Token导航 LogoToken导航TokenDH.com
研究检索需要联网clawhub未标认证来源可访问clear审计通过

meta-ads-audience-analysis元广告受众分析

Agent Skill

meta-ads-audience-analysis 用于查找、检索和筛选相关信息,适合在 OpenClaw 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

3,445

周安装

148

GitHub Stars

公开资料未说明

下载量

1,208
OpenClaw

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

MIT-0

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:meta-ads-audience-analysis(元广告受众分析)
来源仓库:https://github.com/elias-didoo/meta-ads-audience-analysis
安装命令:
openclaw skills install meta-ads-audience-analysis
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 OpenClaw 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。

ClawHubOpenClaw
openclaw skills install meta-ads-audience-analysis

简介

分析元广告受众类型、重叠、人口统计和预算分配效率。

  • 适合审查定位策略或优化受众细分时使用。meta-ads-audience-analysis 属于研究检索类 Skill,可作为该场景下的辅助能力补充。
  • 根据关键词、任务场景检索相关信息,结合来源仓库验证用法。
  • 安装前需确认权限范围及是否会触发联网、命令执行等操作。
  • 建议查阅原始文档了解具体输入参数和输出格式。

SKILL.md

name
meta-ads-audience-analysis
description
[Didoo AI] Analyzes Meta Ads audience efficiency across audience types, overlap, demographics, and budget allocation. Use when reviewing targeting strategy, planning audience expansion or narrowing, or auditing budget distribution across audience segments.
homepage
https://didoo.ai/blog
metadata

Required Credentials

CredentialWhere to GetUsed ForOAuth Scope
META_ACCESS_TOKENMeta Developer Console → Graph API Explorer → Generate TokenFetching audience breakdown and overlap dataads_read (read-only)
META_AD_ACCOUNT_IDAds Manager URL: adsmanager.facebook.com/act_XXXXXXXXXIdentifying which account to query

When to Use

Before launching new campaigns, during quarterly planning, or when a campaign is underperforming and audience mismatch is suspected. Run standalone or after meta-ads-weekly-performance identifies audience-related issues.


Requirements

  • account_id (required)
  • campaign_id (optional — defaults to all campaigns)
  • date_range: last_30d (default)

Step 1: Fetch Audience Data

Pull data broken down by audience dimension:

Audience Type Performance (via breakdown by audience_type)

  • Lookalike, Custom, Broad, Interest, Retargeting
  • Spend, CTR, CPA, ROAS for each

Audience Overlap (via Meta's overlap analysis tool)

  • Pairs of overlapping audiences
  • Overlap percentage, combined spend, ROAS

Demographic Breakdown (via breakdown by age and gender)

  • Age × Gender cells
  • Spend, conversions, CPA, ROAS

Placement Breakdown

  • Feed, Stories, Reels, Instant Articles, Audience Network, Messenger
  • CTR, CVR, CPA, ROAS for each

Step 2: Audience Type Efficiency

Audience TypeSpend% of BudgetCTRCPAROASBudget Should Be

Identify:

  • Which audience type delivers best ROAS
  • Which is over/underallocated relative to efficiency
  • Lookalike quality (which LAL layers performing)

Step 3: Audience Overlap Analysis

Audience NameOverlaps WithOverlap %Combined SpendROASAction
Consolidate / Exclude / Separate / Monitor

Flag:

  • Overlap > 30% with high combined spend → Consolidate
  • Overlap > 50% → Exclude one
  • Different audiences with same LAL source → Separate

Step 4: Demographic Deep Dive

Age and GenderSpendConversionsCPAROASAction
Increase Bid / Decrease Bid / Exclude / Maintain

Identify:

  • Highest and lowest ROAS demographic cells
  • Unexpected performance by gender or age
  • Bid adjustment opportunities

Step 5: Placement Breakdown

PlacementSpend% of BudgetCTRCVRCPAROASAction
Scale / Maintain / Reduce / Exclude

Flag:

  • Audience Network with low ROAS → Exclude
  • Stories vs Feed efficiency gap > 30% → Reduce lower performer
  • Reels growing but underallocated → Scale

Step 6: Budget Reallocation Plan

From Audience/PlacementTo Audience/PlacementShift AmountExpected ROAS ImpactRevenue ImpactPriority
High / Medium / Low

Prioritize moves by:

  1. Highest ROAS differential
  2. Enough data to trust the signal (minimum 50 results per segment)
  3. Realistic shift amounts (don't move more than 20% at once)

Step 7: Output Format

SECTION 1: Audience Type Efficiency

2–3 audience type issues in problem/solution format.

SECTION 2: Audience Overlap and Waste

2 overlap issues in problem/solution format.

SECTION 3: Placement Breakdown

2 placement issues in problem/solution format.

SECTION 4: Demographic Deep Dive

2–3 demographic issues in problem/solution format.

SECTION 5: Budget Reallocation Plan

1–2 reallocation strategy issues in problem/solution format.


Rules

  • No executive summary
  • Minimum 50 results per segment before drawing conclusions on any audience slice
  • Action must be specific: "Increase bid 20% on Women 25-34" not "adjust demographics"
  • This is analysis only — recommendations route to meta-ads-recommendation

Session Context — What This Skill Writes

After completing analysis, store the following in session context:

KeyDescriptionExample
budget_reallocation_planSpecific shifts between audience segments"$200/day from Broad to LAL 1% Women 25-34"
audience_issuesOverlap, waste, or misallocation findings"Audience A overlaps with B by 41% — consolidate"
meta-ads-recommendation reads these keys to produce the audience action plan.

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

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

补充不同宿主或平台的使用分布数据

能力 5

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

OpenClaw

91.53%
按下载量换算1,106

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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