Nocode Expert This workflow corresponds to n8n.io template #17610 — we link there as the canonical source.
This workflow follows the Gmail → Google Sheets recipe pattern — see all workflows that pair these two integrations.
The workflow JSON
Copy or download the full n8n JSON below. Paste it into a new n8n workflow, add your credentials, activate. Full import guide →
{
"id": "adbe2c4e3e47afc2",
"meta": {
"templateCredsSetupCompleted": true
},
"name": "Find Meta ads with below-average quality rankings and fix them with OpenAI",
"tags": [],
"nodes": [
{
"id": "3e9d8435b91452f5",
"name": "Run manually",
"type": "n8n-nodes-base.manualTrigger",
"position": [
-400,
128
],
"parameters": {},
"typeVersion": 1
},
{
"id": "cf2444d06728e9b4",
"name": "Set config: quality digest",
"type": "n8n-nodes-base.code",
"position": [
-160,
128
],
"parameters": {
"jsCode": "// Below-average quality-ranking digest - tune here.\nreturn [{ json: {\n DATE_PRESET: 'last_7d', // MUST be a short window: quality_ranking\n // returns UNKNOWN on last_30d, verified on a\n // 465-ad account incl. ads with 550k impressions\n MIN_SPEND: 20, // only flag ads that spent at least this\n} }];"
},
"typeVersion": 2
},
{
"id": "df31ba9fba2f9660",
"name": "Fetch ad quality rankings (Meta)",
"type": "n8n-nodes-base.httpRequest",
"position": [
80,
128
],
"parameters": {
"url": "=https://graph.facebook.com/{{ $env.META_API_VERSION || 'v21.0' }}/{{ $env.META_AD_ACCOUNT_ID }}/insights",
"options": {},
"sendQuery": true,
"queryParameters": {
"parameters": [
{
"name": "access_token",
"value": "={{ $env.META_ACCESS_TOKEN }}"
},
{
"name": "level",
"value": "ad"
},
{
"name": "date_preset",
"value": "={{ $('Set config: quality digest').first().json.DATE_PRESET }}"
},
{
"name": "fields",
"value": "ad_name,campaign_name,adset_name,spend,impressions,quality_ranking,engagement_rate_ranking,conversion_rate_ranking"
},
{
"name": "limit",
"value": "500"
}
]
}
},
"typeVersion": 4.2
},
{
"id": "527b5fc45e7a507b",
"name": "Flag below-average ads",
"type": "n8n-nodes-base.code",
"position": [
320,
128
],
"parameters": {
"jsCode": "const cfg = $('Set config: quality digest').first().json;\nconst rows = $json.data || [];\nconst isLow = (v) => typeof v === 'string' && v.indexOf('BELOW_AVERAGE') === 0;\nconst flagged = [];\nfor (const r of rows) {\n const spend = Number(r.spend)||0;\n if (spend < cfg.MIN_SPEND) continue;\n const low = [];\n if (isLow(r.quality_ranking)) low.push('quality');\n if (isLow(r.engagement_rate_ranking)) low.push('engagement');\n if (isLow(r.conversion_rate_ranking)) low.push('conversion');\n if (!low.length) continue;\n flagged.push({ ad: r.ad_name, campaign: r.campaign_name, adset: r.adset_name, spend:+spend.toFixed(2),\n quality: r.quality_ranking||'UNKNOWN', engagement: r.engagement_rate_ranking||'UNKNOWN', conversion: r.conversion_rate_ranking||'UNKNOWN', below_on: low });\n}\nflagged.sort((a,b)=> b.spend-a.spend);\nreturn [{ json: { date_preset: cfg.DATE_PRESET, flagged_count: flagged.length, spend_at_risk: +flagged.reduce((n,f)=>n+f.spend,0).toFixed(2), flagged } }];"
},
"typeVersion": 2
},
{
"id": "6347b60ff2f976e7",
"name": "Build AI quality prompt",
"type": "n8n-nodes-base.code",
"position": [
624,
128
],
"parameters": {
"jsCode": "const d = $json;\nconst system = 'You are a Meta creative strategist. Meta rates each ad below/average/above average on quality, engagement-rate and conversion-rate ranking. Below-average quality points to the creative, engagement to the hook/relevance, conversion to the offer/landing page. For each flagged ad give the most likely lever and one fix. Be terse. Return ONLY valid JSON, no prose, no markdown fences.';\nconst user = 'Ads spending with a below-average ranking (worst-spend first):\\n' + JSON.stringify(d.flagged.slice(0,20)) + '\\n\\nReturn JSON exactly: {\"summary\":\"one sentence\",\"fixes\":[{\"ad\":\"\",\"below_on\":\"\",\"lever\":\"\",\"fix\":\"\"}]}';\nconst body = { model:'gpt-4o-mini', max_tokens:3000, response_format:{type:'json_object'},\n messages:[{role:'system',content:system},{role:'user',content:user}] };\nreturn [{ json: { body, _ctx: d } }];"
},
"typeVersion": 2
},
{
"id": "3262fecc9839803e",
"name": "Explain quality fixes with OpenAI",
"type": "n8n-nodes-base.httpRequest",
"position": [
880,
128
],
"parameters": {
"url": "https://api.openai.com/v1/chat/completions",
"method": "POST",
"options": {},
"jsonBody": "={{ $json.body }}",
"sendBody": true,
"sendHeaders": true,
"specifyBody": "json",
"headerParameters": {
"parameters": [
{
"name": "Authorization",
"value": "=Bearer {{ $env.OPENAI_API_KEY }}"
},
{
"name": "content-type",
"value": "application/json"
}
]
}
},
"typeVersion": 4.2
},
{
"id": "fb57e12abfb2221f",
"name": "Print quality digest report",
"type": "n8n-nodes-base.code",
"position": [
1312,
128
],
"parameters": {
"jsCode": "const ctx = $('Build AI quality prompt').first().json._ctx || {};\nfunction aiJson(fallback){ const t=($json.choices&&$json.choices[0]&&$json.choices[0].message&&$json.choices[0].message.content)||''; let s=String(t).trim().replace(/^```(?:json)?/i,'').replace(/```$/,'').trim(); try{return JSON.parse(s)}catch(e){} const m=s.match(/\\{[\\s\\S]*\\}/); if(m){try{return JSON.parse(m[0])}catch(e){}} return fallback; }\nconst ai = aiJson({ summary:'AI response could not be parsed.', fixes:[] });\nconst lines=[];\nlines.push('META QUALITY-RANKING DIGEST (' + (ctx.date_preset||'') + ')');\nlines.push('Below-average ads: ' + (ctx.flagged_count||0) + ' | Spend at risk: $' + (ctx.spend_at_risk||0));\nlines.push('');\nlines.push(ai.summary||'');\nfor (const f of (ai.fixes||[])) { lines.push('- ' + f.ad + ' [' + f.below_on + ' -> ' + (f.lever||'') + ']'); if (f.fix) lines.push(' ' + f.fix); }\nreturn [{ json: { report: lines.join('\\n'), context: ctx, ai } }];"
},
"typeVersion": 2
},
{
"id": "46c3f9405c460a72",
"name": "Sticky Note - Overview",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1776,
-304
],
"parameters": {
"width": 900,
"height": 712,
"content": "## Meta Below-Average Quality-Ranking Digest\n\nSurfaces the ads Meta is quietly penalising. Meta scores every ad below/average/above average on quality, engagement-rate and conversion-rate ranking; below average means higher costs and less reach. This finds them and tells you which lever to pull.\n\n### How it works\n- Pulls ad-level insights with the three Meta ranking diagnostics from the Marketing API.\n- Flags ads spending over a floor that are below average on any ranking.\n- Maps each ranking to its lever: quality to creative, engagement to hook, conversion to offer or landing page.\n- OpenAI gives a specific fix per ad, worst-spend first.\n\n### Setup\n1. Add to your environment: META_ACCESS_TOKEN, META_AD_ACCOUNT_ID, META_API_VERSION, OPENAI_API_KEY.\n2. Run with the manual trigger, or attach a Schedule trigger for a weekly digest.\n\n### Customization\nIn the Config node, tune MIN_SPEND. Swap the report node for a Slack or email node. (Keep DATE_PRESET short. Meta only populates the three rankings on a recent rolling window, so last_30d returns UNKNOWN for every ad regardless of spend or impressions. last_7d is the safe default.)\n\nBuilt by **nocode.expert** - done-for-you automation & tracking.\n\nFull walkthrough of this workflow: https://nocode.expert/resources/n8n-meta-ads-quality-ranking-automation"
},
"typeVersion": 1
},
{
"id": "3d78aa29c9a01430",
"name": "Sticky Note - Section 1",
"type": "n8n-nodes-base.stickyNote",
"position": [
16,
0
],
"parameters": {
"color": 7,
"width": 472,
"height": 380,
"content": "## 1. Fetch rankings\nPull ad-level quality, engagement and conversion rankings."
},
"typeVersion": 1
},
{
"id": "b8b4e657856c4cf7",
"name": "Sticky Note - Section 2",
"type": "n8n-nodes-base.stickyNote",
"position": [
544,
-16
],
"parameters": {
"color": 7,
"width": 520,
"height": 380,
"content": "## 2. Flag below-average\nFind ads below average, then OpenAI gives a fix per ad."
},
"typeVersion": 1
},
{
"id": "faeaeae044f66ffc",
"name": "Sticky Note - Section 3",
"type": "n8n-nodes-base.stickyNote",
"position": [
1168,
-16
],
"parameters": {
"color": 7,
"width": 520,
"height": 380,
"content": "## 3. Report\nPrint the digest (swap for Slack/email)."
