Générer automatiquement des publicités Google avec Claude 3.5, Channable et Relevance AI
Ceci est unContent Creation, Multimodal AIworkflow d'automatisation du domainecontenant 19 nœuds.Utilise principalement des nœuds comme If, Set, Code, Slack, Aggregate. Générer automatiquement des publicités Google avec Claude 3.5, Channable et Relevance AI
- •Token Bot Slack ou URL Webhook
- •Peut nécessiter les informations d'identification d'authentification de l'API cible
- •Informations d'identification Google Sheets API
Nœuds utilisés (19)
Catégorie
{
"meta": {
"instanceId": "04fd795d32aabb18b913b4a3350b5cd0e9313a422ea0e7bdac0da2fb76cac9f7"
},
"nodes": [
{
"id": "b07b40c1-4c8c-4c35-aa07-be7025ed68b2",
"name": "Obtenir le flux produits",
"type": "n8n-nodes-base.httpRequest",
"notes": "CORRECTED: Uses v1 API if available, or replace with your e-commerce platform feed endpoint",
"position": [
-1380,
820
],
"parameters": {
"url": "={{$env.CHANNABLE_API_URL}}/v1/projects/{{$env.PROJECT_ID}}/items",
"options": {},
"authentication": "genericCredentialType",
"genericAuthType": "httpHeaderAuth"
},
"credentials": {
"httpHeaderAuth": {
"id": "luTyM3gZSgwhM8i0",
"name": "Header Auth account"
}
},
"typeVersion": 4.2
},
{
"id": "4b90a3bc-da44-44bc-b19a-d81427545f26",
"name": "Formater pour CSV",
"type": "n8n-nodes-base.set",
"notes": "Formats compliant ads into CSV structure",
"position": [
560,
700
],
"parameters": {
"options": {}
},
"typeVersion": 3.3
},
{
"id": "bea6c2f0-4a95-4848-85a1-780cc048b26d",
"name": "Alerte - Non conforme",
"type": "n8n-nodes-base.slack",
"notes": "Sends Slack alert for non-compliant ads. Can replace with Email node.",
"position": [
560,
900
],
"webhookId": "50d6fb83-2c83-4b57-80f9-5bb46d6b9fdf",
"parameters": {
"text": "=⚠️ Non-Compliant Ad Flagged\n\n*Product ID:* {{$json.product_id}}\n*Product Title:* {{$json.product_title}}\n*Category:* {{$json.category}}\n\n*Generated Headline:* {{$json.validated_headline}}\n*Generated Description:* {{$json.validated_description}}\n\n*Compliance Issues:* Check agent output\n\n*Timestamp:* {{$json.validation_timestamp}}",
"otherOptions": {}
},
"typeVersion": 2.1
},
{
"id": "95f53395-e93d-4309-8f98-d2b2378cbb13",
"name": "Agréger les lots",
"type": "n8n-nodes-base.aggregate",
"notes": "Combines all processed batches into single dataset",
"position": [
760,
700
],
"parameters": {
"options": {},
"aggregate": "aggregateAllItemData"
},
"typeVersion": 1
},
{
"id": "56492a38-1188-4b35-999f-14017ac76484",
"name": "Enregistrer dans Google Sheets",
"type": "n8n-nodes-base.googleSheets",
"notes": "CORRECTED: Saves CSV data to Google Sheets instead of Channable API upload (which doesn't exist). You can then: 1) Import manually to Google Ads, 2) Use Channable scheduled import, or 3) Build Google Ads API direct upload",
"position": [
1160,
700
],
"parameters": {
"options": {},
"sheetName": {
"__rl": true,
"mode": "name",
"value": "Generated Ads"
},
"documentId": {
"__rl": true,
"mode": "id",
"value": "={{$env.GOOGLE_SHEET_ID}}"
}
},
"credentials": {
"googleSheetsOAuth2Api": {
"id": "0xHwEMloLvs9YH5S",
"name": "Google Sheets account"
}
},
"typeVersion": 4.4
},
{
"id": "f25d680c-28ad-43f5-96c8-0f022cf1a710",
"name": "Déclencheur planifié - Quotidien1",
"type": "n8n-nodes-base.scheduleTrigger",
"notes": "Triggers workflow daily at midnight. Can also be triggered manually.",
"position": [
-1600,
820
],
"parameters": {
"rule": {
"interval": [
{
"field": "cronExpression",
"expression": "0 0 * * *"
}
]
}
},
"typeVersion": 1
},
{
"id": "01a193ae-a04f-4a04-9c3e-0a40c8e197c5",
