When your CPA spikes, Claude breaks down exactly what caused it. It looks across your campaign data and isolates the contributing factors — audience fatigue, bid landscape shifts, creative decay, landing page conversion drops, budget distribution changes, or new competitor activity. Platform: Google and Meta.
git clone https://github.com/irinabuht12-oss/marketing-skills.git--- name: cpa-diagnostics description: When your CPA spikes, Claude breaks down exactly what caused it. It looks across your campaign data and isolates the contributing factors — audience fatigue, bid landscape shifts, creative decay, landing page conversion drops, budget distribution changes, or new competitor activity. Platform: Google and Meta. metadata: platform: Google and Meta --- # 1/ CPA Diagnostics — Google + Meta ## What it does When your CPA spikes, Claude breaks down exactly what caused it. It looks across your campaign data and isolates the contributing factors — audience fatigue, bid landscape shifts, creative decay, landing page conversion drops, budget distribution changes, or new competitor activity. ## How it works Feed Claude your campaign data for the current period vs the previous period. It compares metrics at every level — campaign, ad set, ad, keyword, audience, placement — and identifies where the variance is coming from. Results are ranked by impact so you know what to fix first. ## Practical example CPA went from $42 to $67 overnight. Claude identifies that 73% of the increase came from one ad set where frequency hit 4.2 and CTR dropped 31%. Another 18% came from a search campaign where a broad match keyword started matching irrelevant queries. The remaining 9% was a landing page with slower load time after a site update. ## What you get back - Ranked list of factors driving the CPA change with percentage contribution - Specific campaigns, ad sets, keywords, or audiences causing the issue - Recommended fixes for each factor in priority order - Comparison data showing exactly what shifted between periods ## When to use it - CPA jumps more than 15-20% day over day or week over week - Client asks "why did performance drop" and you need an answer fast - Monday morning account checks when you need to catch weekend issues - Before budget calls to identify and fix problems proactively ## Data access (Ryze MCP) This skill works best with live account data. Connect the free Ryze MCP once and Claude reads your Google Ads, Meta Ads, GA4 and Search Console directly: - claude.ai / Claude Desktop: Settings → Connectors → Add custom connector → `https://connector.get-ryze.ai/mcp` - Claude Code: `claude mcp add ryze --transport http https://connector.get-ryze.ai/mcp` - Cursor: Settings → MCP → add the same URL Setup guide: https://www.get-ryze.ai/how-to-connect-claude-to-google-meta-ads-mcp
[{"step":"Identify the campaign and timeframe. Gather platform-specific data (e.g., Meta Ads Manager or Google Ads UI) for the period before and after the CPA spike.","tip":"Export raw data (e.g., cost, conversions, CPA, frequency, CTR) for the last 14 days to ensure accuracy. Use platform filters to isolate the campaign in question."},{"step":"Run the prompt template, replacing [PLACEHOLDERS] with your campaign name, platform, and time period. Include any additional context (e.g., recent creative changes or budget shifts).","tip":"For deeper analysis, add: 'Also compare performance against industry benchmarks for [INDUSTRY] in [REGION].' This helps contextualize the spike."},{"step":"Review the AI's breakdown of contributing factors. Cross-reference with your own notes (e.g., recent A/B tests, competitor launches) to validate findings.","tip":"Use the 'Top Contributing Factors' section to prioritize fixes. Focus on high-impact, low-effort changes first (e.g., pausing underperforming ad sets)."},{"step":"Implement the suggested actions in your ad platform. Set up alerts for key metrics (e.g., CPA, frequency) to catch future spikes early.","tip":"For Meta Ads, use the 'Advantage Campaign Budget' feature to auto-allocate budget away from underperforming ad sets. For Google Ads, adjust bids via 'Bid Simulator' before making changes."},{"step":"Monitor results daily for 3-5 days. Re-run the analysis if the CPA doesn’t improve to identify missed factors or new issues.","tip":"Use Google Sheets or Meta Ads' 'Custom Reports' to track metrics like CPA, CTR, and frequency over time. Set up automated alerts for CPA thresholds."}]
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/irinabuht12-oss/marketing-skills/tree/main/skills/cpa-diagnosticsCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
Analyze the recent CPA spike for [CAMPAIGN_NAME] in [PLATFORM: Google Ads/Meta Ads] during [TIME_PERIOD: e.g., last 7 days]. Break down the root causes by evaluating: audience fatigue metrics, bid landscape shifts, creative performance decay, landing page conversion rates, budget distribution changes, and competitor activity. Rank the factors by impact and suggest immediate corrective actions for each. Include a prioritized action plan for this week.
### CPA Spike Analysis for 'Summer Sale' Campaign (Meta Ads) - July 15-22, 2024 **Key Findings:** - **CPA Increased by 42%** (from $12.45 to $17.58) compared to the prior week. - **Top Contributing Factors:** 1. **Audience Fatigue (35% impact):** Frequency rose from 3.2 to 5.1, with CTR dropping 18% in the last 3 days. The 'Past Purchasers' segment (30% of spend) saw a 22% CPA increase despite no bid changes. 2. **Creative Decay (28% impact):** The top-performing video ad (15-second product demo) had a 30% drop in engagement after 14 days, with CPM increasing 25% due to audience saturation. 3. **Bid Landscape Shift (20% impact):** Competitor bids for 'summer sale' keywords rose 12% in the last 5 days, forcing our bids up by 8% to maintain position. Our average CPC increased from $1.45 to $1.72. 4. **Landing Page Conversion Drop (12% impact):** Post-click conversion rate fell from 4.2% to 2.9% due to a broken mobile optimization (discovered via heatmaps). 5. **Budget Reallocation (5% impact):** 15% of budget was shifted to a new 'Lookalike Audience' test, which underperformed (CPA: $22.10). **Immediate Actions (This Week):** 1. **Pause 2 underperforming ad sets** targeting 'Past Purchasers' and 'Lookalike Audience' (combined spend: $1,200/week). Replace with fresh creatives and broader lookalike audiences. 2. **Refresh top-performing video ad** with a new 10-second teaser and updated product showcase. Budget: $800 for new creative production. 3. **Adjust bids downward by 10%** for high-CPA keywords (e.g., 'summer sale discount') to reduce CPA while maintaining volume. Monitor impression share closely. 4. **Fix landing page issues** by reverting to the previous mobile-optimized version and adding a sticky CTA button. Expected lift: +1.5% conversion rate. 5. **Run a competitor analysis** to identify new entrants in the 'summer sale' space. Flag any aggressive new players for bid adjustments. **Expected Outcome:** CPA should stabilize at ~$15.20 by next Monday if all actions are implemented. Monitor daily and adjust bids based on real-time performance.
skills-collection
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan