wallacedobbs428/ads-scrape

Ad Scrape — Creative Deep Analysis

Deep-analyze competitor ad creative for: $ARGUMENTS

Your Mission

You are The Call Taker's ad creative analyst. Take the competitor data from /ads-research and break down exactly WHY the winning ads work — hook by hook, line by line.

Context

The Call Taker is an AI receptionist SaaS. $97/$297/$497/mo. Demo line: (615) 784-5747. 14-day free pilot. NEVER use "AI" in ad copy — use "24/7 receptionist", "smart receptionist", "virtual receptionist."

Step 1: Load Research Data

Read the research file:

~/thecalltaker-ops/ads/research/{vertical}-competitors.json

If it doesn't exist, tell the user to run /ads-research {vertical} first.

Step 2: Scrape Each Competitor's Ads

For each competitor from the research:

• Use WebSearch/WebFetch to find their Facebook Ad Library page and landing pages
• Search: "site:facebook.com/ads/library {advertiser name}" and their website
• Extract from each active ad:
  • Full primary text (the main copy)
  • Headline
  • Description
  • CTA button text
  • Format (image/video/carousel)
  • Link destination (landing page URL)
  • Estimated run duration

Step 3: Classify Every Ad

For each ad, classify:

Hook Type (first line of copy):

• Stat hook: "87% of callers hang up if no one answers"
• Question hook: "How many calls did you miss last week?"
• Pain hook: "Every missed call is money walking to your competitor"
• Story hook: "Last month, a plumber in Nashville..."
• Direct hook: "Your phone rings. Nobody answers. You just lost $500."

Body Structure:

• Problem → Solution → CTA
• Social proof → Benefit → CTA
• Pain → Agitate → Solve → CTA
• Benefit stack → Proof → CTA

Offer: Free trial/demo, specific price, ROI guarantee, no commitment

CTA Strength (1-5):

• 1 = "Learn More" (weak)
• 3 = "Start Free Trial" (medium)
• 5 = "Get Your AI Receptionist Now" (strong + specific)

Step 4: Identify the #1 Performing Angle

Based on longevity (longest running = most profitable):

• Which hook type appears most in 30+ day ads?
• Which body structure converts best?
• Which offer type has the longest survival?
• Which CTA gets used by the top spenders?

Step 5: Save Analysis

Save to:

~/thecalltaker-ops/ads/scrape/{vertical}-ad-analysis.json

JSON structure:

{
  "vertical": "{vertical}",
  "analyzed_at": "ISO timestamp",
  "total_ads_analyzed": 0,
  "ads": [
    {
      "advertiser": "",
      "primary_text": "",
      "headline": "",
      "description": "",
      "cta_button": "",
      "format": "",
      "landing_page": "",
      "days_running": 0,
      "hook_type": "",
      "body_structure": "",
      "offer": "",
      "cta_strength": 0
    }
  ],
  "winning_hook_type": "",
  "winning_body_structure": "",
  "winning_offer": "",
  "top_3_hooks": ["", "", ""],
  "burned_angles": [],
  "opportunity_gaps": []
}

Update ~/thecalltaker-ops/ads/intelligence.json with winning/burned angles.

Create directories if needed: mkdir -p ~/thecalltaker-ops/ads/scrape

Output to User

Print:

1. "Analyzed X ads from Y competitors in {vertical}"
2. "The winning angle: {hook_type} + {body_structure} + {offer}"
3. "Top 3 hooks that have run the longest:" (with actual copy)
4. "Burned angles to AVOID:" (short-lived ads = losers)
5. "Opportunity gaps The Call Taker can exploit:"
6. "Saved to ~/thecalltaker-ops/ads/scrape/{vertical}-ad-analysis.json"
7. "Next step: /ads-brief {vertical}"