This workflow follows the Chainllm → Gmail 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": "Intelligent Email Routing System",
"nodes": [
{
"parameters": {
"pollTimes": {
"item": [
{
"mode": "everyMinute"
}
]
},
"filters": {
"q": "in:inbox -label:Processed "
}
},
"type": "n8n-nodes-base.gmailTrigger",
"typeVersion": 1.3,
"position": [
-560,
48
],
"id": "YOUR_ID_HERE",
"name": "Gmail Trigger",
"credentials": {
"gmailOAuth2": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"model": {
"__rl": true,
"value": "gpt-4.1-mini",
"mode": "list",
"cachedResultName": "gpt-4.1-mini"
},
"builtInTools": {},
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"typeVersion": 1.3,
"position": [
-40,
272
],
"id": "YOUR_ID_HERE",
"name": "OpenAI Chat Model",
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"promptType": "define",
"text": "={{ $json.Email_snippet }}",
"hasOutputParser": true,
"messages": {
"messageValues": [
{
"message": "=## ROLE\n\nYou're an intelligent email classifier for KOKA Retail Store that classifies emails based on contexts, not just keyowrds.\n\n## TASK\nClassify the incoming email into exactly ONE of these:\n- Sales\n- Logistics\n- Billing\n- Customer Support\n- Office/Admin\n\n### KEYWORDS\n- Sales: Product inquiries, bulk orders, pricing discussions, partnerships, purchase requests.\n- Logistics: Delivery issues, shipment tracking, inventory movement, dispatch coordination.\n- Billing: Invoices, payments, receipts, refunds, financial questions.\n- Customer Support: Complaints, product issues, returns, general assistance.\n- Office/Admin: Internal communication, vendor messages, unclear or uncategorized emails.\n\n### CONFIDENCE SCORING\nAssign a confidence score between 0.00 and 1.00 based on how certain you are on your classification:\n- 0.75 \u2013 1.00 \u2192 High Confidence: clear intent, route directly\n- 0.00 \u2013 0.74 \u2192 Risky: unclear intent, default label must be Office/Admin\n\n## RULES\n- If confidence is below 0.75 you MUST set label to \"Office/Admin\"\n- Never guess \u2014 when in doubt, use Office/Admin\n- Ignore email signatures, greetings, and formatting\n- Focus only on the core intent of the message\n- Do not rely on KEYWORDS alone, analyze email context as well\n\n## OUTPUT\nReturn this JSON:\n{\"label\": \"Department Name\", \"confidence\": 0.00}\nFrom: {{ $json.From }}\nSubject: {{ $json.Subject }}\nBody: {{ $json.Email_snippet }}\nReceived: {{ $json.Received_at }}\nMessage_id: {{ $json.Message_id }}"
}
]
},
"batching": {}
},
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"typeVersion": 1.9,
"position": [
-112,
48
],
"id": "YOUR_ID_HERE",
"name": "Handle routing logic and confidence score"
},
{
"parameters": {
"jsCode": "const llmOutput = $input.first().json;\nconst emailData = $('Get email info').first().json;\n\nconst raw = llmOutput.text || llmOutput.message?.content || llmOutput.content || \"\";\nconst clean = raw.replace(/```json|```/g, \"\").trim();\nconst parsed = JSON.parse(clean);\n\nconst confidence = parsed.confidence;\nconst label = confidence >= 0.75 ? parsed.label : \"Office/Admin\";\nconst confidence_level = confidence >= 0.75 ? \"High Confidence\" : \"Risky\";\n\nreturn [{ json: {\n message_id : emailData.Message_id,\n from : emailData.From,\n subject : emailData.Subject,\n body : emailData.Email_snippet,\n received_at : emailData.Received_at,\n finalLabel : label,\n confidence : confidence,\n confidence_level : confidence_level\n}}];"
},
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
240,
48
],
"id": "YOUR_ID_HERE",
"name": "Split logic reasoning"
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "YOUR_ID_HERE",
"name": "Message_id",
"value": "={{ $json.id }}",
"type": "string"
},
{
"id": "YOUR_ID_HERE",
"name": "From",
"value": "={{ $json.From }}",
"type": "string"
},
{
"id": "YOUR_ID_HERE",
"name": "Subject",
