PollupAI This workflow corresponds to n8n.io template #10199 — we link there as the canonical source.
This workflow follows the Agent → Chainllm 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 →
{
"nodes": [
{
"id": "c00e5d2a-415e-40c1-9e9b-60f34aa3d5b4",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
-240,
0
],
"parameters": {
"width": 992,
"height": 368,
"content": "# Customer Health Scoring & Churn Alert\nThis workflow automatically calculates a customer health score by analyzing data from HubSpot and Google Sheets. If it detects a potential churn risk based on factors like ticket sentiment and product usage, it sends an alert email.\n\n## \ud83d\ude80 How to Set Up\n1. **Configure Credentials:** Add your credentials for the **HubSpot: Get All Deals**, **Config: Set LLM for Agent & Chains**, **Tool: Get Feature Usage from Sheets**, and **Email: Send Churn Alert** nodes.\n2. **Set Tool URLs:**\n * In **Tool: Calculate Sentiment Score**, enter the Webhook URL for your sentiment scoring workflow.\n * In **Tool: Get HubSpot Data**, enter the Endpoint URL for your MCP HubSpot data workflow.\n3. **Set Your Google Sheet:** In **Tool: Get Feature Usage from Sheets**, enter the Document ID for your own Google Sheet.\n4. **Update Email Details:** In **Email: Send Churn Alert**, change the `From` and `To` email addresses.\n5. **Activate Workflow:** Once configured, make sure to activate the workflow."
},
"typeVersion": 1
},
{
"id": "9e0e813a-b908-4025-b57d-a724289673ae",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
-240,
1232
],
"parameters": {
"color": 6,
"width": 432,
"height": 176,
"content": "## Contact me\n- If you need any modification to this workflow\n- if you need some help with this workflow\n- Or if you need any workflow in n8n, Make, or Langchain / Langgraph\n\nWrite to me: [thomas@pollup.net](<mailto:thomas@pollup.net>)"
},
"typeVersion": 1
},
{
"id": "50f33598-bcb5-4afd-8792-9368cfd0ce70",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"position": [
672,
1200
],
"parameters": {
"color": 7,
"height": 272,
"content": "This tool calls this workflow's own Webhook trigger. It sends all tickets at once to be processed and returns a single sentiment score."
},
"typeVersion": 1
},
{
"id": "fec2e35b-2cb4-43ae-82c8-af9f6ffec53b",
"name": "Sticky Note3",
"type": "n8n-nodes-base.stickyNote",
"position": [
1584,
816
],
"parameters": {
"color": 7,
"width": 352,
"height": 272,
"content": "This AI prompt checks for three churn signals:\n- Deal age > 1 year\n- A negative sentiment score\n- Declining feature usage over time"
},
"typeVersion": 1
},
{
"id": "34112bf3-f326-454d-a34f-cc59583e4ded",
"name": "Manual Trigger: Run Churn Analysis",
"type": "n8n-nodes-base.manualTrigger",
"position": [
-192,
960
],
"parameters": {},
"typeVersion": 1
},
{
"id": "124ca622-1418-4c72-abdf-1026dd1fdb53",
"name": "HubSpot: Get All Deals",
"type": "n8n-nodes-base.hubspot",
"position": [
32,
960
],
"parameters": {
"filters": {},
"resource": "deal",
"operation": "getAll",
"authentication": "oAuth2"
},
"typeVersion": 2.1
},
{
"id": "10109472-0242-4d47-bb89-b39683107f91",
"name": "Start Loop: For Each Deal",
"type": "n8n-nodes-base.splitInBatches",
"position": [
320,
960
],
"parameters": {
"options": {}
},
"typeVersion": 3
},
{
"id": "b76656fb-f5e9-4aff-bb49-39ed6b6edc1f",
"name": "Set: Isolate Current Deal ID Export to Sheets",
"type": "n8n-nodes-base.set",
"position": [
608,
960
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "a587d50b-5a8f-4650-a803-f9459d4f2342",
"name": "dealId",
"type": "string",
"value": "={{ $json.dealId }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "2a4d7904-4375-47a8-9797-1715b3596484",
"name": "Trigger: Receive Tickets for Scoring",
"type": "n8n-nodes-base.webhook",
"position": [
-192,
608
],
"parameters": {
"path": "9696956a-460a-4c45-aa3c-e5f83ce95e54",
"options": {},
"httpMethod": "POST",
