Create High-Converting Meta Ad Scripts with GPT & Gemini Using Performance Data

Created by

Robert Breen

Last update

Last update 3 hours ago

🛠️ Step-by-Step Setup & Execution

1️⃣ Run Workflow (Manual Trigger)

Node	Type	Purpose
`Run Workflow`	Manual Trigger	Start the workflow on demand while you test.

2️⃣ Set Your Topic

Node	Type	How to configure
`Set Topic`	Set	Add a string field Topic – e.g. `"n8n"`. This keyword drives every subsequent step.

3️⃣ Search Google (Results 1-10)

Node	Type	API Credential
`Search Google (top 10)`	SerpAPI	Create SerpAPI credential<br>1. Sign up → copy API key → n8n → Credentials → New → SerpAPI → paste.<br>2. Select the credential in this node.
Key Params
`q`		`={{ $json.Topic }} Expert`
`location`		Region code (ex `585069efee19ad271e9c9b36`)
`additionalFields.start`		`"10"` (Google position 1-10)

4️⃣ Search Google (Results 11-20)

Node	Type	Notes
`Search Google (11-20)`	SerpAPI (same credential)	Remove `start` or set to 20+ to fetch next page.

5️⃣ Extract URL Lists

Node	Type	Script Purpose
`Extract Url` & `Extract Url 2`	Code	Loop `data.organic_results` → output `{ title, link, displayed_link }` for each result.

6️⃣ Combine Both Result Sets

Node	Type	Details
`Append Results`	Merge (combineAll)	Merges arrays from steps 3 & 4 into a single list for processing.

7️⃣ Loop Over Every URL

Node	Type	Configuration
`Loop Over Items1`	Split In Batches	Default batch = 1 (process one page at a time).<br>`onError = continueRegularOutput` keeps loop alive on failures.

8️⃣ Scrape Webpage Content (Apify)

Node	Type	API Credential
`Scrape URL with apify`	HTTP Request	Create Apify credential<br>1. Sign up at https://console.apify.com<br>2. Account → API tokens → copy.<br>3. n8n → Credentials → New → HTTP Query Auth → set query param `token=YOUR_TOKEN`.
Request Details
Method	POST
URL	`https://api.apify.com/v2/acts/6sigmag~fast-website-content-crawler/run-sync-get-dataset-items`
JSON Body	`json { "startUrls": ["{{ $json.link }}"] }`

9️⃣ Extract Email with OpenAI

Node	Type	API Credential
`Extract Email from webpage`	LangChain Agent	Create OpenAI credential<br>1. Generate key at https://platform.openai.com/account/api-keys<br>2. n8n → Credentials → New → OpenAI API → paste key.
Prompt (system)	`extract the email address from the text. if there is no email address, output null.`
Output Parser	`Structured Output Parser2` expects → `{ "email": "address OR null" }`

🔟 Loop Continues & Final Data

The extracted result returns to Loop Over Items1 until every URL is processed.

Typical final item JSON:

{
  "title": "How to Build n8n Workflows",
  "link": "https://example.com",
  "email": "[email protected]"
}

💡 Optional Enhancements
Idea How
Save Leads Add a Google Sheets or Airtable node after the loop.
Validate Emails Chain a ZeroBounce / Hunter.io verification API before saving.
Parallel Crawling Increase SplitInBatches size (watch Apify rate limits).

🙋‍♂️ Need More Help?
Robert Breen – Automation Consultant & n8n Expert
📧 [email protected]
🔗 https://www.linkedin.com/in/robert-breen-29429625/
🌐 https://ynteractive.com