},
"typeVersion": 1
},
{
"id": "47939dc8335a4c0b",
"name": "Optional: run this weekly",
"type": "n8n-nodes-base.scheduleTrigger",
"disabled": true,
"position": [
-368,
592
],
"parameters": {
"rule": {
"interval": [
{
"field": "weeks",
"triggerAtDay": [
1
],
"triggerAtHour": 9
}
]
}
},
"typeVersion": 1.2
},
{
"id": "47b5945c12544191",
"name": "Optional: log flagged ads to Google Sheets",
"type": "n8n-nodes-base.googleSheets",
"disabled": true,
"position": [
-128,
608
],
"parameters": {
"operation": "append",
"sheetName": {
"__rl": true,
"mode": "list",
"value": ""
},
"documentId": {
"__rl": true,
"mode": "list",
"value": ""
}
},
"typeVersion": 4.5
},
{
"id": "93ef7a37fe1c4d55",
"name": "Sticky Note - Optional add-ons",
"type": "n8n-nodes-base.stickyNote",
"position": [
-416,
464
],
"parameters": {
"width": 900,
"height": 60,
"content": "## Optional add-ons (not wired up)\n\nLeft disconnected so you can plug in only what you need.\n\n- **Run this weekly** - quality ranking moves slowly and the default window is 30 days, so weekly is the right cadence. Swap this in for the manual trigger.\n- **Log flagged ads to Google Sheets** - connect after the report node. This is the one that pays off: a ranking is only useful once you can see whether last week's creative fix actually moved it.\n- **Email the digest via Outlook** - connect after the report node to send rather than read in the editor.\n\nNone are required. The workflow runs end to end from the manual trigger with all three disconnected."
},
"typeVersion": 1
},
{
"id": "adf33bca-b9c9-4835-8820-c08a7ba3adc3",
"name": "Send a message",
"type": "n8n-nodes-base.gmail",
"position": [
1520,
128
],
"parameters": {
"message": "={{ $json.report }}",
"options": {}
},
"typeVersion": 2.2
}
],
"active": false,
"settings": {
"binaryMode": "separate",
"executionOrder": "v1"
},
"versionId": "41a403dc-af76-4f59-99be-a2a13b1ba8ab",
"nodeGroups": [],
"connections": {
"Run manually": {
"main": [
[
{
"node": "Set config: quality digest",
"type": "main",
"index": 0
}
]
]
},
"Flag below-average ads": {
"main": [
[
{
"node": "Build AI quality prompt",
"type": "main",
"index": 0
}
]
]
},
"Build AI quality prompt": {
"main": [
[
{
"node": "Explain quality fixes with OpenAI",
"type": "main",
"index": 0
}
]
]
},
"Set config: quality digest": {
"main": [
[
{
"node": "Fetch ad quality rankings (Meta)",
"type": "main",
"index": 0
}
]
]
},
"Print quality digest report": {
"main": [
[
{
"node": "Send a message",
"type": "main",
"index": 0
}
]
]
},
"Fetch ad quality rankings (Meta)": {
"main": [
[
{
"node": "Flag below-average ads",
"type": "main",
"index": 0
}
]
]
},
"Explain quality fixes with OpenAI": {
"main": [
[
{
"node": "Print quality digest report",
"type": "main",
"index": 0
}
]
]
}
}
}
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
This workflow pulls daily reach and frequency insights from the Meta Marketing API, detects early signs of audience saturation, asks Anthropic Claude to summarize the status and recommended actions, and emails a formatted saturation forecast report via Gmail. Runs when triggered…
Source: https://n8n.io/workflows/17610/ — original creator credit. Request a take-down →
Related workflows
Workflows that share integrations, category, or trigger type with this one. All free to copy and import.
This workflow monitors unread Gmail support emails, uses Google Gemini to classify and summarize each message and draft a reply, logs the ticket details to Google Sheets, creates a Gmail draft respons
Gmail Trigger, HTTP Request, Google Sheets +2
This workflow polls the CoinDesk RSS feed every 10 minutes, filters for crypto-related stories, checks Google Sheets to avoid duplicates, uses OpenAI to score and summarize risk, logs results to Googl
Rss Feed Read Trigger, Google Sheets, HTTP Request +1
This workflow hosts an n8n intake form, sends submissions to Anthropic Claude for practice-area classification and qualification, then routes results to Gmail and Google Sheets before returning a conf
Form Trigger, HTTP Request, Gmail +2
This workflow polls Gmail for emails with the subject “Trade Instruction”, uses OpenAI Chat Completions to extract trade fields from the email body, logs valid and failed extractions to separate tabs
Gmail Trigger, HTTP Request, Google Sheets +1
This workflow collects SIP and mutual fund details via an n8n Form, calculates goal and performance metrics, generates a personalized advisory using OpenAI, emails a color-coded HTML report via Gmail,
Form Trigger, HTTP Request, Google Sheets +1