"value": "={{ $json.Subject }}",
"type": "string"
},
{
"id": "YOUR_ID_HERE",
"name": "Email_snippet",
"value": "={{ $json.snippet }}",
"type": "string"
},
{
"id": "YOUR_ID_HERE",
"name": "Received_at",
"value": "={{ new Date(parseInt($json.internalDate)).toLocaleString('en-GB') }}",
"type": "string"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
-336,
48
],
"id": "YOUR_ID_HERE",
"name": "Get email info"
},
{
"parameters": {
"authentication": "oAuth2",
"select": "channel",
"channelId": {
"__rl": true,
"value": "={{ $json.target_slack }}",
"mode": "id"
},
"text": "=\ud83d\udce7 *New Email Routed to the {{ $json.target_dept }} Team\n\n*From:* {{ $json.from.match(/<([^>]+)>/)?.[1] || $json.from }}\n*Subject:* {{ $json.subject }}\n*Received:* {{ $json.received_at }}\n*Email ID:* {{ $json.message_id }}",
"otherOptions": {
"includeLinkToWorkflow": false
}
},
"type": "n8n-nodes-base.slack",
"typeVersion": 2.4,
"position": [
912,
48
],
"id": "YOUR_ID_HERE",
"name": "Send notification",
"credentials": {
"slackOAuth2Api": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"documentId": {
"__rl": true,
"value": "YOUR_SHEET_ID_HERE",
"mode": "list",
"cachedResultName": "Routing Config Sheet",
"cachedResultUrl": "https://docs.google.com/spreadsheets/d/YOUR_SHEET_ID_HERE/edit"
},
"sheetName": {
"__rl": true,
"value": "YOUR_SHEET_TAB_ID",
"mode": "list",
"cachedResultName": "AI routing system",
"cachedResultUrl": "https://docs.google.com/spreadsheets/d/YOUR_SHEET_ID_HERE"
},
"filtersUI": {
"values": [
{
"lookupColumn": "Department",
"lookupValue": "={{ $json.finalLabel }}"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.googleSheets",
"typeVersion": 4.7,
"position": [
464,
48
],
"id": "YOUR_ID_HERE",
"name": "Get dept. email and slack channel",
"credentials": {
"googleSheetsOAuth2Api": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"jsCode": "const item = $input.first().json;\nconst emailData = $('Get email info').first().json;\n\nconsole.log(JSON.stringify(Object.keys(item)));\n\nreturn [{ json: {\n message_id : emailData.Message_id,\n from : emailData.From,\n subject : emailData.Subject,\n body : emailData.Email_snippet,\n received_at : emailData.Received_at,\n finalLabel : item.finalLabel,\n confidence : item.confidence,\n confidence_level : item.confidence_level,\n target_dept : item['Department'],\n target_slack : item['Slack Channel'],\n target_email : item['Email'],\n email_label : item['Label ID'],\n}}];"
},
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
688,
48
],
"id": "YOUR_ID_HERE",
"name": "Preserve email"
},
{
"parameters": {
"operation": "addLabels",
"messageId": "={{ $('Preserve email').item.json.message_id }}",
"labelIds": "={{ $('Preserve email').item.json.email_label }}"
},
"type": "n8n-nodes-base.gmail",
"typeVersion": 2.2,
"position": [
1136,
48
],
"id": "YOUR_ID_HERE",
"name": "Attach email label to dept",
"credentials": {
"gmailOAuth2": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"operation": "addLabels",
"messageId": "={{ $('Preserve email').item.json.message_id }}",
"labelIds": "=YOUR_PROCESSED_LABEL_ID"
},
"type": "n8n-nodes-base.gmail",
"typeVersion": 2.2,
"position": [
1360,
48
],
"id": "YOUR_ID_HERE",
"name": "Attach processed/routed label to email",
"credentials": {
"gmailOAuth2": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"content": "## Email intake\n",
"height": 320,
"width": 432
},
"type": "n8n-nodes-base.stickyNote",
"position": [
-592,
-80
],
"typeVersion": 1,
"id": "YOUR_ID_HERE",
"name": "Sticky Note"
},
{
"parameters": {
"content": "## AI Classification and confidence scoring\n",
"height": 320,
"width": 528,
"color": 6
},
"type": "n8n-nodes-base.stickyNote",
"position": [
-128,
-80
],
"typeVersion": 1,
"id": "YOUR_ID_HERE",
"name": "Sticky Note1"
},
{
"parameters": {
"content": "## Routing & Delivery\n",
"height": 320,
"width": 592,
"color": 3