"name": "Diviser en lots1",
"type": "n8n-nodes-base.splitInBatches",
"notes": "Processes 50 products at a time to avoid API rate limits",
"position": [
-1140,
820
],
"parameters": {
"options": {},
"batchSize": 50
},
"typeVersion": 3
},
{
"id": "50792985-2a29-4aeb-8191-2a1ef3d7ec97",
"name": "Générer texte publicitaire - Relevance AI1",
"type": "n8n-nodes-base.httpRequest",
"notes": "CORRECTED: Uses /trigger endpoint with tool ID from environment variable. Get your actual tool ID after creating in Relevance AI.",
"position": [
-880,
800
],
"parameters": {
"url": "={{$env.RELEVANCE_AI_API_URL}}/tools/google_text_ad_copy_generator/run",
"method": "POST",
"options": {
"timeout": 60000
},
"jsonBody": "={\n \"params\": {\n \"product_title\": \"{{$json.title}}\",\n \"product_description\": \"{{$json.description}}\",\n \"price\": \"{{$json.price}}\",\n \"category\": \"{{$json.category}}\",\n \"brand\": \"{{$json.brand}}\"\n }\n}",
"sendBody": true,
"specifyBody": "json",
"authentication": "genericCredentialType",
"genericAuthType": "httpHeaderAuth"
},
"credentials": {
"httpHeaderAuth": {
"id": "luTyM3gZSgwhM8i0",
"name": "Header Auth account"
}
},
"typeVersion": 4.2
},
{
"id": "17d67f83-8300-41c5-a3ac-9b7ea4bc25c2",
"name": "Valider les limites de caractères1",
"type": "n8n-nodes-base.code",
"notes": "Validates headline ≤30 chars and description ≤90 chars. This is the fix from the YouTube video for accurate character counting.",
"position": [
-560,
800
],
"parameters": {
"jsCode": "// Character limit validation with accurate counting\n\nconst headline = $input.item.json.headline || '';\nconst description = $input.item.json.description || '';\n\n// Function to accurately count displayed characters\nfunction getCharacterCount(str) {\n str = str.trim();\n return str.length;\n}\n\n// Validate headline (max 30 characters)\nlet headlineResult = {};\nconst headlineCount = getCharacterCount(headline);\n\nif (headlineCount > 30) {\n let truncated = headline.substring(0, 27).trim();\n truncated = truncated.replace(/[,.\\-:]$/, '');\n headlineResult = {\n headline: truncated + '...',\n char_count: truncated.length + 3,\n truncated: true,\n original: headline\n };\n} else {\n headlineResult = {\n headline: headline,\n char_count: headlineCount,\n truncated: false,\n original: headline\n };\n}\n\n// Validate description (max 90 characters)\nlet descriptionResult = {};\nconst descriptionCount = getCharacterCount(description);\n\nif (descriptionCount > 90) {\n let truncated = description.substring(0, 87).trim();\n truncated = truncated.replace(/[,.\\-:]$/, '');\n descriptionResult = {\n description: truncated + '...',\n char_count: truncated.length + 3,\n truncated: true,\n original: description\n };\n} else {\n descriptionResult = {\n description: description,\n char_count: descriptionCount,\n truncated: false,\n original: description\n };\n}\n\n// Return validated data with original product info\nreturn {\n product_id: $input.item.json.product_id || $input.item.json.id,\n product_title: $input.item.json.product_title || $input.item.json.title,\n validated_headline: headlineResult.headline,\n headline_char_count: headlineResult.char_count,\n headline_truncated: headlineResult.truncated,\n validated_description: descriptionResult.description,\n description_char_count: descriptionResult.char_count,\n description_truncated: descriptionResult.truncated,\n category: $input.item.json.category,\n final_url: $input.item.json.product_url || $input.item.json.link,\n validation_timestamp: new Date().toISOString()\n};"
},
"typeVersion": 2
},
{
"id": "534ea658-ea94-4d4c-a4e3-5d5fa1c699b6",
"name": "Agent de vérification de conformité1",
"type": "n8n-nodes-base.httpRequest",
"notes": "CORRECTED: Uses /agents/trigger with agent_id in body. Get your actual agent ID from Relevance AI.",
"position": [
-160,
800
],
"parameters": {
"url": "={{$env.RELEVANCE_AI_API_URL}}/agents/google_ads_compliance_checker/run",
"method": "POST",
"options": {
"timeout": 60000
},
"jsonBody": "={\n \"message\": {\n \"role\": \"user\",\n \"content\": \"Check compliance for this ad: Headline: {{$json.validated_headline}}, Description: {{$json.validated_description}}, Category: {{$json.category}}\"\n },\n \"agent_id\": \"{{$env.RELEVANCE_AGENT_COMPLIANCE_ID}}\"\n}",
"sendBody": true,
"specifyBody": "json",
"authentication": "genericCredentialType",
"genericAuthType": "httpHeaderAuth"
},
"credentials": {
"httpHeaderAuth": {
"id": "luTyM3gZSgwhM8i0",
"name": "Header Auth account"
}
},
"typeVersion": 4.2
},
{
"id": "c9bfe847-7215-4ae5-a61e-4eb24ae1da0c",
"name": "SI conforme1",
"type": "n8n-nodes-base.if",
"notes": "Routes ads based on compliance status. APPROVED ads go to formatting, others trigger alert.",
"position": [
160,
800
],
"parameters": {
"options": {},
"conditions": {
"string": [
{
"value1": "={{$json.compliance_status || $json.output}}",
"value2": "APPROVED",
"operation": "contains"
}
]
}
},
"typeVersion": 2
},
{
"id": "5a27e0f1-d4e7-4a81-9700-dd1c368fbb0b",
"name": "Générer fichier CSV1",
"type": "n8n-nodes-base.code",
"notes": "Converts JSON to properly escaped CSV format",
"position": [
960,
700
],
"parameters": {
"jsCode": "// Generate CSV from aggregated data\n\nconst items = $input.all();\n\n// CSV headers\nconst headers = ['product_id', 'headline', 'description', 'final_url', 'display_url'];\n\n// Create CSV rows\nconst rows = items.map(item => {\n const json = item.json;\n return [\n json.product_id || '',\n json.headline || '',\n json.description || '',\n json.final_url || '',\n json.display_url || ''\n ];\n});\n\n// Convert to CSV string\nconst csvRows = [headers, ...rows];\nconst csvString = csvRows\n .map(row => row.map(cell => `\"${String(cell).replace(/\"/g, '\"\"')}\"`).join(','))\n .join('\\n');\n\n// Return CSV data\nreturn {\n csv_data: csvString,\n total_ads: rows.length,\n generated_at: new Date().toISOString(),\n filename: `google_ads_${new Date().toISOString().split('T')[0]}.csv`\n};"
},
"typeVersion": 2
},
{
"id": "55d4429b-b4a9-4a19-bb34-2f58d7881a5c",
"name": "Notification de succès1",
"type": "n8n-nodes-base.slack",
"notes": "Sends success notification with summary",
"position": [
1360,
700
],
"webhookId": "04afb760-a7e8-4d2d-9827-1ed01108405d",
"parameters": {
"text": "=✅ *Google Ads Generation Complete*\n\n📊 *Summary:*\n• Total Ads Generated: {{$node['Generate CSV File1'].json.total_ads}}\n• Saved to Google Sheets: {{$env.GOOGLE_SHEET_ID}}\n• Timestamp: {{$node['Generate CSV File1'].json.generated_at}}\n\n🎯 *Next Steps:*\n• Review ads in Google Sheet\n• Import to Google Ads (manual or scheduled)\n• Monitor for disapprovals in 24 hours\n\n💡 CSV filename: {{$node['Generate CSV File1'].json.filename}}",
"otherOptions": {}
},
"typeVersion": 2.1
},
{
"id": "9af47779-8ca3-4e7c-9d3d-76f419c45361",
"name": "Note adhésive",
"type": "n8n-nodes-base.stickyNote",
"position": [
-2060,
760
],
"parameters": {
"width": 360,
"height": 220,
"content": "## 🟦 Schedule Trigger - Daily\n\n### 🕓 Purpose: Automatically runs every night at midnight (0 0 * * *).\nTip: You can also execute manually for testing.\n\n💡 Use to refresh and generate new ad copy daily."
