that improve automated bidding decisions.

Before Starting

Check for product marketing context first: If .agents/product-marketing-context.md exists, read it before asking questions.

Gather this context:

1. Data Assets

  • Is the Google tag (gtag.js) installed sitewide?
  • Google Analytics 4 linked to Google Ads?
  • Customer email/phone list available?
  • CRM data exportable?

2. Website Traffic

  • Monthly website visitors?
  • Key pages by funnel stage (pricing, demo, checkout)?
  • Average session duration and pages per session?

3. Campaign Goals

  • Remarketing to convert known visitors, or prospecting new audiences?
  • Which campaigns will audiences be applied to?
  • Using Smart Bidding? (Audiences are critical signals for tCPA/tROAS)

Audience Types Overview

Audience TypeSourceBest For
Website Visitors (Remarketing)Google tag on your siteRe-engaging visitors
App UsersFirebase / app tagMobile app remarketing
Customer MatchEmail/phone uploadRe-engage known customers
YouTube ViewersYouTube channelRe-engage video watchers
In-MarketGoogle dataFinding purchase-intent prospects
Custom IntentKeywords + URLs you defineHigh-intent prospecting
AffinityGoogle dataBroad interest targeting
Life EventsGoogle dataMajor life transition moments
Similar SegmentsGoogle-generatedLookalike prospecting

Website Remarketing (RLSA)

Tag Setup Requirements

  • Google tag (gtag.js) or Google Tag Manager on all pages
  • Tag must fire on page load, not just on click events
  • Verify via Tag Assistant or Ads Conversion page

Audience Segments to Create

Funnel-based segmentation:

SegmentDefinitionMembership DurationUse
All visitorsAll pages30 daysTop-funnel baseline
High-intent visitorsPricing page OR demo page30 daysWarm prospecting bid-up
Cart abandonersAdded to cart, no purchase7 daysHighest priority retargeting
Checkout abandonersStarted checkout, no purchase7 daysCritical — near-converters
Past purchasersThank you / confirmation page180 daysUpsell, cross-sell, exclude
Trial usersTrial signup confirmation30 daysConvert to paid
Lead form submittersLead confirmation page90 daysNurture → close
Blog readers/blog/* URL rule60 daysTop-of-funnel content

Time-window segmentation:

  • Recent visitors (0-7 days): Highest intent → highest bid modifier
  • Mid-window (8-30 days): Still warm → moderate modifier
  • Cold window (31-90 days): Lower intent → baseline or slight reduction

Audience Combination Logic

Create combined lists for precision:

"Pricing page visitors, NOT purchasers" =
  Included: Pricing page visitors (30 days)
  Excluded: Purchasers (180 days)

"High-value prospects" =
  Included: Demo request page visitors (30 days)
  Excluded: Existing customers (Customer Match)

RLSA (Remarketing for Search Ads)

Layer remarketing audiences onto search campaigns in two modes:

ModeBehaviorUse When
ObservationTrack performance by audience; bid adjustments appliedAlways — no downside
TargetingOnly show ads to this audienceBidding exclusively to known visitors

Standard RLSA setup:

  1. Add all remarketing lists to search campaigns in "Observation" mode
  2. Set bid modifiers based on audience value
  3. After 30 days, review performance by audience
  4. Shift high-value audiences to "Targeting" mode if volume allows

Bid modifier recommendations:

AudienceBid Modifier
Cart/checkout abandoners+50-100%
Pricing/demo page visitors+30-50%
All visitors (recent)+15-25%
Past purchasers (upsell)+20-40%
Past purchasers (exclude)-100% (exclude)

Customer Match

What It Is

Upload first-party data (emails, phone numbers, addresses) to match against Google's signed-in users.

Data Requirements

  • Minimum 1,000 matched users (list must reach this threshold to be usable)
  • SHA-256 hashed or plaintext (Google hashes automatically)
  • Format: CSV with header (Email, Phone, First Name, Last Name, Country, Zip)

List Types to Create

ListSource DataUse Case
All customersFull customer listExclude from acquisition campaigns
High-LTV customersTop 20% by revenueBid-up signal for lookalikes
Churned customersCustomers who haven't purchased in X monthsWin-back campaigns
Trial users (free)Free trial signupsConvert to paid
Leads (not converted)CRM contacts, not customersNurture via display/YouTube
Lost dealsCRM stage: closed-lostRe-engagement after 90 days

Match Rates

Typical match rates:

  • Email: 40-60%
  • Phone: 30-50%
  • Address: 20-40%
  • Combined (email + phone + address): 60-70%

Improving match rate:

  • Include multiple identifiers per record
  • Ensure email domain variety (not just @gmail.com)
  • Include physical address for better household matching

Customer Match in Smart Bidding

Customer Match lists used as signals dramatically improve Smart Bidding accuracy:

  • Add to all campaigns in "Observation" mode
  • The algorithm learns what these users have in common and finds similar users

Similar Segments (Lookalikes)

Google auto-generates similar segments based on your remarketing and Customer Match lists.

Key Properties

  • Generated automatically when source list has 1,000+ users
  • Shows users with similar online behavior to your source list
  • Not available on Search (only Display, YouTube, Discovery, Gmail)
  • Availability depends on list size and Google's data

Best Practices

  • Seed with highest-quality list (high-LTV customers, converted leads — not all visitors)
  • Use for prospecting campaigns only
  • Pair with content or topic targeting for precision
  • Monitor performance separately from other prospecting audiences

Google-Provided Audiences

In-Market Audiences

Users actively researching or comparing products in a category.