"responseMode": "responseNode"
},
"typeVersion": 2.1
},
{
"id": "e9507eaa-294f-4ba8-a1d9-759f6ae58a5d",
"name": "Code: Format Tickets for Analysis",
"type": "n8n-nodes-base.code",
"position": [
32,
608
],
"parameters": {
"jsCode": "// Loop over input items and add a new field called 'myNewField' to the JSON of each one\n\nlet tickets = JSON.parse($input.first().json.body.tickets) \nlet tickets_text = []\nfor(var ticket of tickets) {\n if(ticket.properties) tickets_text.push(ticket.properties.subject.value + \"\\n\" + ticket.properties.content.value)\n else tickets_text.push(ticket.subject + \"\\n\" + ticket.content)\n \n}\n\nreturn {tickets_text}"
},
"typeVersion": 2
},
{
"id": "c34f65fd-2c46-47c0-9b06-87b090feca10",
"name": "Loop: For Each Ticket",
"type": "n8n-nodes-base.splitOut",
"position": [
256,
608
],
"parameters": {
"options": {},
"fieldToSplitOut": "tickets_text"
},
"typeVersion": 1
},
{
"id": "173c07bc-884e-48d3-b927-31cdc800c5b2",
"name": "AI: Analyze Ticket Sentiment",
"type": "@n8n/n8n-nodes-langchain.sentimentAnalysis",
"position": [
480,
480
],
"parameters": {
"options": {},
"inputText": "={{ $json.tickets_text }}"
},
"typeVersion": 1.1
},
{
"id": "db3e0604-ba4f-4f09-b281-42848441a5c6",
"name": "Merge: Consolidate Results",
"type": "n8n-nodes-base.merge",
"position": [
832,
592
],
"parameters": {
"numberInputs": 3
},
"typeVersion": 3.2
},
{
"id": "2be479b2-d019-43d9-a06a-052991743f4d",
"name": "Code: Convert Sentiment to Score",
"type": "n8n-nodes-base.code",
"position": [
1248,
608
],
"parameters": {
"mode": "runOnceForEachItem",
"jsCode": "// Loop over input items and add a new field called 'myNewField' to the JSON of each one\n \nreturn {\n json: {\n score:\n $json.sentimentAnalysis.category === 'Positive'\n ? 10\n : $json.sentimentAnalysis.category === 'Neutral'\n ? 5\n : -10\n }\n }"
},
"typeVersion": 2
},
{
"id": "a52e3b3c-a73d-459e-9d68-9dc201383860",
"name": "Code: Sum All Ticket Scores",
"type": "n8n-nodes-base.code",
"position": [
1648,
608
],
"parameters": {
"jsCode": "// Loop over input items and add a new field called 'myNewField' to the JSON of each one\nlet total = 0;\nfor (const item of $input.all()) {\n total += item.json.score;\n}\nreturn [{ json: { totalScore: total } }];"
},
"typeVersion": 2
},
{
"id": "87806b1f-1246-4c82-9fec-71bd668368e9",
"name": "Respond: Return Total Score Export to Sheets",
"type": "n8n-nodes-base.respondToWebhook",
"position": [
1952,
608
],
"parameters": {
"options": {}
},
"typeVersion": 1.4
},
{
"id": "5099176a-7e74-4cfe-9e2e-82e7144db3b1",
"name": "AI Agent: Gather Customer Data",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
960,
960
],
"parameters": {
"text": "=You collect data for a customer scoring.\n1.\tDeal data\n\t\u2022\tGet the info for the deal id \"{{ $json.dealId }}\" using the \"Tool: Get HubSpot Data\".\n\t\u2022\tReturn the default fields useful for customer scoring.\n \u2022 Extract the `dealname` and return it as `companyName`.\n\t\u2022\tEnsure `associatedVids` is included.\n \u2022 Also return the timestamp of the closing of the deal which is in `hs_is_closed.timestamp`.\n2.\tContact email\n\t\u2022\tFrom the `associatedVids`, fetch the primary contact using the \"Tool: Get HubSpot Data\".\n\t\u2022\tReturn the contact\u2019s email address.\n3. Ticket search and scoring\n \u2022 Find all the tickets by using the tool \"Tool: Get HubSpot Data\".\n \u2022 Then send all the tickets at once to the tool \"Tool: Calculate Sentiment Score\".\n4. Feature search\n \u2022 Then search for **ALL** the features data usage for the contact\u2019s email address in the google sheet tool \"Tool: Get Feature Usage from Sheets\".",
"options": {
"returnIntermediateSteps": false
},
"promptType": "define",
"hasOutputParser": true
},
"typeVersion": 2.2
},
{
"id": "37a5efac-6dbf-4af0-830e-466e31561113",
"name": "Config: Set LLM for Agent & Chains",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