},
"typeVersion": 1
},
{
"id": "d563b4b9-7239-4bb1-ab62-73b924643715",
"name": "Note adhésive1",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1500,
400
],
"parameters": {
"width": 340,
"height": 320,
"content": "## 🟦 Get Product Feed (Channable)\n### 📦 Purpose: Fetches live product data (title, price, brand, description, category).\nAPI: GET {{$env.CHANNABLE_API_URL}}/v1/projects/{{$env.PROJECT_ID}}/items\n\n#### ⚙️ Replace or connect to your store’s API if Channable isn’t used.\n✅ Returns all active products for ad generation."
},
"typeVersion": 1
},
{
"id": "538ff6d6-5496-42db-90c4-2f22a1f228f4",
"name": "Note adhésive2",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1240,
980
],
"parameters": {
"width": 320,
"height": 240,
"content": "## 🟦 Split Into Batches\n\n### 🧮 Purpose: Processes up to 50 products per batch.\n\nPrevents hitting API rate limits from Relevance AI or Channable.\n⚡ Adjust batch size depending on product volume and plan limits."
},
"typeVersion": 1
},
{
"id": "bb106886-79a8-437c-8f74-50939befe720",
"name": "Note adhésive4",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1020,
400
],
"parameters": {
"width": 340,
"height": 320,
"content": "## 🟦 Generate Ad Copy - Relevance AI\n### ✍️ Purpose: Calls your Relevance AI Tool to generate Google Ads headlines & descriptions.\nAPI: POST /tools/google_text_ad_copy_generator/run\n\nUses product title, description, price, brand, and category.\n🧠 Model: Claude 3.5 / GPT-4 via Relevance AI."
},
"typeVersion": 1
},
{
"id": "eabe3ce0-d216-426d-8cc3-e1411f36c385",
"name": "Note adhésive3",
"type": "n8n-nodes-base.stickyNote",
"position": [
-680,
980
],
"parameters": {
"width": 320,
"height": 300,
"content": "## 🟦 Validate Character Limits\n\n🔍 Purpose: Ensures headlines ≤30 chars and descriptions ≤90 chars.\n\nUses JavaScript to count accurately and truncate gracefully.\n✨ Automatically cleans trailing punctuation and spaces.\n💡 Fixes the common issue where ads get disapproved for “too long” text."
},
"typeVersion": 1
},
{
"id": "adb78971-8b7e-4eaf-9d9a-c47439b35c12",
"name": "Note adhésive5",
"type": "n8n-nodes-base.stickyNote",
"position": [
-320,
400
],
"parameters": {
"width": 340,
"height": 320,
"content": "## 🟦 Compliance Check Agent\n\n🧠 Purpose: Checks generated ad text for Google Ads policy compliance.\nAPI: POST /agents/google_ads_compliance_checker/run\n\nFlags emojis, exaggerated claims, restricted terms, etc.\nReturns \"APPROVED\" or \"REJECTED\".\n🔐 Uses your Relevance AI Agent ID."
