This workflow follows the Gmail → OpenAI 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 →
{
"name": "26-ai-customer-churn-prediction-retention",
"nodes": [
{
"parameters": {
"rule": {
"interval": [
{
"field": "days",
"triggerAt": "08:00"
}
]
}
},
"id": "f6g7h8i9-1111-4444-8888-000000000001",
"name": "Schedule",
"type": "n8n-nodes-base.schedule",
"typeVersion": 1,
"position": [
240,
400
]
},
{
"parameters": {
"operation": "executeQuery",
"query": "SELECT customer_id, name, email, phone, last_visit_date, visit_count, avg_spend, complaint_count, survey_score FROM customer_interactions WHERE is_active = true;",
"options": {}
},
"id": "f6g7h8i9-2222-4444-8888-000000000002",
"name": "PostgreSQL: Fetch Customers",
"type": "n8n-nodes-base.postgres",
"typeVersion": 2.5,
"position": [
460,
400
],
"credentials": {
"postgres": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"jsCode": "// \u0622\u0645\u0627\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u062f\u0627\u062f\u0647\u200c\u0647\u0627 \u0628\u0631\u0627\u06cc \u062a\u062d\u0644\u06cc\u0644 AI\nconst items = $input.all();\nconst today = new Date();\n\nreturn items.map(item => {\n const data = item.json;\n const daysSinceLastVisit = Math.floor((today - new Date(data.last_visit_date)) / (1000 * 60 * 60 * 24));\n \n return {\n json: {\n customer_id: data.customer_id,\n name: data.name,\n email: data.email,\n phone: data.phone,\n days_since_last_visit: daysSinceLastVisit,\n visit_count: data.visit_count,\n avg_spend: data.avg_spend,\n complaint_count: data.complaint_count,\n survey_score: data.survey_score,\n ai_prompt: `Customer: ${data.name}. Days since last visit: ${daysSinceLastVisit}. Total visits: ${data.visit_count}. Avg spend: $${data.avg_spend}. Complaints: ${data.complaint_count}. Last survey score: ${data.survey_score}/10. Predict churn probability (0-100), risk_level (High, Medium, Low), and generate a personalized retention offer.`\n }\n };\n});"
},
"id": "f6g7h8i9-3333-4444-8888-000000000003",
"name": "Code: Feature Engineering",
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
680,
400
]
},
{
"parameters": {
"model": "gpt-4o",
"prompt": {
"messages": [
{
"role": "system",
"content": "You are an expert Customer Retention AI. Analyze the customer data. Output ONLY valid JSON with these exact keys: 'churn_probability' (integer 0-100), 'risk_level' (string: 'High', 'Medium', or 'Low'), 'personalized_offer' (string: a specific, compelling discount or perk based on their history), and 'reasoning' (string: 1 sentence explaining the risk)."
},
{
"role": "user",
"content": "={{ $json.ai_prompt }}"
}
]
},
"options": {
"responseFormat": "json_object"
}
},
"id": "f6g7h8i9-4444-4444-8888-000000000004",
"name": "OpenAI: Churn Prediction",
"type": "n8n-nodes-base.openAi",
"typeVersion": 1.2,
"position": [
900,
400
],
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"rules": {
"values": [
{
"conditions": {
"options": {
"caseSensitive": true,
"leftValue": "",
"typeValidation": "strict"
},
"conditions": [
{
"id": "c1",
"leftValue": "={{ JSON.parse($json.content).risk_level }}",
"rightValue": "High",
"operator": {
"type": "string",
"operation": "equals"
}
}
]
}
},
{
"conditions": {
"options": {
"caseSensitive": true,
"leftValue": "",
"typeValidation": "strict"
},
"conditions": [
{
"id": "c2",
"leftValue": "={{ JSON.parse($json.content).risk_level }}",
"rightValue": "Medium",
"operator": {
"type": "string",
"operation": "equals"
}
}
]
}
}
]
},