544,
736
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-5-mini",
"cachedResultName": "gpt-5-mini"
},
"options": {}
},
"typeVersion": 1.2
},
{
"id": "536718ef-3912-49fd-95c2-d172902e1b3a",
"name": "Tool: Get HubSpot Data",
"type": "@n8n/n8n-nodes-langchain.mcpClientTool",
"position": [
944,
1312
],
"parameters": {
"options": {},
"serverTransport": "httpStreamable"
},
"typeVersion": 1.1
},
{
"id": "6c13beac-207b-48fd-8514-242c2ed30c87",
"name": "Tool: Calculate Sentiment Score",
"type": "n8n-nodes-base.httpRequestTool",
"position": [
784,
1312
],
"parameters": {
"method": "POST",
"options": {},
"sendBody": true,
"bodyParameters": {
"parameters": [
{
"name": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('parameters0_Name', ``, 'string') }}",
"value": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('parameters0_Value', ``, 'string') }}"
}
]
},
"toolDescription": "Send all the info and return the score"
},
"typeVersion": 4.2
},
{
"id": "4584effb-8688-4f06-a495-e2b4b7ebfd2e",
"name": "Tool: Get Feature Usage from Sheets",
"type": "n8n-nodes-base.googleSheetsTool",
"position": [
1104,
1312
],
"parameters": {
"options": {},
"sheetName": {
"__rl": true,
"mode": "list",
"value": "gid=0",
"cachedResultUrl": "",
"cachedResultName": ""
},
"documentId": {
"__rl": true,
"mode": "id",
"value": ""
}
},
"typeVersion": 4.7
},
{
"id": "7f03cc9a-09a2-4d19-b386-09302bcbd699",
"name": "AI: Structure Agent's Findings Export to Sheets",
"type": "@n8n/n8n-nodes-langchain.outputParserStructured",
"position": [
1264,
1312
],
"parameters": {
"jsonSchemaExample": "{\n \"companyName\": \"\",\n \"email_of_contact\": \"\",\n \"dealstage\": \"\",\n \"amount_in_home_currency\": \"\",\n \"hs_forecast_amount\": \"\",\n \"Is closed status\": \"\",\n \"hs_is_closed_timestamp\": 12345,\n \"associatedVids\": [\"1234\"],\n \"score\": 10,\n \"tickets\": [{\n \"hs_client_id\": 123456,\n \"content\": \"\",\n \"subject\": \"\"\n }],\n \"feature\": [{\n \"Feature name\": \"\",\n \"Monthly Usage Frequency\": 12,\n \"Number of Users\": 12,\n \"Date\": \"29/09/2025\"\n }]\n}"
},
"typeVersion": 1.3
},
{
"id": "1147efd1-d880-4c75-a2da-7f6b74df3c43",
"name": "Code: Group Feature Data by Name",
"type": "n8n-nodes-base.code",
"position": [
1424,
960
],
"parameters": {
"jsCode": "// Loop over input items and add a new field called 'myNewField' to the JSON of each one\nlet features = $input.all()[0].json.output['feature']\nconsole.log($input.all()[0].json.output['feature'])\n\nlet acc = {}\nfor (const item of features) {\n if (!acc[item[\"Feature name\"]]) {\n acc[item[\"Feature name\"]] = [];\n }\n acc[item[\"Feature name\"]].push(item)\n}\n\nreturn { res: JSON.stringify(acc)};\n"
},
"typeVersion": 2
},
{
"id": "e1f8b15e-3be2-4727-aaba-040d18016057",
"name": "AI Chain: Analyze for Churn Risk",
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"position": [
1648,
960
],
"parameters": {
"text": "=You are an AI assistant for Customer Success teams. Your role is to generate clear, actionable email alerts when signs of potential customer churn are detected.\n \nTrigger conditions:\nGenerate an alert titled \u201cCustomer Churn Factors Detected\u201d if any of the following are true:\n \nThe time since {{ $('AI Agent: Gather Customer Data').item.json.output.hs_is_closed_timestamp }} is more than 1 year.\nThe score {{ $('AI Agent: Gather Customer Data').item.json.output.score }} is below 0.\nMonthly usage or number of users has decreased according to this data: {{ $json.res }}.\n\nEmail Output Instructions\nIf an alert is triggered, write a short, professional email for the Customer Success Manager.\nYour goal is to help them act quickly, not overwhelm them with raw data.\nDo not include any signature, sign-off, or sender name at the end of the email body.\n\nTone:\nClear, confident, and proactive.\nFriendly but concise \u2014 suitable for internal use.\nAvoid data dumps; summarize what matters.\n\nEmail structure:\nSubject:\n\u201c\u26a0\ufe0f Customer Churn Alert: [Company Name] \u2013 Thresholds Exceeded\u201d\n\nBody:\nContext / Summary \u2014 Briefly explain that churn risk indicators crossed a threshold (e.g. usage drop, low score, long inactivity).\nKey Signals (1\u20133 bullets) \u2014 Summarize the main patterns detected (e.g. \u201cUsage dropped 25% MoM\u201d, \u201cNo engagement since July\u201d, \u201cScore below 0 threshold\u201d).\nImpact / Risk Level \u2014 Classify risk as Low / Medium / High depending on how many thresholds were exceeded.\nRecommended Actions \u2014 Suggest 2\u20133 concrete next steps to help retain the customer, such as:\n* Schedule a check-in or QBR\n* Offer a training or usage review session\n* Share a new feature relevant to their goals\n* Investigate potential technical or value gaps\n\nClose:\nEnd with a short motivational note (\u201cAddressing this now can help restore engagement before renewal.\u201d).\n\nNote: Use the word \u201cThreshold\u201d naturally 2\u20133 times throughout to signal data awareness (e.g., \u201cUsage has dropped below the activity threshold\u201d).\n \n\n\ud83d\udca1 Example Output\nSubject: \u26a0\ufe0f Customer Churn Alert: Acme Corp \u2013 Usage Threshold Exceeded\nBody:\nHi Team,\nOur system detected that Acme Corp\u2019s engagement has dropped below several key thresholds.\n\nSignals:\n* Product usage fell 28% over the last 30 days\n* Active users decreased from 45 \u2192 32\n* Health score is currently \u20130.4 (below threshold)\n\nRisk Level: High\n\nRecommended Actions:\n* Reach out for a check-in to understand usage drop-off\n* Share recent case studies or new feature releases\n* Offer a short workflow review to identify blockers\n\nTaking action this week could help recover engagement before renewal.",
"batching": {},
"promptType": "define",
"hasOutputParser": true
},
"typeVersion": 1.7
},
{
"id": "c20a8716-68ce-4d76-911e-576a86ce5f7f",
"name": "AI: Structure Alert Email",
"type": "@n8n/n8n-nodes-langchain.outputParserStructured",
"position": [
1792,
1168
],
"parameters": {
"jsonSchemaExample": "{\n\t\"subject\": \"\",\n\t\"message\": \"\"\n}"
},
"typeVersion": 1.3
},
{
"id": "78bb2588-c4e0-4737-b21d-36889df26a28",
"name": "Format: Convert Alert to HTML",
"type": "n8n-nodes-base.markdown",
"position": [
2016,
960
],
"parameters": {
"mode": "markdownToHtml",
"options": {},
"markdown": "={{ $json.output.message }}",
"destinationKey": "message"
},
"typeVersion": 1
},
{
"id": "62eb0b37-a18b-4d49-b321-948ae60b8e79",
"name": "Email: Send Churn Alert Export to Sheets",
"type": "n8n-nodes-base.emailSend",
"position": [
2224,
960
],
"parameters": {
"html": "={{ $json.message }}",
"options": {
"appendAttribution": false
},
"subject": "={{ $json.output.subject }}"
},
"credentials": {
"smtp": {
"name": "<your credential>"
}
},
"typeVersion": 2.1
}
],
"connections": {
"Loop: For Each Ticket": {
"main": [
[
{
"node": "AI: Analyze Ticket Sentiment",
"type": "main",
"index": 0
}
]
]
},
"HubSpot: Get All Deals": {
"main": [
[
{
"node": "Start Loop: For Each Deal",
"type": "main",
"index": 0
}
]
]
},
"Tool: Get HubSpot Data": {
"ai_tool": [
[
{
"node": "AI Agent: Gather Customer Data",
"type": "ai_tool",
"index": 0
}
]
]
},
"AI: Structure Alert Email": {
"ai_outputParser": [
[
{
"node": "AI Chain: Analyze for Churn Risk",
"type": "ai_outputParser",
"index": 0
}
]
]
},
"Start Loop: For Each Deal": {
"main": [
[],
[
{
"node": "Set: Isolate Current Deal ID Export to Sheets",
"type": "main",
"index": 0
}
]
]
},
"Merge: Consolidate Results": {
"main": [
[
{
"node": "Code: Convert Sentiment to Score",
"type": "main",
"index": 0
}
]
]
},
"Code: Sum All Ticket Scores": {
"main": [
[
{
"node": "Respond: Return Total Score Export to Sheets",
"type": "main",
"index": 0
}
]
]
},
"AI: Analyze Ticket Sentiment": {
"main": [
[
{
"node": "Merge: Consolidate Results",
"type": "main",
"index": 0
}
],
[
{
"node": "Merge: Consolidate Results",
"type": "main",
"index": 1
}
],
[
{
"node": "Merge: Consolidate Results",
"type": "main",
"index": 2
}
]
]
},
"Format: Convert Alert to HTML": {
"main": [
[
{
"node": "Email: Send Churn Alert Export to Sheets",
"type": "main",
"index": 0
}
]
]
},
"AI Agent: Gather Customer Data": {
"main": [
[
{
"node": "Code: Group Feature Data by Name",
"type": "main",
"index": 0
}
]
]
},
"Tool: Calculate Sentiment Score": {
"ai_tool": [
[
{
"node": "AI Agent: Gather Customer Data",
"type": "ai_tool",
"index": 0
}
]
]
},
"AI Chain: Analyze for Churn Risk": {
"main": [
[
{
"node": "Format: Convert Alert to HTML",
"type": "main",
"index": 0
}
]
]
},
"Code: Convert Sentiment to Score": {
"main": [
[
{
"node": "Code: Sum All Ticket Scores",
"type": "main",
"index": 0
}
]
]
},
"Code: Group Feature Data by Name": {
"main": [
[
{
"node": "AI Chain: Analyze for Churn Risk",
"type": "main",
"index": 0
}
]
]
},
"Code: Format Tickets for Analysis": {
"main": [
[
{
"node": "Loop: For Each Ticket",
"type": "main",
"index": 0
}
]
]
},
"Config: Set LLM for Agent & Chains": {
"ai_languageModel": [
[
{
"node": "AI: Analyze Ticket Sentiment",
"type": "ai_languageModel",
"index": 0
},
{
"node": "AI Agent: Gather Customer Data",
"type": "ai_languageModel",
"index": 0
},
{
"node": "AI Chain: Analyze for Churn Risk",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Manual Trigger: Run Churn Analysis": {
"main": [
[
{
"node": "HubSpot: Get All Deals",
"type": "main",
"index": 0
}
]
]
},
"Tool: Get Feature Usage from Sheets": {
"ai_tool": [
[
{
"node": "AI Agent: Gather Customer Data",
"type": "ai_tool",
"index": 0
}
]
]
},
"Trigger: Receive Tickets for Scoring": {
"main": [
[
{
"node": "Code: Format Tickets for Analysis",
"type": "main",
"index": 0
}
]
]
},
"Email: Send Churn Alert Export to Sheets": {
"main": [
[
{
"node": "Start Loop: For Each Deal",
"type": "main",
"index": 0
}
]
]
},
"Set: Isolate Current Deal ID Export to Sheets": {
"main": [
[
{
"node": "AI Agent: Gather Customer Data",
"type": "main",
"index": 0
}
]
]
},
"AI: Structure Agent's Findings Export to Sheets": {
"ai_outputParser": [
[
{
"node": "AI Agent: Gather Customer Data",
"type": "ai_outputParser",
"index": 0
}
]
]
}
}
}
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.
smtp
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
Built for Customer Success and Account Management teams focused on proactive retention. This workflow helps you automatically identify at-risk customers – before they churn – by combining CRM, usage, and sentiment data into one actionable alert.
Source: https://n8n.io/workflows/10199/ — 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 enables natural-language-driven automation using Bright Data's MCP tools, triggered directly by new leads in HubSpot. It dynamically extracts and executes the right tool based on lead co
Google Sheets Trigger, Output Parser Structured, Output Parser Autofixing +7
This workflow is perfect for creators, solopreneurs, and personal brands who want to consistently publish bold, high-performing content on X (Twitter) — without writing a single line themselves. After
OpenAI Chat, Memory Buffer Window, Tool Workflow +10
This workflow transforms natural language queries into research reports through a five-stage AI pipeline. When triggered via webhook (typically from Google Sheets using the companion [](https://gist.g
Redis, Agent, Output Parser Structured +7
AI Blog Publisher – Automated Blog Content Workflow This workflow is designed for individuals and teams who regularly publish content on their blog and want to automate the entire process from start t
WordPress, HTTP Request, Memory Buffer Window +9
Awesome N8N Templates. Uses agent, telegramTrigger, mcpClientTool, mcpTrigger. Event-driven trigger; 33 nodes.
Agent, Telegram Trigger, Mcp Client Tool +9