borghei/Claude-Skills / marketing/conversational-ads
conversational-ads
VerifiedPlan, write and measure ads inside AI assistants and AI search (ChatGPT Ads, Google AI Overviews and AI Mode, Microsoft Copilot). Use when testing ChatGPT ads, answer-adjacent copy, or incrementality for AI placements.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 13.3 KB |
| assets/sample_ad_copy.json | config | 1.9 KB |
| assets/sample_plan_input.json | config | 1.2 KB |
| assets/test-plan-template.md | markdown | 2.5 KB |
| references/anti-patterns.md | markdown | 3.3 KB |
| references/conversational-ads-playbook.md | markdown | 7.6 KB |
| references/platform-specs.md | markdown | 9.6 KB |
| scripts/answer_adjacent_copy_linter.py | script | 10.2 KB |
| scripts/conversational_ad_planner.py | script | 12.1 KB |
assets/test-plan-template.md
View on GitHub# Conversational Ads Test Plan — [Product]
**Owner:** [name] · **Dates:** [start] → [readout] · **Platform facts verified on:** [date, links re-checked]
## 1. Hypothesis
Adding [ChatGPT Ads / broad-match + AI Max / PMax for Copilot] in [markets] will deliver incremental conversions at an incremental CPA ≤ [target] within [N] weeks.
## 2. Prompts in scope
| # | Prompt (as customers phrase it) | Stage | Weight | Sensitive? |
|---|----------------------------------|-------|--------|-----------|
| 1 | | research / compare / purchase / support | | yes / no |
## 3. Eligibility (from `conversational_ad_planner.py`)
| Platform | Eligible | Buy mechanism | Notes |
|----------|----------|---------------|-------|
| ChatGPT Ads | | direct | |
| Google AI Overviews / AI Mode | | via Search broad / AI Max / Shopping / PMax | no opt-out, no segment reporting |
| Microsoft Copilot | | via PMax / Search with logo / Shopping | no opt-out, no Copilot metrics |
## 4. Design
- Type: geo holdout / time-based on-off
- Treated regions: [list] · Holdout regions: [list] ([x]% of volume)
- Matching method: [past 8 weeks conversions, correlation ≥ 0.9]
- Freeze rules: no bid, budget or creative changes from week 3 to readout
## 5. Budget and power
| Arm | Test spend | Expected conversions | Readable (≥ 50)? |
|-----|-----------|----------------------|------------------|
## 6. KPIs
| KPI | Target | Source |
|-----|--------|--------|
| Incremental CPA | ≤ [target] | geo readout |
| Platform-reported CPA | ≤ [target × 1.3] in learning | platform |
| Brand-search lift | > 0 | search console / ads |
| CTR baseline | set in weeks 1-2 | platform |
## 7. Readiness checklist
- [ ] Server-side conversions live and deduplicated (OpenAI Conversions API / Google enhanced conversions / Microsoft UET)
- [ ] UTMs: `utm_source=<platform>&utm_medium=cpc&utm_campaign=<macro>`
- [ ] Product feed audited (attributes customers ask about, price, availability)
- [ ] Landing pages answer the top prompts in the first screen
- [ ] Logo + image assets uploaded (Copilot eligibility)
- [ ] Copy passed `answer_adjacent_copy_linter.py` (exit 0)
- [ ] AI-generated creative labelled where required
## 8. Readout (fill at end)
| Arm | Treated conv. | Holdout conv. (scaled) | Incremental | Incremental CPA | Decision |
|-----|---------------|------------------------|-------------|-----------------|----------|
Decision rule: scale if incremental CPA ≤ target; iterate inputs if within 30%; stop otherwise.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the work | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation code | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |