Agent Skills › aitytech/agentkits-marketing › analytics-attribution

analytics-attribution

GitHub

提供营销分析、归因建模及ROI评估能力，涵盖追踪设置、渠道表现分析、仪表盘创建及数据差异排查，支持多模型归因与漏斗阶段KPI计算。

skills/analytics-attribution/SKILL.md aitytech/agentkits-marketing

Trigger Scenarios

设置营销追踪和测量分析活动或渠道表现构建归因模型创建仪表板和报告计算营销ROI和CAC/LTV

Install

npx skills add aitytech/agentkits-marketing --skill analytics-attribution -g -y

More Options

Use without installing

npx skills use aitytech/agentkits-marketing@analytics-attribution

指定 Agent (Claude Code)

npx skills add aitytech/agentkits-marketing --skill analytics-attribution -a claude-code -g -y

安装 repo 全部 skill

npx skills add aitytech/agentkits-marketing --all -g -y

预览 repo 内 skill

npx skills add aitytech/agentkits-marketing --list

SKILL.md

Frontmatter

{
    "name": "analytics-attribution",
    "brand": "AgentKits Marketing by AityTech",
    "agents": [
        "project-manager",
        "researcher"
    ],
    "version": "1.0.0",
    "category": "core",
    "triggers": [
        "analytics",
        "attribution",
        "tracking",
        "ROI",
        "CAC",
        "LTV",
        "dashboard",
        "reporting",
        "conversion tracking",
        "marketing metrics"
    ],
    "difficulty": "advanced",
    "description": "Performance measurement, attribution modeling, and marketing ROI analysis. Use when setting up tracking, analyzing campaign performance, building attribution models, or creating marketing reports.",
    "prerequisites": [
        "marketing-fundamentals"
    ],
    "related_skills": [
        "ab-test-setup",
        "paid-advertising"
    ],
    "success_metrics": [
        "tracking_accuracy",
        "attribution_confidence"
    ],
    "mcp_integrations": {
        "optional": [
            "google-analytics",
            "google-search-console"
        ]
    }
}

Analytics & Attribution

Performance measurement and attribution modeling for data-driven marketing decisions.

Language & Quality Standards

CRITICAL: Respond in the same language the user is using. If Vietnamese, respond in Vietnamese. If Spanish, respond in Spanish.

Standards: Token efficiency, sacrifice grammar for concision, list unresolved questions at end.

When to Use This Skill

Apply analytics expertise when:

Setting up marketing tracking and measurement
Analyzing campaign or channel performance
Building attribution models
Creating dashboards and reports
Calculating marketing ROI and CAC/LTV
Troubleshooting data discrepancies

Core Concepts

Analytics Framework

Dimensions (What you're measuring by):

Channel, campaign, source/medium
Device, geography, time period
Audience segment, persona
Content type, landing page

Metrics (What you're measuring):

Traffic: Sessions, users, pageviews
Engagement: Time on site, bounce rate, pages/session
Conversion: Goal completions, conversion rate
Revenue: Transaction value, ROAS, ROI
Cost: CPC, CPL, CAC

Key Marketing Reports

Report	Questions Answered	Frequency
Acquisition	Where do visitors come from?	Weekly
Behavior	What do they do on site?	Weekly
Conversion	Do they complete goals?	Daily
Attribution	What drove the conversion?	Monthly
Funnel	Where do they drop off?	Weekly
Cohort	How do segments perform over time?	Monthly

Attribution Models

Model	Credit Distribution	Best For
Last Click	100% to final touchpoint	Short cycles, direct response
First Click	100% to first touchpoint	Brand awareness, TOFU
Linear	Equal across all	Understanding full journey
Time Decay	More to recent touches	Long sales cycles
Position-Based	40/20/40 first-mid-last	Balanced view
Data-Driven	ML-based distribution	High volume, mature programs

Marketing KPIs by Funnel Stage

TOFU (Awareness)

Impressions, reach, traffic
CPM, cost per visitor
Brand search volume

MOFU (Consideration)

Leads, MQLs, engagement
CPL, cost per MQL
Content downloads, webinar registrations

BOFU (Decision)

SQLs, opportunities, customers
CAC, cost per opportunity
Demo requests, trial signups

Retention

NPS, retention rate, churn
LTV, expansion revenue
Referrals, advocacy

Best Practices

Setup Excellence

UTM Discipline: Consistent naming convention across all campaigns
Goal Hierarchy: Primary conversions > secondary > micro-conversions
Cross-Domain Tracking: Proper setup for checkout/payment flows
Event Taxonomy: Clear naming for custom events

Reporting Excellence

Context Always: Never report numbers without comparison (vs target, vs previous)
Action-Oriented: Every insight should suggest an action
Visualization: Use appropriate chart types (trends=line, comparison=bar)
Segmentation: Break down by meaningful dimensions

Attribution Excellence

Window Matching: Attribution window matches sales cycle
Model Selection: Choose model based on marketing maturity
Multi-Touch Visibility: Track full journey, not just last touch
Offline Integration: Include phone, events, direct sales

Agent Integration

Agent	How They Use This Skill
`researcher`	Compiling performance data, competitive benchmarks
`lead-qualifier`	Funnel conversion analysis, lead source quality
`planner`	Budget allocation based on channel ROI
`project-manager`	Campaign performance tracking

Anti-Patterns to Avoid

Anti-Pattern	Why It's Wrong	Do This Instead
Vanity metrics only	Impressions ≠ impact	Focus on conversion metrics
Last-click bias	Ignores awareness touchpoints	Use multi-touch attribution
No control groups	Can't prove causation	A/B test when possible
Siloed data	Missing full picture	Integrate CRM + analytics
Report without action	Wastes time and attention	Include recommendations