"options": {
"fallbackOutput": "default"
}
},
"id": "f6g7h8i9-5555-4444-8888-000000000005",
"name": "Switch: Risk Routing",
"type": "n8n-nodes-base.switch",
"typeVersion": 3,
"position": [
1120,
400
]
},
{
"parameters": {
"sendTo": "={{ $json.email }}",
"subject": "We Miss You, {{ $json.name }}! Here's a Special Gift \ud83c\udf81",
"emailType": "text",
"message": "Dear {{ $json.name }},\n\nWe noticed it's been a while since your last visit. We value you as a customer!\n\nAs a token of our appreciation, here is your exclusive offer: {{ JSON.parse($node['OpenAI: Churn Prediction'].json.content).personalized_offer }}\n\nReply to this email or call us to book your next appointment.\n\nWarm regards,\nClinic Management Team",
"options": {}
},
"id": "f6g7h8i9-6666-4444-8888-000000000006",
"name": "Gmail: Retention Email",
"type": "n8n-nodes-base.gmail",
"typeVersion": 2.1,
"position": [
1340,
320
],
"credentials": {
"gmailOAuth2": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"fromNumber": "+1234567890",
"toNumber": "={{ $json.phone }}",
"message": "Hi {{ $json.name }}! We miss you. Enjoy {{ JSON.parse($node['OpenAI: Churn Prediction'].json.content).personalized_offer }}. Book now: https://clinic.example.com/book",
"options": {}
},
"id": "f6g7h8i9-7777-4444-8888-000000000007",
"name": "Twilio: SMS Alert",
"type": "n8n-nodes-base.twilio",
"typeVersion": 1,
"position": [
1560,
320
],
"credentials": {
"twilioApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"operation": "executeQuery",
"query": "INSERT INTO churn_predictions (customer_id, prediction_date, churn_probability, risk_level, offer_sent, channels_used) VALUES ('{{ $json.customer_id }}', NOW(), {{ JSON.parse($node['OpenAI: Churn Prediction'].json.content).churn_probability }}, '{{ JSON.parse($node['OpenAI: Churn Prediction'].json.content).risk_level }}', '{{ JSON.parse($node['OpenAI: Churn Prediction'].json.content).personalized_offer }}', 'Email,SMS') ON CONFLICT (customer_id, prediction_date) DO NOTHING;",
"options": {}
},
"id": "f6g7h8i9-8888-4444-8888-000000000008",
"name": "PostgreSQL: Log Prediction",
"type": "n8n-nodes-base.postgres",
"typeVersion": 2.5,
"position": [
1340,
480
],
"credentials": {
"postgres": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"conditions": {
"options": {
"caseSensitive": true,
"leftValue": "",
"typeValidation": "strict"
},
"conditions": [
{
"id": "cond1",
"leftValue": "={{ $json.has_responded_or_booked }}",
"rightValue": false,
"operator": {
"type": "boolean",
"operation": "equal"
}
},
{
"id": "cond2",
"leftValue": "={{ JSON.parse($node['OpenAI: Churn Prediction'].json.content).risk_level }}",
"rightValue": "High",
"operator": {
"type": "string",
"operation": "equals"
}
}
],
"combinator": "and"
},
"options": {}
},
"id": "f6g7h8i9-9999-4444-8888-000000000009",
"name": "IF: 7-Day Response Check",
"type": "n8n-nodes-base.if",
"typeVersion": 2,
"position": [
1560,
480
]
},
{
"parameters": {
"select": "channel",
"channelId": {
"__rl": true,
"value": "C1111222233",
"mode": "list",
"cachedResultName": "branch-manager-escalations"
},
"text": "\ud83d\udea8 *CHURN ESCALATION REQUIRED* \ud83d\udea8\n\n\ud83d\udc64 *Customer:* {{ $json.name }} (ID: {{ $json.customer_id }})\n\ud83d\udcc9 *Churn Probability:* {{ JSON.parse($node['OpenAI: Churn Prediction'].json.content).churn_probability }}%\n\u26a0\ufe0f *Risk Level:* High\n\ud83d\udcdd *Reasoning:* {{ JSON.parse($node['OpenAI: Churn Prediction'].json.content).reasoning }}\n\ud83c\udf81 *Offer Sent:* {{ JSON.parse($node['OpenAI: Churn Prediction'].json.content).personalized_offer }}\n\n\u23f3 *Status:* No response after 7 days. Please make a direct phone call.",
"otherOptions": {}
},
"id": "f6g7h8i9-0000-4444-8888-000000000010",
"name": "Slack: Manager Escalation",
"type": "n8n-nodes-base.slack",
"typeVersion": 2.2,
"position": [
1780,
420
],
"credentials": {
"slackApi": {
"name": "<your credential>"
}
}
}
],
"connections": {
"Schedule": {
"main": [
[
{
"node": "PostgreSQL: Fetch Customers",
"type": "main",
"index": 0
}
]
]
},
"PostgreSQL: Fetch Customers": {
"main": [
[
{
"node": "Code: Feature Engineering",
"type": "main",
"index": 0
}
]
]
},
"Code: Feature Engineering": {
"main": [
[
{
"node": "OpenAI: Churn Prediction",
"type": "main",
"index": 0
}
]
]
},
"OpenAI: Churn Prediction": {
"main": [
[
{
"node": "Switch: Risk Routing",
"type": "main",
"index": 0
}
]
]
},
"Switch: Risk Routing": {
"main": [
[
{
"node": "Gmail: Retention Email",
"type": "main",
"index": 0
}
],
[
{
"node": "Gmail: Retention Email",
"type": "main",
"index": 0
}
],
[
{
"node": "PostgreSQL: Log Prediction",
"type": "main",
"index": 0
}
]
]
},
"Gmail: Retention Email": {
"main": [
[
{
"node": "Twilio: SMS Alert",
"type": "main",
"index": 0
}
]
]
},
"Twilio: SMS Alert": {
"main": [
[
{
"node": "PostgreSQL: Log Prediction",
"type": "main",
"index": 0
}
]
]
},
"PostgreSQL: Log Prediction": {
"main": [
[
{
"node": "IF: 7-Day Response Check",
"type": "main",
"index": 0
}
]
]
},
"IF: 7-Day Response Check": {
"main": [
[
{
"node": "Slack: Manager Escalation",
"type": "main",
"index": 0
}
],
[]
]
}
},
"active": false,
"settings": {
"executionOrder": "v1",
"saveManualExecutions": true,
"saveDataErrorExecution": "all",
"saveDataSuccessExecution": "all"
},
"id": "26-ai-customer-churn-prediction-retention",
"tags": []
}
Credentials you'll need
Each integration node will prompt for credentials when you import. We strip credential IDs before publishing — you'll add your own.
gmailOAuth2openAiApipostgresslackApitwilioApi
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
26-ai-customer-churn-prediction-retention. Uses schedule, postgres, openAi, gmail. Scheduled trigger; 10 nodes.
Source: https://github.com/kooroosh1363/agentic-automation-lab/blob/main/26-ai-customer-churn-prediction-retention/workflow.json — 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.
Automates sales data analysis and strategic insight generation for sales managers and strategists needing actionable intelligence. Fetches multi-source data from sales, marketing, and financial system
HTTP Request, Agent, OpenAI Chat +6
Scheduled runs collect data from oil markets, global shipping movements, news sources, and official reports. The system performs statistical checks to detect anomalies and volatility shifts. An AI-dri
HTTP Request, Agent, Gmail +3
This workflow automates end-to-end patient care coordination by monitoring appointment schedules, clinical events, and care milestones while orchestrating personalized communications across multiple c
HTTP Request, Agent, OpenAI Chat +5
Automatically compare AI-generated email drafts against what your support team actually sent, learn from the differences, and improve future drafts over time — without any model fine-tuning.
Postgres, Gmail, Agent +3
Business Post Pipeline. Uses postgres, httpRequest, agent, lmChatGoogleGemini. Scheduled trigger; 30 nodes.
Postgres, HTTP Request, Agent +4