$npx -y skills add aitytech/agentkits-marketing --skill analytics-attributionPerformance measurement, attribution modeling, and marketing ROI analysis. Use when setting up tracking, analyzing campaign performance, building attribution models, or creating marketing reports.
| 1 | # Analytics & Attribution |
| 2 | |
| 3 | Performance measurement and attribution modeling for data-driven marketing decisions. |
| 4 | |
| 5 | ## Language & Quality Standards |
| 6 | |
| 7 | **CRITICAL**: Respond in the same language the user is using. If Vietnamese, respond in Vietnamese. If Spanish, respond in Spanish. |
| 8 | |
| 9 | **Standards**: Token efficiency, sacrifice grammar for concision, list unresolved questions at end. |
| 10 | |
| 11 | --- |
| 12 | |
| 13 | ## When to Use This Skill |
| 14 | |
| 15 | Apply analytics expertise when: |
| 16 | - Setting up marketing tracking and measurement |
| 17 | - Analyzing campaign or channel performance |
| 18 | - Building attribution models |
| 19 | - Creating dashboards and reports |
| 20 | - Calculating marketing ROI and CAC/LTV |
| 21 | - Troubleshooting data discrepancies |
| 22 | |
| 23 | ## Core Concepts |
| 24 | |
| 25 | ### Analytics Framework |
| 26 | |
| 27 | **Dimensions** (What you're measuring by): |
| 28 | - Channel, campaign, source/medium |
| 29 | - Device, geography, time period |
| 30 | - Audience segment, persona |
| 31 | - Content type, landing page |
| 32 | |
| 33 | **Metrics** (What you're measuring): |
| 34 | - Traffic: Sessions, users, pageviews |
| 35 | - Engagement: Time on site, bounce rate, pages/session |
| 36 | - Conversion: Goal completions, conversion rate |
| 37 | - Revenue: Transaction value, ROAS, ROI |
| 38 | - Cost: CPC, CPL, CAC |
| 39 | |
| 40 | ### Key Marketing Reports |
| 41 | |
| 42 | | Report | Questions Answered | Frequency | |
| 43 | |--------|-------------------|-----------| |
| 44 | | Acquisition | Where do visitors come from? | Weekly | |
| 45 | | Behavior | What do they do on site? | Weekly | |
| 46 | | Conversion | Do they complete goals? | Daily | |
| 47 | | Attribution | What drove the conversion? | Monthly | |
| 48 | | Funnel | Where do they drop off? | Weekly | |
| 49 | | Cohort | How do segments perform over time? | Monthly | |
| 50 | |
| 51 | ### Attribution Models |
| 52 | |
| 53 | | Model | Credit Distribution | Best For | |
| 54 | |-------|-------------------|----------| |
| 55 | | Last Click | 100% to final touchpoint | Short cycles, direct response | |
| 56 | | First Click | 100% to first touchpoint | Brand awareness, TOFU | |
| 57 | | Linear | Equal across all | Understanding full journey | |
| 58 | | Time Decay | More to recent touches | Long sales cycles | |
| 59 | | Position-Based | 40/20/40 first-mid-last | Balanced view | |
| 60 | | Data-Driven | ML-based distribution | High volume, mature programs | |
| 61 | |
| 62 | ### Marketing KPIs by Funnel Stage |
| 63 | |
| 64 | **TOFU (Awareness)** |
| 65 | - Impressions, reach, traffic |
| 66 | - CPM, cost per visitor |
| 67 | - Brand search volume |
| 68 | |
| 69 | **MOFU (Consideration)** |
| 70 | - Leads, MQLs, engagement |
| 71 | - CPL, cost per MQL |
| 72 | - Content downloads, webinar registrations |
| 73 | |
| 74 | **BOFU (Decision)** |
| 75 | - SQLs, opportunities, customers |
| 76 | - CAC, cost per opportunity |
| 77 | - Demo requests, trial signups |
| 78 | |
| 79 | **Retention** |
| 80 | - NPS, retention rate, churn |
| 81 | - LTV, expansion revenue |
| 82 | - Referrals, advocacy |
| 83 | |
| 84 | ## Best Practices |
| 85 | |
| 86 | ### Setup Excellence |
| 87 | 1. **UTM Discipline**: Consistent naming convention across all campaigns |
| 88 | 2. **Goal Hierarchy**: Primary conversions > secondary > micro-conversions |
| 89 | 3. **Cross-Domain Tracking**: Proper setup for checkout/payment flows |
| 90 | 4. **Event Taxonomy**: Clear naming for custom events |
| 91 | |
| 92 | ### Reporting Excellence |
| 93 | 1. **Context Always**: Never report numbers without comparison (vs target, vs previous) |
| 94 | 2. **Action-Oriented**: Every insight should suggest an action |
| 95 | 3. **Visualization**: Use appropriate chart types (trends=line, comparison=bar) |
| 96 | 4. **Segmentation**: Break down by meaningful dimensions |
| 97 | |
| 98 | ### Attribution Excellence |
| 99 | 1. **Window Matching**: Attribution window matches sales cycle |
| 100 | 2. **Model Selection**: Choose model based on marketing maturity |
| 101 | 3. **Multi-Touch Visibility**: Track full journey, not just last touch |
| 102 | 4. **Offline Integration**: Include phone, events, direct sales |
| 103 | |
| 104 | ## Agent Integration |
| 105 | |
| 106 | | Agent | How They Use This Skill | |
| 107 | |-------|------------------------| |
| 108 | | `researcher` | Compiling performance data, competitive benchmarks | |
| 109 | | `lead-qualifier` | Funnel conversion analysis, lead source quality | |
| 110 | | `planner` | Budget allocation based on channel ROI | |
| 111 | | `project-manager` | Campaign performance tracking | |
| 112 | |
| 113 | ## Anti-Patterns to Avoid |
| 114 | |
| 115 | | Anti-Pattern | Why It's Wrong | Do This Instead | |
| 116 | |--------------|----------------|-----------------| |
| 117 | | Vanity metrics only | Impressions ≠ impact | Focus on conversion metrics | |
| 118 | | Last-click bias | Ignores awareness touchpoints | Use multi-touch attribution | |
| 119 | | No control groups | Can't prove causation | A/B test when possible | |
| 120 | | Siloed data | Missing full picture | Integrate CRM + analytics | |
| 121 | | Report without ac |