},
"type": "n8n-nodes-base.stickyNote",
"position": [
432,
-80
],
"typeVersion": 1,
"id": "YOUR_ID_HERE",
"name": "Sticky Note2"
},
{
"parameters": {
"content": "## Label & Close\n",
"height": 320,
"width": 544,
"color": 4
},
"type": "n8n-nodes-base.stickyNote",
"position": [
1040,
-80
],
"typeVersion": 1,
"id": "YOUR_ID_HERE",
"name": "Sticky Note3"
}
],
"connections": {
"Gmail Trigger": {
"main": [
[
{
"node": "Get email info",
"type": "main",
"index": 0
}
]
]
},
"OpenAI Chat Model": {
"ai_languageModel": [
[
{
"node": "Handle routing logic and confidence score",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Handle routing logic and confidence score": {
"main": [
[
{
"node": "Split logic reasoning",
"type": "main",
"index": 0
}
]
]
},
"Split logic reasoning": {
"main": [
[
{
"node": "Get dept. email and slack channel",
"type": "main",
"index": 0
}
]
]
},
"Get email info": {
"main": [
[
{
"node": "Handle routing logic and confidence score",
"type": "main",
"index": 0
}
]
]
},
"Send notification": {
"main": [
[
{
"node": "Attach email label to dept",
"type": "main",
"index": 0
}
]
]
},
"Get dept. email and slack channel": {
"main": [
[
{
"node": "Preserve email",
"type": "main",
"index": 0
}
]
]
},
"Preserve email": {
"main": [
[
{
"node": "Send notification",
"type": "main",
"index": 0
}
]
]
},
"Attach email label to dept": {
"main": [
[
{
"node": "Attach processed/routed label to email",
"type": "main",
"index": 0
}
]
]
}
},
"active": false,
"settings": {
"executionOrder": "v1",
"binaryMode": "separate",
"availableInMCP": false
},
"versionId": "YOUR_VERSION_ID",
"meta": {
"templateCredsSetupCompleted": true
},
"id": "YOUR_WORKFLOW_ID",
"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.
gmailOAuth2googleSheetsOAuth2ApiopenAiApislackOAuth2Api
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
How this works
This workflow automates the intelligent routing of incoming emails to the appropriate department or team, saving hours of manual sorting and ensuring timely responses. It leverages AI to analyse email content, sender details, and context, assigning a confidence score to determine the best route with minimal errors. Ideal for support teams, customer service managers, or growing businesses handling high email volumes, it integrates seamlessly with Gmail for triggering and Slack for instant notifications. The key step involves the AI-powered chain that processes the email and decides the routing path based on predefined logic.
Use this workflow when you receive diverse customer enquiries via Gmail and need to distribute them efficiently to specialised teams, such as IT support or sales. Avoid it for very low-volume inboxes or scenarios requiring human oversight for sensitive decisions, like legal matters. Common variations include adapting the Google Sheets lookup for custom department mappings or adding follow-up automations in tools like Zendesk.
About this workflow
Intelligent Email Routing System. Uses gmailTrigger, lmChatOpenAi, chainLlm, slack. Event-driven trigger; 14 nodes.
Source: https://github.com/ChibugoOhanyiri/AI-automation-portfolio/blob/main/ai-email-routing-system/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.
This template is ideal for HR teams, startup founders, operations leads, remote-first companies, and freelancers managing onboarding manually or across multiple tools.
✨🔪 Advanced AI Powered Document Parsing & Text Extraction with Llama Parse. Uses gmail, gmailTrigger, limit, stickyNote. Webhook trigger; 54 nodes.
n8n Graphic Design Team. Uses googleSheets, googleDrive, httpRequest, outputParserStructured. Event-driven trigger; 37 nodes.
Wait Splitout. Uses lmOpenAi, outputParserStructured, httpRequest, stickyNote. Event-driven trigger; 26 nodes.
Wait Splitout. Uses lmOpenAi, outputParserStructured, httpRequest, stickyNote. Event-driven trigger; 26 nodes.