Setup: Export Your Data
Before using these prompts, export your campaign data:
- Google Ads: Reports → Custom → Download as CSV
- Meta Ads: Ads Manager → Export → CSV
- GA4: Explore → Export → CSV
Include: Campaign name, spend, impressions, clicks, conversions, revenue (if available), and dates. The more columns, the richer the analysis.
The 7 Prompts
1. Instant Performance Summary
Analyze this campaign data and give me:
1. Top 3 best performing campaigns by ROAS (or conversion rate if no revenue)
2. Top 3 worst performing campaigns that are still spending
3. Total spend, conversions, and blended CPA
4. One surprising insight I might have missed
Format as a brief executive summary I can share with my team.
Use case: Weekly check-ins, stakeholder updates, quick pulse checks.
2. Anomaly Detection
Review this data for anomalies and red flags:
1. Any campaigns with spend but zero conversions?
2. Any sudden CPA spikes (>50% above average)?
3. Any campaigns where CTR dropped significantly?
4. Any budget pacing issues (underspend or overspend)?
List each issue with the specific campaign name and recommended action.
Use case: Monday morning audits, catching problems before they burn budget.
3. Budget Reallocation Recommendations
Based on this performance data, recommend how I should reallocate my budget:
1. Which campaigns should get MORE budget? (high efficiency, room to scale)
2. Which campaigns should get LESS budget? (poor efficiency, diminishing returns)
3. Which campaigns should be paused entirely?
Assume my total budget stays the same. Show the % shift you recommend and expected impact.
Use case: Monthly budget planning, optimization cycles.
4. Creative Fatigue Analysis
Analyze this data for signs of creative fatigue:
1. Which campaigns show declining CTR over time?
2. Which campaigns show increasing CPA over time?
3. Which campaigns have high frequency but dropping engagement?
For each fatigued campaign, estimate how urgent the refresh is (critical/moderate/low).
Use case: Creative refresh planning, fighting ad blindness.
5. Day-of-Week & Hour Analysis
Break down performance by day of week (and hour if available):
1. Which days have the best conversion rate?
2. Which days have the worst CPA?
3. Are there any days where we're overspending relative to performance?
Recommend a day-parting strategy based on this data.
Use case: Ad scheduling optimization, reducing wasted weekend spend.
6. Audience/Segment Deep Dive
Segment this data by [audience/geo/device/age] and tell me:
1. Which segment has the highest ROAS?
2. Which segment has the lowest CPA?
3. Which segment is underinvested relative to its performance?
4. Which segment should we exclude or reduce?
Create a simple table comparing all segments.
Use case: Audience optimization, finding hidden high-performers.
7. Executive Summary Generator
Create a 1-page executive summary of this campaign data for my CMO:
- Overall performance vs. previous period (if data available)
- Key wins (what worked)
- Key challenges (what didn't)
- Top 3 recommended actions for next period
- One strategic insight
Keep it under 300 words. Use bullet points. No jargon.
Use case: Monthly reports, board updates, stakeholder communication.
Pro Tips for Better Results
- Be specific about metrics: "Analyze ROAS" beats "Analyze performance."
- Include date ranges: ChatGPT can spot trends if you include time-series data.
- Ask for tables: Add "Format as a table" for easy copy-paste to slides.
- Iterate: Ask follow-up questions. "Drill deeper into Campaign X" works great.
- Request visualizations: "Create a bar chart of spend by campaign" generates actual charts.
A Note on Data Privacy
ChatGPT Plus and Team don't train on your uploads by default. For extra safety: anonymize campaign names (Campaign_A, Campaign_B), remove any PII, and consider ChatGPT Enterprise for strict compliance requirements.
FAQ
Can ChatGPT analyze my Google Ads data?
Yes. Export your campaign data as CSV, upload it to ChatGPT (Plus or Team), and use specific prompts to get instant analysis of performance, anomalies, budget allocation, and optimization opportunities.
Is my data safe when I upload it to ChatGPT?
ChatGPT Plus and Team plans don't train on your data by default. For sensitive data, use ChatGPT Team or Enterprise which have explicit data privacy agreements. You can also anonymize campaign names before uploading.
What file formats does ChatGPT accept?
ChatGPT accepts CSV, Excel (.xlsx), PDF, and plain text. For best results with marketing data, use CSV exports from Google Ads, Meta Ads Manager, or Google Analytics.