},
"typeVersion": 1
}
],
"pinData": {},
"connections": {
"c9bfe847-7215-4ae5-a61e-4eb24ae1da0c": {
"main": [
[
{
"node": "4b90a3bc-da44-44bc-b19a-d81427545f26",
"type": "main",
"index": 0
}
],
[
{
"node": "bea6c2f0-4a95-4848-85a1-780cc048b26d",
"type": "main",
"index": 0
}
]
]
},
"4b90a3bc-da44-44bc-b19a-d81427545f26": {
"main": [
[
{
"node": "95f53395-e93d-4309-8f98-d2b2378cbb13",
"type": "main",
"index": 0
}
]
]
},
"b07b40c1-4c8c-4c35-aa07-be7025ed68b2": {
"main": [
[
{
"node": "01a193ae-a04f-4a04-9c3e-0a40c8e197c5",
"type": "main",
"index": 0
}
]
]
},
"95f53395-e93d-4309-8f98-d2b2378cbb13": {
"main": [
[
{
"node": "5a27e0f1-d4e7-4a81-9700-dd1c368fbb0b",
"type": "main",
"index": 0
}
]
]
},
"5a27e0f1-d4e7-4a81-9700-dd1c368fbb0b": {
"main": [
[
{
"node": "56492a38-1188-4b35-999f-14017ac76484",
"type": "main",
"index": 0
}
]
]
},
"01a193ae-a04f-4a04-9c3e-0a40c8e197c5": {
"main": [
[
{
"node": "50792985-2a29-4aeb-8191-2a1ef3d7ec97",
"type": "main",
"index": 0
}
]
]
},
"56492a38-1188-4b35-999f-14017ac76484": {
"main": [
[
{
"node": "55d4429b-b4a9-4a19-bb34-2f58d7881a5c",
"type": "main",
"index": 0
}
]
]
},
"534ea658-ea94-4d4c-a4e3-5d5fa1c699b6": {
"main": [
[
{
"node": "c9bfe847-7215-4ae5-a61e-4eb24ae1da0c",
"type": "main",
"index": 0
}
]
]
},
"f25d680c-28ad-43f5-96c8-0f022cf1a710": {
"main": [
[
{
"node": "b07b40c1-4c8c-4c35-aa07-be7025ed68b2",
"type": "main",
"index": 0
}
]
]
},
"17d67f83-8300-41c5-a3ac-9b7ea4bc25c2": {
"main": [
[
{
"node": "534ea658-ea94-4d4c-a4e3-5d5fa1c699b6",
"type": "main",
"index": 0
}
]
]
},
"50792985-2a29-4aeb-8191-2a1ef3d7ec97": {
"main": [
[
{
"node": "17d67f83-8300-41c5-a3ac-9b7ea4bc25c2",
"type": "main",
"index": 0
}
]
]
}
}
}Comment utiliser ce workflow ?
Copiez le code de configuration JSON ci-dessus, créez un nouveau workflow dans votre instance n8n et sélectionnez "Importer depuis le JSON", collez la configuration et modifiez les paramètres d'authentification selon vos besoins.
Dans quelles scénarios ce workflow est-il adapté ?
Avancé - Création de contenu, IA Multimodale
Est-ce payant ?
Ce workflow est entièrement gratuit et peut être utilisé directement. Veuillez noter que les services tiers utilisés dans le workflow (comme l'API OpenAI) peuvent nécessiter un paiement de votre part.
Workflows recommandés
Nikan Noorafkan
@nikkannooraHey, I’m Nikan Noorafkan — a creator passionate about building smart, automated workflows that drive business outcomes. With a background in performance marketing, user acquisition, and retention strategies, I use n8n to connect data, automate repetitive tasks, and scale growth across the funnel.
Partager ce workflow