ad-spend-optimizer
About
This skill analyzes advertising performance data across platforms like Google Ads and Meta to recommend budget reallocations that maximize return on ad spend (ROAS) and minimize customer acquisition cost (CAC). Developers can use it for quarterly planning, diagnosing underperforming channels, and rebalancing the media mix after performance shifts. Its key capability is calculating marginal ROI to inform data-driven decisions on scaling or reducing spend.
Quick Install
Claude Code
Recommendednpx skills add guia-matthieu/clawfu-skills -a claude-code/plugin add https://github.com/guia-matthieu/clawfu-skillsgit clone https://github.com/guia-matthieu/clawfu-skills.git ~/.claude/skills/ad-spend-optimizerCopy and paste this command in Claude Code to install this skill
Documentation
Ad Spend Optimizer
Analyze paid advertising performance across channels and recommend budget reallocation to maximize ROAS and minimize CAC.
When to Use This Skill
- Quarterly budget planning — reallocate spend based on performance data
- Channel mix optimization — find the right balance across platforms
- Performance troubleshooting — diagnose why CAC is rising or ROAS declining
- Scaling decisions — determine if a channel has headroom to scale
- New channel testing — structure test budgets with clear success criteria
Methodology Foundation
| Aspect | Details |
|---|---|
| Source | Marginal ROI optimization + portfolio theory for marketing |
| Core Principle | Allocate each dollar where the marginal return is highest — shift spend from diminishing-returns channels to underspent ones |
| Framework | 70/20/10 — 70% proven channels, 20% optimization tests, 10% new channel experiments |
What Claude Does vs What You Decide
| Claude Does | You Decide |
|---|---|
| Calculates ROAS, CAC, and CPL per channel and campaign | Total budget constraints |
| Identifies diminishing returns and reallocation opportunities | Risk tolerance for new channels |
| Models projected outcomes for different allocation scenarios | Business priorities and brand considerations |
| Creates monitoring dashboards and alert thresholds | Platform selection and creative direction |
Instructions
Step 1: Audit Current Performance
Collect these metrics per channel and campaign:
| Metric | Formula | Healthy Range |
|---|---|---|
| ROAS | Revenue ÷ Ad Spend | >3:1 for most B2B/B2C |
| CAC | Ad Spend ÷ New Customers | <LTV ÷ 3 |
| CPL | Ad Spend ÷ Leads | Varies by industry |
| CTR | Clicks ÷ Impressions | >1% search, >0.5% social |
| Conv Rate | Conversions ÷ Clicks | >2% landing pages |
Validation checkpoint: If data is missing for any channel, flag it — incomplete data leads to wrong reallocations.
Step 2: Attribution Analysis
Choose the model that matches the business:
| Model | Best For | Trade-off |
|---|---|---|
| Last Click | Direct response, short cycles | Ignores awareness |
| First Click | Awareness campaigns | Ignores conversion assist |
| Linear | Balanced multi-touch view | Dilutes signal |
| Time Decay | Shorter sales cycles | Biases toward bottom-funnel |
| Position-Based | Balanced with emphasis | May miss mid-funnel |
| Data-Driven | Sophisticated, enough data | Requires volume |
Step 3: Calculate Marginal ROI
For each channel, answer: Where does the next $1 produce the most return?
| Signal | Meaning | Action |
|---|---|---|
| CAC well below target | Headroom to scale | Increase spend 50%, monitor weekly |
| CAC at target | Optimized | Maintain, test creative |
| CAC above target | Diminishing returns | Reduce spend, reallocate |
| Low volume, good CAC | Underinvested | Scale cautiously (2x) |
| High volume, rising CAC | Hitting ceiling | Cap spend, diversify |
Step 4: Model Reallocation Scenarios
Build 3 scenarios (conservative, moderate, aggressive) showing projected leads, CAC, and ROAS at each budget level. Include:
- Per-channel breakdowns with expected performance
- Warning thresholds — CAC levels that trigger spend cuts
- Implementation timeline — weekly changes, not all at once
Step 5: Implement and Monitor
Weekly monitoring checklist:
- Spend pacing vs. plan
- CAC by channel vs. target
- Lead volume vs. forecast
- Any channel crossing warning threshold?
Scaling rule: If CAC stays 15%+ below target for 2 consecutive weeks, increase spend by 25%. If CAC exceeds target for 2 weeks, reduce by 25%.
Examples
Example: B2B SaaS Budget Reallocation
Input: $100K/month — Google ($50K), Meta ($30K), LinkedIn ($15K), Other ($5K). Target: $200 CAC, 500 leads/month. Current: 395 leads, $253 CAC.
Diagnosis:
- Google Display ($15K → 30 leads, $500 CAC) — cut entirely
- Meta Lookalike ($15K → 85 leads, $176 CAC) — star performer, scale
- LinkedIn Lead Gen ($5K → 10 leads, $500 CAC) — cut
Proposed reallocation:
| Channel | Current | Proposed | Expected CAC |
|---|---|---|---|
| Google Ads | $50K | $35K | $206 |
| Meta | $30K | $50K | $196 |
| $15K | $8K | $286 | |
| Testing | $5K | $7K | Variable |
Projected result: 473 leads (+20%), $211 CAC (-17%).
Skill Boundaries
What This Skill Does Well
- Analyzing multi-channel ad performance from provided data
- Recommending budget shifts based on marginal ROI
- Modeling reallocation scenarios with projected outcomes
- Creating monitoring frameworks with alert thresholds
What This Skill Cannot Do
- Access ad platform accounts or pull live data
- Make real-time bid adjustments or campaign changes
- Evaluate creative quality (headlines, images, video)
- Account for brand lift or offline conversion effects
References
- Google Ads Optimization Guide
- Meta Business Suite Best Practices
- LinkedIn Marketing Solutions
- Common Thread Collective — ad spend allocation methodology
Related Skills
google-ads-expert— Google-specific campaign optimizationaarrr-metrics— Full funnel view beyond paid acquisitiongrowth-loops— Sustainable growth beyond paid channels
GitHub Repository
Related Skills
evaluating-llms-harness
TestingThis Claude Skill runs the lm-evaluation-harness to benchmark LLMs across 60+ standardized academic tasks like MMLU and GSM8K. It's designed for developers to compare model quality, track training progress, or report academic results. The tool supports various backends including HuggingFace and vLLM models.
cloudflare-cron-triggers
TestingThis skill provides comprehensive knowledge for implementing Cloudflare Cron Triggers to schedule Workers using cron expressions. It covers setting up periodic tasks, maintenance jobs, and automated workflows while handling common issues like invalid cron expressions and timezone problems. Developers can use it for configuring scheduled handlers, testing cron triggers, and integrating with Workflows and Green Compute.
webapp-testing
TestingThis Claude Skill provides a Playwright-based toolkit for testing local web applications through Python scripts. It enables frontend verification, UI debugging, screenshot capture, and log viewing while managing server lifecycles. Use it for browser automation tasks but run scripts directly rather than reading their source code to avoid context pollution.
finishing-a-development-branch
TestingThis skill helps developers complete finished work by verifying tests pass and then presenting structured integration options. It guides the workflow for merging, creating PRs, or cleaning up branches after implementation is done. Use it when your code is ready and tested to systematically finalize the development process.
