The core of Meta Ads Kit is a simple framework: 5 questions that replace your entire daily Ads Manager routine .
90% of ad management is pattern recognition. Spend trending up or down. CTR declining (creative fatigue). CPA spiking (audience exhaustion). Winners emerging. Losers bleeding.
You don’t need to stare at Ads Manager to spot these patterns. The AI agent does it and tells you what matters.
The Framework
Question 1 Am I on track? Today’s spend vs. pacing expectations
Question 2 What’s running? Active campaigns at a glance
Question 3 How’s performance? 7-day metrics by campaign
Question 4 Who’s winning/losing? Ad-level performance sorted
Question 5 Any fatigue? CTR trends, frequency, CPC movement
What Each Question Tells You 1. Am I on track? Compares today’s spend rate to your daily budget. If you’re pacing high or low, the system flags it immediately.
What you see:
Current spend today vs. expected at this hour
Percentage over/under pace
Whether you’ll exhaust budget before end of day or underspend
Why it matters: Catch overspend before you blow through your budget. Catch underspend before you miss opportunity.Example output: 🟡 Spend pacing 18% above target Current: $287 (as of 2:30pm) Expected: $243 at this time Projected EOD: $524 (budget: $450)
2. What’s running? Lists all active campaigns with their current status. Flags anything that should be off or paused.
What you see:
Campaign names and statuses
Daily budgets
Any campaigns that were supposed to stop but are still running
Why it matters: Sometimes campaigns stay active when they shouldn’t. Or an old test is still burning budget. This catches it.Example output: Active Campaigns (4): ✅ Summer Launch - Traffic ($150/day) ✅ Retargeting - Conversions ($200/day) ⚠️ Q1 Test Campaign ($50/day) — Note: This was scheduled to end 3/1 ✅ Brand Awareness ($75/day)
3. How’s the last 7 days? Campaign-level metrics for the previous week, compared to the week before if data is available.
What you see:
Spend, impressions, clicks, CTR
CPC, CPA (if applicable)
Week-over-week trends
Why it matters: Shows momentum. Is performance improving or declining? Are you getting more efficient or less?Example output: Last 7 Days Performance: Summer Launch - Traffic Spend: $1,050 (↑12% vs. prior week) CTR: 1.8% (↓0.2pp) CPC: $1.22 (↑$0.15) Retargeting - Conversions Spend: $1,400 CTR: 2.4% (↑0.3pp) CPA: $18.50 (target: $25) 🏆
4. Who’s winning and who’s losing? Ad-level performance sorted by efficiency. Top 3 winners ready to scale. Bottom 3 losers bleeding budget.
What you see:
Best performing ads by CTR and cost efficiency
Worst performing ads with high spend + low results
Specific ad IDs and creative names
Why it matters: This is where the money decisions happen. Which ads deserve more budget? Which need to be paused?Example output: 🏆 Winners (Top 3): 1. Ad #23847111 "Notes App Hero" CTR: 3.2% | CPC: $0.87 | Spend: $124 → Scale candidate 2. Ad #23847112 "Kitchen Before/After" CTR: 2.9% | CPC: $0.95 | Spend: $98 3. Ad #23847113 "Testimonial Video" CTR: 2.7% | CPC: $1.10 | Spend: $156 🩸 Bleeders (Bottom 3): 1. Ad #23847120 "Generic Product Shot" CTR: 0.6% | CPC: $3.20 | Spend: $187 Frequency: 4.2 — audience exhausted → Pause candidate 2. Ad #23847121 "Headline Only" CTR: 0.8% | CPC: $2.80 | Spend: $142
Bleeders are defined as: Ads with CTR < 1.0% OR frequency > 3.5, combined with significant spend (typically >$10 in the period). These are actively hurting your ROAS.
5. Any fatigue signals? Daily breakdown to spot creative fatigue before it kills your performance . Tracks CTR declining day-over-day, frequency climbing, CPC rising.
What you see:
CTR trend over the past 7 days (per ad)
Frequency progression
CPC movement
Specific days where fatigue started
Why it matters: Creative fatigue is the silent killer. CTR drops 0.1% per day, frequency ticks up, CPC quietly inflates — and by the time you notice manually, you’ve burned budget. This catches it early.Example output: 😴 Fatigue Detected (2 ads): Ad #23847111 "Notes App Hero" CTR trend (7d): 3.2% → 3.0% → 2.8% → 2.5% → 2.2% 🔴 → 31% decline from peak Frequency: 2.1 → 2.8 → 3.4 → 3.9 🟡 Recommendation: Rotate creative this week Ad #23847115 "Receipt Close-up" CPC trend: $1.10 → $1.25 → $1.55 → $1.82 🟡 → 65% inflation over 4 days Recommendation: Pause or refresh copy
Fatigue thresholds (configurable in ad-config.json):
CTR dropping >20% over 3 days = 🔴 Critical
Frequency > 3.5 = 🟡 Warning
CPC rising >15% over 3 days = 🟡 Warning
Why This Replaces Ads Manager Here’s what you typically do in Ads Manager:
Log in (1 min)
Navigate to account and date range (30 sec)
Scroll through campaigns (2 min)
Check today’s spend vs. yesterday (1 min)
Sort ads by performance (2 min)
Identify what’s working, what’s not (5 min)
Look for creative fatigue patterns (3 min)
Check frequency caps and delivery (2 min)
Export data to analyze elsewhere (3 min)
Decide on actions (2 min)
Total: ~20 minutes of clicking, context-switching, and mental overhead. With the 5 Daily Questions:
Run: openclaw start → “Daily ads check”
Read the summary (2 min)
Approve/reject recommendations
Total: 2 minutes over coffee. The agent watches the patterns. You make the calls.
Running the Daily Check You can trigger the 5 questions in multiple ways:
Via OpenClaw Agent openclaw start # Then message: "Daily ads check" # or "How are my Meta ads doing?" # or "Run the 5 questions on my ads"
The agent orchestrates all 5 questions, interprets the data, and presents findings with recommendations.
Via Script (Direct) cd meta-ads-kit ./skills/meta-ads/scripts/meta-ads.sh daily-check
Runs all 5 reports and outputs raw data. Good for piping into other tools or logging.
Automated (Cron) Set up a daily cron job to run automatically every morning:
# In OpenClaw: "Run my daily ads check every morning at 8am and send me the summary"
The agent will run the check, analyze results, and message you with findings + recommendations. You approve from your phone.
Customizing Your Benchmarks The 5 questions use your configured benchmarks to determine what’s “good” vs. “bad”. Edit ad-config.json:
{ "benchmarks" : { "target_cpa" : 25.00 , "target_roas" : 3.0 , "max_frequency" : 3.5 , "min_ctr" : 1.0 , "max_cpc" : 2.50 }, "alerts" : { "bleeder_ctr_threshold" : 1.0 , "bleeder_frequency_threshold" : 3.5 , "fatigue_ctr_drop_pct" : 20 , "spend_pace_alert_pct" : 15 } }
Or just tell the OpenClaw agent your benchmarks conversationally — it’ll figure it out and update the config.
What Happens After the Questions The 5 questions diagnose . The skills act .
Once you’ve identified:
This is the full loop. See Workflow for how it all chains together.
Next Steps
Skills Overview Learn about the 5 skills that act on these insights
Workflow See the full loop from monitoring to creative refresh
Setup Guide Get your ad account connected and run your first check
Configuration Customize benchmarks and thresholds