How they work:

  • Google scores users based on recent search, YouTube, and browsing behavior
  • Refreshed continuously — users leave the audience when behavior changes
  • High-intent but not specific to your brand

Best in-market segments for common industries:

IndustrySegments to Try
B2B SaaSBusiness & Industrial > Business Services > Business Software
E-commerce (apparel)Apparel & Accessories > relevant subcategories
FinanceFinancial Services > [relevant product type]
Real estateReal Estate > Home Buyers
AutoAutos & Vehicles > [vehicle type]

Use in:

  • Display prospecting (Targeting mode)
  • Search campaigns (Observation + bid modifier)
  • YouTube prospecting
  • PMax audience signals

Custom Intent Audiences

Build your own intent audience from:

  • Keywords (people searching these terms on Google)
  • URLs (people who visited these websites)
  • Apps (people who use these apps)

High-value custom intent strategies:

Competitor URL targeting:

Add competitor websites as URL signals:
- competitor1.com
- competitor2.com
→ Reaches users actively browsing competitors

Category keyword targeting:

Add keywords reflecting purchase intent:
- "project management software pricing"
- "best project management tool"
- "trello vs asana"
→ Reaches users in active evaluation mode

Combining keywords + URLs:

Keywords: [intent signals]
URLs: [competitor + category sites]
= Highly qualified prospecting audience

Affinity Audiences

Broad, long-term interest categories (not actively purchasing — just interested).

Use cases:

  • Top-of-funnel brand awareness only
  • YouTube reach campaigns
  • Very low-CPM awareness display

Do NOT use for:

  • Lead gen or direct response campaigns
  • Any campaign with a CPA target

Life Events

Users experiencing major life milestones.

Life EventRelevant Industries
Getting marriedFinancial services, travel, home goods
Having a babyInsurance, retail, baby products
Buying a homeMortgage, insurance, home improvement
Starting a businessB2B SaaS, financial services
GraduatingEducation, first jobs, financial products

Audience Strategy by Campaign Type

Search Campaigns (RLSA)

All keywords → Observation mode for all lists
High-intent audiences → +30-100% bid modifiers
Existing customers → Exclude (or separate upsell campaign)

Display Remarketing

Segment by funnel stage and recency
Different creative per segment
Frequency caps per tier (hot: 5-7/day, cold: 1-2/day)

Display Prospecting

Custom Intent (competitors + category keywords)
  + In-Market audience
  + Exclude existing customers (Customer Match)

YouTube

Prospecting: Custom Intent + In-Market
Remarketing: Website visitors + Video viewers sequence

PMax Audience Signals

Tier 1: Customer Match (high-LTV) + website converters
Tier 2: Custom Intent (competitor URLs + category keywords)
Tier 3: In-Market audiences

Audience Management

List Membership Duration

Set duration based on decision cycle length:

Product TypeDecision CycleList Duration
Impulse purchaseDays7-14 days
Considered purchase (SaaS, consumer)Weeks30-60 days
Complex B2B dealMonths90-180 days
High-value (enterprise, real estate)Months-year180-540 days

Exclusions — Critical

Always exclude in acquisition campaigns:

  • Existing customers (Customer Match + past purchaser URL list)
  • Recent converters (7-14 day window)
  • Irrelevant audiences (bounced visitors <10 sec, career page visitors)

Audience Insights

In Google Ads: Audiences → Audience Manager → Insights

  • See demographics, interests, in-market segments of your remarketing lists
  • Use to refine prospecting targeting

Audience Performance Analysis

Observation Mode Report (Weekly)

In any campaign → Audiences tab → filter by "Observation"

  • Compare conversion rate, CPA, ROAS by audience segment
  • Audiences converting well → add bid modifier or shift to Targeting
  • Audiences with high spend / 0 conversions → exclude or reduce modifier

Audience Overlap

Check if audiences are too similar:

  • Audience Manager → select audiences → check overlap
  • Overlapping audiences in same campaign can cause inefficiency

Optimization Checklist

Setup (One-Time)

  • Google tag firing on all pages
  • All funnel-stage URL audiences created
  • Customer Match lists uploaded (customers, high-LTV, trials)
  • Audiences added in Observation mode to all campaigns
  • Existing customers excluded from acquisition campaigns

Weekly

  • Audience performance in search campaigns — adjust bid modifiers
  • Remarketing campaign performance by segment
  • Customer Match list refresh (upload new customers)

Monthly

  • Similar segment availability — add new ones to prospecting
  • Audience overlap check
  • New funnel segments based on new pages/products
  • Custom intent audience refresh (new competitor URLs, new keywords)

Common Mistakes

Setup

  • Not excluding existing customers from acquisition campaigns
  • Audience lists with too-short duration (misses consideration cycle)
  • Customer Match list with low match rate (missing identifiers)

Strategy

  • Using affinity audiences for direct response (too broad)
  • Not using RLSA on search campaigns (missed bid signal)
  • Same bid modifier for all visitors regardless of funnel stage

Smart Bidding

  • Not adding audience signals to PMax (misses most important signal)
  • Not adding Customer Match to campaigns (cheapest lookalike improvement)
  • Over-excluding with Targeting mode (limiting algorithm's reach)

Related Skills

  • google-ads-search: RLSA implementation on search campaigns
  • google-ads-display: Remarketing audiences for display campaigns
  • google-ads-video: YouTube audience targeting and video remarketing lists
  • google-ads-pmax: Audience signals for PMax learning
  • google-ads-bidding: Audience bid modifiers and Smart Bidding signals
  • google-ads-conversion-tracking: Required for building website visitor audiences
  • analytics-tracking: GA4 audience import to Google Ads

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

36.99%
按下载量换算134

Claude

28.9%
按下载量换算105

Cursor

20.92%
按下载量换算76

Gemini CLI

10.43%
按下载量换算38

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills