{
"meta": {
"instanceId": "408f9fb9940c3cb18ffdef0e0150fe342d6e655c3a9fac21f0f644e8bedabcd9"
},
"nodes": [
{
"id": "0404384b-10b6-4666-84a4-8870db30c607",
"name": "Embeddings OpenAI",
"type": "@n8n/n8n-nodes-langchain.embeddingsOpenAi",
"position": [
1220,
280
],
"parameters": {
"model": "text-embedding-3-small",
"options": {}
},
"credentials": {
"openAiApi": {
"id": "8gccIjcuf3gvaoEr",
"name": "OpenAi account"
}
},
"typeVersion": 1
},
{
"id": "a6741f04-5a5b-47a9-ac08-eb562f9f6052",
"name": "Default Data Loader",
"type": "@n8n/n8n-nodes-langchain.documentDefaultDataLoader",
"position": [
1340,
280
],
"parameters": {
"options": {
"metadata": {
"metadataValues": [
{
"name": "question",
"value": "={{ $json.question }}"
},
{
"name": "participant",
"value": "={{ $json.participant }}"
},
{
"name": "survey",
"value": "={{ $('Get Survey Results').params.documentId.cachedResultName }}"
}
]
}
},
"jsonData": "={{ $json.answer }}",
"jsonMode": "expressionData"
},
"typeVersion": 1
},
{
"id": "7663c3dd-f713-4034-bef6-0c000285f54f",
"name": "Convert to Question Answer Pairs",
"type": "n8n-nodes-base.set",
"position": [
720,
160
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "6b593ffb-ffbd-4cf5-a508-cd4f2a6d1004",
"name": "data",
"type": "array",
"value": "={{\n Object.keys($json)\n .filter(key => !['row_number', 'Participant'].includes(key))\n .map(key => ({ question: key, answer: $json[key], participant: $json.Participant }))\n}}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "84873f0c-81ce-442f-a33c-d7c6c2efa11b",
"name": "Recursive Character Text Splitter",
"type": "@n8n/n8n-nodes-langchain.textSplitterRecursiveCharacterTextSplitter",
"position": [
1340,
420
],
"parameters": {
"options": {}
},
"typeVersion": 1
},
{
"id": "da9a8ee8-5e3f-49db-8d1f-26a61ca82344",
"name": "Get Survey Results",
"type": "n8n-nodes-base.googleSheets",
"position": [
540,
160
],
"parameters": {
"options": {},
"sheetName": {
"__rl": true,
"mode": "list",
"value": "gid=0",
"cachedResultUrl": "https://docs.google.com/spreadsheets/d/1-168Vm-1kCeHkqGLAs6odha4DhPE93njfHlYIviKE50/edit#gid=0",
"cachedResultName": "Sheet1"
},
"documentId": {
"__rl": true,
"mode": "list",
"value": "1-168Vm-1kCeHkqGLAs6odha4DhPE93njfHlYIviKE50",
"cachedResultUrl": "https://docs.google.com/spreadsheets/d/1-168Vm-1kCeHkqGLAs6odha4DhPE93njfHlYIviKE50/edit?usp=drivesdk",
"cachedResultName": "Remote Working Survey Responses"
}
},
"credentials": {
"googleSheetsOAuth2Api": {
"id": "XHvC7jIRR8A2TlUl",
"name": "Google Sheets account"
}
},
"typeVersion": 4.4
},
{
"id": "4bad90b2-eefe-49c8-8caa-41cd4cb5e60f",
"name": "Get Survey Headers",
"type": "n8n-nodes-base.googleSheets",
"position": [
740,
940
],
"parameters": {
"options": {
"dataLocationOnSheet": {
"values": {
"range": "A1:Z2",
"rangeDefinition": "specifyRangeA1"
}
}
},
"sheetName": {
"__rl": true,
"mode": "id",
"value": "={{ $('Execute Workflow Trigger').first().json.sheetName }}"
},
"documentId": {
"__rl": true,
"mode": "id",
"value": "={{ $('Execute Workflow Trigger').first().json.sheetID }}"
}
},
"credentials": {
"googleSheetsOAuth2Api": {
"id": "XHvC7jIRR8A2TlUl",
"name": "Google Sheets account"
}
},
"typeVersion": 4.4
},
{
"id": "47c64994-9d1f-42ca-a849-3eeab5335b66",
"name": "Extract Questions",
"type": "n8n-nodes-base.set",
"position": [
940,
940
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "d655b165-dfa2-46cb-bc27-140399bc4227",
"name": "question",
"type": "array",
"value": "={{\n Object.keys($('Get Survey Headers').item.json)\n .filter(key => key.includes('?'))\n}}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "c237d523-b290-41ca-b323-4cc7c7f6ff37",
"name": "Questions to List",
"type": "n8n-nodes-base.splitOut",
"position": [
940,
1120
],
"parameters": {
"options": {},
"fieldToSplitOut": "question"
},
"typeVersion": 1
},
{
"id": "7f44a770-4c5d-4404-ae95-d9dee8348380",
"name": "Find All Answers",
"type": "n8n-nodes-base.httpRequest",
"position": [
1460,
1120
],
"parameters": {
"url": "=http://qdrant:6333/collections/{{ $('Set Variables').item.json.collectionName }}/points/scroll",
"method": "POST",
"options": {},
"jsonBody": "={\n \"limit\": 500,\n \"filter\":{\n \"must\": [\n {\n \"key\": \"metadata.question\",\n \"match\": { \"value\": \"{{ $('For Each Question...').item.json.question }}\" }\n },\n {\n \"key\": \"metadata.survey\",\n \"match\": { \"value\": \"{{ $('Set Variables').item.json.surveyName }}\" }\n }\n ]\n },\n \"with_vector\":true\n}",
"sendBody": true,
"specifyBody": "json",
"authentication": "predefinedCredentialType",
"nodeCredentialType": "qdrantApi"
},
"credentials": {
"qdrantApi": {
"id": "NyinAS3Pgfik66w5",
"name": "QdrantApi account"
}
},
"typeVersion": 4.2
},
{
"id": "2b6dc317-f8f3-4201-a9e1-d35ee578e79e",
"name": "Get Payload of Points",
"type": "n8n-nodes-base.httpRequest",
"position": [
2380,
800
],
"parameters": {
"url": "=http://qdrant:6333/collections/{{ $('Set Variables').first().json.collectionName }}/points",
"method": "POST",
"options": {},
"jsonBody": "={{\n {\n \"ids\": $json.points,\n \"with_payload\": true\n }\n}}",
"sendBody": true,
"specifyBody": "json",
"authentication": "predefinedCredentialType",
"nodeCredentialType": "qdrantApi"
},
"credentials": {
"qdrantApi": {
"id": "NyinAS3Pgfik66w5",
"name": "QdrantApi account"
}
},
"typeVersion": 4.2
},
{
"id": "d4a37d97-975a-4243-a7ea-81b3e30558a5",
"name": "Clusters To List",
"type": "n8n-nodes-base.splitOut",
"position": [
2180,
800
],
"parameters": {
"options": {},
"fieldToSplitOut": "output"
},
"typeVersion": 1
},
{
"id": "c78f1bf6-8390-48ee-88f4-7d1a893a8ade",
"name": "Set Variables",
"type": "n8n-nodes-base.set",
"position": [
200,
1060
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "b77c94a0-d865-4bd6-b078-a09b2ddb2a99",
"name": "collectionName",
"type": "string",
"value": "ux_survey_insights"
},
{
"id": "7b0a4d14-b5f9-4597-84c0-8cfdb363c3d3",
"name": "surveyName",
"type": "string",
"value": "={{ $json.properties.title }}"
},
{
"id": "45434b3b-3b74-4262-82e0-7ed02155caad",
"name": "insightsSheetName",
"type": "string",
"value": "=Insights-{{ $now.format('yyyyMMdd') }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "fbb1f3c3-06ad-44b5-b020-6fc3c8eda7c4",
"name": "OpenAI Chat Model",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
2560,
980
],
"parameters": {
"model": "gpt-4o-mini",
"options": {}
},
"credentials": {
"openAiApi": {
"id": "8gccIjcuf3gvaoEr",
"name": "OpenAi account"
}
},
"typeVersion": 1
},
{
"id": "83d3b413-a661-4c4c-9b8d-6ee395a15348",
"name": "Prep Output For Export",
"type": "n8n-nodes-base.set",
"position": [
3160,
1300
],
"parameters": {
"mode": "raw",
"options": {},
"jsonOutput": "={{ {\n ...$json.output,\n \"Number of Response\": $('Get Payload of Points').item.json.result.length,\n \"Participant IDs\": $('Get Payload of Points').item.json.result.map(item =>\n item.payload.metadata.participant\n ).join(','),\n \"Raw Responses\": $('Get Payload of Points').item.json.result.map(item =>\n `Participant ${item.payload.metadata.participant},${item.payload.content.replaceAll('\"', '\\\"')}`\n ).join('\\n')\n} }}\n"
},
"typeVersion": 3.4
},
{
"id": "14784dff-a8ea-4b6b-8379-b0c9051a8f98",
"name": "Export To Sheets",
"type": "n8n-nodes-base.googleSheets",
"position": [
3360,
1300
],
"parameters": {
"columns": {
"value": {},
"schema": [
{
"id": "What is your name?",
"type": "string",
"display": true,
"removed": false,
"required": false,
"displayName": "What is your name?",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "The responses indicate that two participants have the same name, 'Kwame Nkosi', which suggests a commonality in names or cultural naming traditions among the respondents. This could highlight the importance of understanding cultural context in survey responses.",
"type": "string",
"display": true,
"removed": false,
"required": false,
"displayName": "The responses indicate that two participants have the same name, 'Kwame Nkosi', which suggests a commonality in names or cultural naming traditions among the respondents. This could highlight the importance of understanding cultural context in survey responses.",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "neutral",
"type": "string",
"display": true,
"removed": false,
"required": false,
"displayName": "neutral",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "3",
"type": "string",
"display": true,
"removed": false,
"required": false,
"displayName": "3",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "77,17,54",
"type": "string",
"display": true,
"removed": false,
"required": false,
"displayName": "77,17,54",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "Participant 77,Kwame Nkosi\nParticipant 17,Kwame Nkosi\nParticipant 54,Kwame Nkansah",
"type": "string",
"display": true,
"removed": false,
"required": false,
"displayName": "Participant 77,Kwame Nkosi\nParticipant 17,Kwame Nkosi\nParticipant 54,Kwame Nkansah",
"defaultMatch": false,
"canBeUsedToMatch": true
}
],
"mappingMode": "autoMapInputData",
"matchingColumns": []
},
"options": {},
"operation": "append",
"sheetName": {
"__rl": true,
"mode": "name",
"value": "={{ $('Set Variables').first().json.insightsSheetName }}"
},
"documentId": {
"__rl": true,
"mode": "id",
"value": "={{ $('Execute Workflow Trigger').first().json.sheetID }}"
}
},
"credentials": {
"googleSheetsOAuth2Api": {
"id": "XHvC7jIRR8A2TlUl",
"name": "Google Sheets account"
}
},
"typeVersion": 4.4
},
{
"id": "779b9411-db3e-44f3-ad2a-c9d40a70580d",
"name": "Export To Sheets1",
"type": "n8n-nodes-base.googleSheets",
"position": [
2360,
1300
],
"parameters": {
"columns": {
"value": {},
"schema": [],
"mappingMode": "autoMapInputData",
"matchingColumns": []
},
"options": {},
"operation": "append",
"sheetName": {
"__rl": true,
"mode": "name",
"value": "={{ $('Set Variables').first().json.insightsSheetName }}"
},
"documentId": {
"__rl": true,
"mode": "id",
"value": "={{ $('Execute Workflow Trigger').first().json.sheetID }}"
}
},
"credentials": {
"googleSheetsOAuth2Api": {
"id": "XHvC7jIRR8A2TlUl",
"name": "Google Sheets account"
}
},
"typeVersion": 4.4
},
{
"id": "a31ab677-f57c-4b78-a290-d4a913ed4f8e",
"name": "For Each Question...",
"type": "n8n-nodes-base.splitInBatches",
"position": [
1280,
940
],
"parameters": {
"options": {}
},
"typeVersion": 3
},
{
"id": "dcfaf927-6ecd-4ebe-aee0-5fb3367b2725",
"name": "Trigger Insights",
"type": "n8n-nodes-base.executeWorkflow",
"position": [
1980,
160
],
"parameters": {
"options": {},
"workflowId": "={{ $workflow.id }}"
},
"typeVersion": 1
},
{
"id": "2579adf0-9c00-4b87-b53e-740044577ab0",
"name": "Prep Values For Trigger",
"type": "n8n-nodes-base.set",
"position": [
1800,
160
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "24dd90ad-390f-444e-ba6c-8c06a41e836e",
"name": "sheetID",
"type": "string",
"value": "={{ $('Get Survey Results').params.documentId.value }}"
},
{
"id": "90199bbb-3938-411c-a7a8-faa7ccba6059",
"name": "sheetName",
"type": "string",
"value": "={{ $('Get Survey Results').params.sheetName.value }}"
}
]
}
},
"executeOnce": true,
"typeVersion": 3.4
},
{
"id": "9b29e850-b9d0-4358-af62-92c20ab3b088",
"name": "Execute Workflow Trigger",
"type": "n8n-nodes-base.executeWorkflowTrigger",
"position": [
20,
900
],
"parameters": {},
"typeVersion": 1
},
{
"id": "70a0dcec-9f74-4af2-bd64-0ab762a77e51",
"name": "Create Insights Sheet",
"type": "n8n-nodes-base.googleSheets",
"position": [
420,
900
],
"parameters": {
"title": "={{ $('Set Variables').first().json.insightsSheetName }}",
"options": {},
"operation": "create",
"documentId": {
"__rl": true,
"mode": "id",
"value": "={{ $('Execute Workflow Trigger').first().json.sheetID }}"
}
},
"credentials": {
"googleSheetsOAuth2Api": {
"id": "XHvC7jIRR8A2TlUl",
"name": "Google Sheets account"
}
},
"typeVersion": 4.4,
"alwaysOutputData": true
},
{
"id": "f31400fb-dd7a-4c62-90ec-e9d78bbaa5e8",
"name": "Prep Values For Export",
"type": "n8n-nodes-base.set",
"position": [
2180,
1300
],
"parameters": {
"mode": "raw",
"options": {},
"jsonOutput": "={\n \"Question\": \"{{ $('For Each Question...').item.json.question }}\",\n \"Insight\": \"No Insight Found\"\n}\n"
},
"typeVersion": 3.4
},
{
"id": "506c20df-5109-422c-8c9e-0eb50fbd3ff9",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
459.27570452141345,
-42.168106366729035
],
"parameters": {
"color": 7,
"width": 617.2130261221611,
"height": 420.7389587470384,
"content": "## Step 1. Import Survey Responses\n[Read more about Google Sheets](https://docs.n8n.io/integrations/builtin/app-nodes/n8n-nodes-base.googlesheets)\n\nOur approach requires to import all participant responses as vectors with metadata linking them to the questions being answered. To do this, we'll generate questiona and answer pairs from the survey."
},
"typeVersion": 1
},
{
"id": "bddcafa8-6f54-4829-93c9-37bbb9e7edf3",
"name": "QA Pairs to List",
"type": "n8n-nodes-base.splitOut",
"position": [
900,
160
],
"parameters": {
"options": {},
"fieldToSplitOut": "data"
},
"typeVersion": 1
},
{
"id": "8d6e6bf6-c94c-43cb-a29e-5d10207cb8bd",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
1100,
-102.05898437632061
],
"parameters": {
"color": 7,
"width": 563.8350682199533,
"height": 678.1641960508446,
"content": "## Step 2. Vectorize Each Response Into Qdrant\n[Read more about using Qdrant](https://docs.n8n.io/integrations/builtin/cluster-nodes/root-nodes/n8n-nodes-langchain.vectorstoreqdrant)\n\nSpecial attention is given to how metadata is captured as it becomes key to this workflow is being able to retrieve subsets of the data for analysis."
},
"typeVersion": 1
},
{
"id": "613d4a32-a87a-423e-a1d1-ee23db0de6d1",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"position": [
1680,
-30.440883940004255
],
"parameters": {
"color": 7,
"width": 519.6419932444072,
"height": 429.11782776909047,
"content": "## Step 3. Trigger Insights SubWorkflow\n[Learn more about Workflow Triggers](https://docs.n8n.io/integrations/builtin/core-nodes/n8n-nodes-base.executeworkflow)\n\nA subworkflow is used to trigger the analysis for the survey. This separation is optional but used here to better demonstrate the two part process."
},
"typeVersion": 1
},
{
"id": "1e858e4a-b91b-4411-8e2a-6eb76647b796",
"name": "Sticky Note3",
"type": "n8n-nodes-base.stickyNote",
"position": [
-57.47778952966382,
710.393394209128
],
"parameters": {
"color": 7,
"width": 668.3083616841852,
"height": 528.2386658883073,
"content": "## Step 4. Create Insights Sheet\n[Learn more about Workflow Triggers](https://docs.n8n.io/integrations/builtin/core-nodes/n8n-nodes-base.executeworkflowtrigger)\n\nTo capture the generated insights, we'll create a new unique sheet within the survey spreadsheet. This is optional and you may want to capture in other datastores depending on your needs."
},
"typeVersion": 1
},
{
"id": "9170c566-07d3-49dc-aafb-2dbe79940d2c",
"name": "Sticky Note4",
"type": "n8n-nodes-base.stickyNote",
"position": [
640,
683.5153164275844
],
"parameters": {
"color": 7,
"width": 536.9288458983389,
"height": 622.1362463986454,
"content": "## Step 5. Get List Of Questions From Survey\n[Read more about using Google Sheets](https://docs.n8n.io/integrations/builtin/app-nodes/n8n-nodes-base.googlesheets)\n\nNext we'll fetch the survey for metadata and questions, splitting them into separate workflow items. Our intention is to process each question end-to-end before moving to the next. This approach is a little \"safer\" in the scenario where an interruption occurs we won't lose all our work."
},
"typeVersion": 1
},
{
"id": "8488df77-055d-41cc-94f1-92ac5d54ef10",
"name": "Sticky Note5",
"type": "n8n-nodes-base.stickyNote",
"position": [
1200,
673.291535602609
],
"parameters": {
"color": 7,
"width": 823.147012265536,
"height": 868.2579789328703,
"content": "## Step 6. Find Groups of Similar Answers For Each Question\n[Learn more about using the Code Node](https://docs.n8n.io/integrations/builtin/core-nodes/n8n-nodes-base.code/)\n\nGiving all the responses to an LLM to analyse is the common but naive approach; the summarisation is usually too high level for real insights and loses a lot of detail such as the number and identity of respondants. What we want to do instead is find and group popular answers for each question to ensure all perspectives are considered.\n\nOur approach does this by mapping our answer vectors to a 2D grid and then identifying where the vector points are \"clustered\"; where a group of points are within close proximity to each other."
},
"typeVersion": 1
},
{
"id": "f4748b6d-5bd8-48cf-942f-3a0dc681078d",
"name": "Sticky Note6",
"type": "n8n-nodes-base.stickyNote",
"position": [
2060,
1180
],
"parameters": {
"color": 7,
"width": 536.9288458983389,
"height": 359.90385684071794,
"content": "## Step 7b. Skip If No Clusters Found\nWhere no clusters were found, it means the answers were unique enough to not show any pattern. eg. \"What's you name?\""
},
"typeVersion": 1
},
{
"id": "d55d6a47-da8c-46ae-bd10-0eb671dcd121",
"name": "Sticky Note7",
"type": "n8n-nodes-base.stickyNote",
"position": [
2060,
611.6915003841909
],
"parameters": {
"color": 7,
"width": 871.451300407926,
"height": 541.1135860445843,
"content": "## Step 7a. Summarise the Top Groups of Similar Answers\n[Read more about using the Information Extractor Node](https://docs.n8n.io/integrations/builtin/cluster-nodes/root-nodes/n8n-nodes-langchain.information-extractor)\n\nEach discovered cluster will return a reference vector which is used to fetch all related answers in the group.\nThe group is then sent to the LLM to summarise as well as assign a sentiment score."
},
"typeVersion": 1
},
{
"id": "e5d5f88f-5832-43fc-a5b9-f747d08e7e77",
"name": "Sticky Note8",
"type": "n8n-nodes-base.stickyNote",
"position": [
2620,
1180
],
"parameters": {
"color": 7,
"width": 924.2798021207429,
"height": 363.07347551845976,
"content": "## Step 8. Write To Insights Sheet\nFinally, our completed insights to appended to\nthe Insights Sheet we created earlier in the workflow."
},
"typeVersion": 1
},
{
"id": "49ac1504-7b43-4fa1-b4ce-15c7a53c9018",
"name": "Sticky Note9",
"type": "n8n-nodes-base.stickyNote",
"position": [
460,
400
],
"parameters": {
"color": 5,
"width": 323.2987132716669,
"height": 80,
"content": "### Run this once! \nIf for any reason you need to run more than once, be sure to clear the existing data first."
},
"typeVersion": 1
},
{
"id": "450f89c5-ef0f-4bf8-8db9-6347247c7f4d",
"name": "Has Clusters?",
"type": "n8n-nodes-base.if",
"position": [
1820,
1120
],
"parameters": {
"options": {},
"conditions": {
"options": {
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "40b6bb62-a2d6-4422-8fbb-7ae49898bad9",
"operator": {
"type": "array",
"operation": "notEmpty",
"singleValue": true
},
"leftValue": "={{ $json.output }}",
"rightValue": ""
}
]
}
},
"typeVersion": 2
},
{
"id": "1652a108-8fb8-4229-a76d-abf9fbcff626",
"name": "Sticky Note10",
"type": "n8n-nodes-base.stickyNote",
"position": [
20,
-400
],
"parameters": {
"width": 400.381109509268,
"height": 679.5610243514676,
"content": "## Try It Out!\n\n### This workflow generates highly-detailed insights from survey responses. Works best when dealing with a large number of participants.\n\n* Import survey responses and vectorise in Qdrant vectorstore.\n* Identify clusters of popular responses to questions using K-means clustering algorithm. \n* Each valid cluster is analysed and summarised by LLM.\n* Export LLM response and cluster results back into sheet.\n\nCheck out the reference google sheet here: https://docs.google.com/spreadsheets/d/e/2PACX-1vT6m8XH8JWJTUAfwojc68NAUGC7q0lO7iV738J7aO5fuVjiVzdTRRPkMmT1C4N8TwejaiT0XrmF1Q48/pubhtml\n\n### Need Help?\nJoin the [Discord](https://discord.com/invite/XPKeKXeB7d) or ask in the [Forum](https://community.n8n.io/)!\n\nHappy Hacking!"
},
"typeVersion": 1
},
{
"id": "6eef981e-b2ce-433c-b71f-78be64812a56",
"name": "Sticky Note11",
"type": "n8n-nodes-base.stickyNote",
"position": [
1260,
1340
],
"parameters": {
"color": 5,
"width": 323.2987132716669,
"height": 110.05160146874424,
"content": "### First Time Running?\nThere is a slight delay on first run because the code node has to download the required packages."
},
"typeVersion": 1
},
{
"id": "fa0c14be-03f4-4ed2-bd60-e93817382ded",
"name": "When clicking ‘Test workflow’",
"type": "n8n-nodes-base.manualTrigger",
"position": [
240,
100
],
"parameters": {},
"typeVersion": 1
},
{
"id": "30323019-59ba-4a19-a46e-196d469f097d",
"name": "Get Sheet Details",
"type": "n8n-nodes-base.httpRequest",
"position": [
200,
900
],
"parameters": {
"url": "=https://sheets.googleapis.com/v4/spreadsheets/{{ $json.sheetID }}",
"options": {},
"authentication": "predefinedCredentialType",
"nodeCredentialType": "googleSheetsOAuth2Api"
},
"credentials": {
"googleSheetsOAuth2Api": {
"id": "XHvC7jIRR8A2TlUl",
"name": "Google Sheets account"
}
},
"typeVersion": 4.2
},
{
"id": "6ced8012-1dd3-4da3-8c27-e4f4dfc959f6",
"name": "Only Clusters With 3+ points",
"type": "n8n-nodes-base.filter",
"position": [
2180,
960
],
"parameters": {
"options": {},
"conditions": {
"options": {
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "328f806c-0792-4d90-9bee-a1e10049e78f",
"operator": {
"type": "array",
"operation": "lengthGt",
"rightType": "number"
},
"leftValue": "={{ $json.points }}",
"rightValue": 2
}
]
}
},
"typeVersion": 2
},
{
"id": "8ae81a55-75e2-40a3-bef6-0935ff08128f",
"name": "Apply K-means Clustering Algorithm",
"type": "n8n-nodes-base.code",
"position": [
1640,
1120
],
"parameters": {
"language": "python",
"pythonCode": "import numpy as np\nfrom sklearn.cluster import KMeans\n\n# get vectors for all answers\npoint_ids = [item.id for item in _input.first().json.result.points]\nvectors = [item.vector.to_py() for item in _input.first().json.result.points]\nvectors_array = np.array(vectors)\n\n# apply k-means clustering where n_clusters = 10\n# this is a max and we'll discard some of these clusters later\nkmeans = KMeans(n_clusters=min(len(vectors), 10), random_state=42).fit(vectors_array)\nlabels = kmeans.labels_\nunique_labels = set(labels)\n\n# Extract and print points in each cluster\nclusters = {}\nfor label in set(labels):\n clusters[label] = vectors_array[labels == label]\n\n# return Qdrant point ids for each cluster\n# we'll use these ids to fetch the payloads from the vector store.\noutput = []\nfor cluster_id, cluster_points in clusters.items():\n points = [point_ids[i] for i in range(len(labels)) if labels[i] == cluster_id]\n output.append({\n \"id\": f\"Cluster {cluster_id}\",\n \"total\": len(cluster_points),\n \"points\": points\n })\n\nreturn {\"json\": {\"output\": output } }"
},
"typeVersion": 2
},
{
"id": "cbb42384-d46b-471f-a7d8-27e3de042492",
"name": "Qdrant Vector Store",
"type": "@n8n/n8n-nodes-langchain.vectorStoreQdrant",
"position": [
1220,
100
],
"parameters": {
"mode": "insert",
"options": {},
"qdrantCollection": {
"__rl": true,
"mode": "list",
"value": "ux_survey_insights",
"cachedResultName": "ux_survey_insights"
}
},
"credentials": {
"qdrantApi": {
"id": "NyinAS3Pgfik66w5",
"name": "QdrantApi account"
}
},
"typeVersion": 1
},
{
"id": "17584901-15d6-421f-ad69-3ba872276055",
"name": "Survey Insights Agent",
"type": "@n8n/n8n-nodes-langchain.informationExtractor",
"position": [
2580,
800
],
"parameters": {
"text": "=The {{ $json.result.length }} participant responses were:\n{{\n$json.result.map(item =>\n`* Participant ${item.payload.metadata.participant}: \"${item.payload.content.replaceAll('\"', '\\\"')}\"`\n).join('\\n')\n}}",
"options": {
"systemPromptTemplate": "=You help summarise a selection of participant responses to a specific question for a survey called \"{{ $json.result[0].payload.metadata.survey }}\".\nThe question asked was \"{{ $json.result[0].payload.metadata.question }}\".\nThe {{ $json.result.length }} participant responses were selected because their answers were similar in context.\n\nYour task is to: \n* summarise the given participant responses into a short paragraph. Provide an insight from this summary and what we could learn from the answers.\n* determine if the overall sentiment of all the listed responses to be either negative, mildy negative, neutral, mildy positive or positive."
},
"schemaType": "fromJson",
"jsonSchemaExample": "{\n\t\"Question\": \"What do you enjoy most about working remotely, and why?\",\n\t\"Insight\": \"\",\n \"Sentiment\": \"Positive\"\n}"
},
"typeVersion": 1
}
],
"pinData": {},
"connections": {
"Has Clusters?": {
"main": [
[
{
"node": "Clusters To List",
"type": "main",
"index": 0
}
],
[
{
"node": "Prep Values For Export",
"type": "main",
"index": 0
}
]
]
},
"Set Variables": {
"main": [
[
{
"node": "Create Insights Sheet",
"type": "main",
"index": 0
}
]
]
},
"Clusters To List": {
"main": [
[
{
"node": "Only Clusters With 3+ points",
"type": "main",
"index": 0
}
]
]
},
"Export To Sheets": {
"main": [
[
{
"node": "For Each Question...",
"type": "main",
"index": 0
}
]
]
},
"Find All Answers": {
"main": [
[
{
"node": "Apply K-means Clustering Algorithm",
"type": "main",
"index": 0
}
]
]
},
"QA Pairs to List": {
"main": [
[
{
"node": "Qdrant Vector Store",
"type": "main",
"index": 0
}
]
]
},
"Embeddings OpenAI": {
"ai_embedding": [
[
{
"node": "Qdrant Vector Store",
"type": "ai_embedding",
"index": 0
}
]
]
},
"Export To Sheets1": {
"main": [
[
{
"node": "For Each Question...",
"type": "main",
"index": 0
}
]
]
},
"Extract Questions": {
"main": [
[
{
"node": "Questions to List",
"type": "main",
"index": 0
}
]
]
},
"Get Sheet Details": {
"main": [
[
{
"node": "Set Variables",
"type": "main",
"index": 0
}
]
]
},
"OpenAI Chat Model": {
"ai_languageModel": [
[
{
"node": "Survey Insights Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Questions to List": {
"main": [
[
{
"node": "For Each Question...",
"type": "main",
"index": 0
}
]
]
},
"Get Survey Headers": {
"main": [
[
{
"node": "Extract Questions",
"type": "main",
"index": 0
}
]
]
},
"Get Survey Results": {
"main": [
[
{
"node": "Convert to Question Answer Pairs",
"type": "main",
"index": 0
}
]
]
},
"Default Data Loader": {
"ai_document": [
[
{
"node": "Qdrant Vector Store",
"type": "ai_document",
"index": 0
}
]
]
},
"Qdrant Vector Store": {
"main": [
[
{
"node": "Prep Values For Trigger",
"type": "main",
"index": 0
}
]
]
},
"For Each Question...": {
"main": [
null,
[
{
"node": "Find All Answers",
"type": "main",
"index": 0
}
]
]
},
"Create Insights Sheet": {
"main": [
[
{
"node": "Get Survey Headers",
"type": "main",
"index": 0
}
]
]
},
"Get Payload of Points": {
"main": [
[
{
"node": "Survey Insights Agent",
"type": "main",
"index": 0
}
]
]
},
"Survey Insights Agent": {
"main": [
[
{
"node": "Prep Output For Export",
"type": "main",
"index": 0
}
]
]
},
"Prep Output For Export": {
"main": [
[
{
"node": "Export To Sheets",
"type": "main",
"index": 0
}
]
]
},
"Prep Values For Export": {
"main": [
[
{
"node": "Export To Sheets1",
"type": "main",
"index": 0
}
]
]
},
"Prep Values For Trigger": {
"main": [
[
{
"node": "Trigger Insights",
"type": "main",
"index": 0
}
]
]
},
"Execute Workflow Trigger": {
"main": [
[
{
"node": "Get Sheet Details",
"type": "main",
"index": 0
}
]
]
},
"Only Clusters With 3+ points": {
"main": [
[
{
"node": "Get Payload of Points",
"type": "main",
"index": 0
}
]
]
},
"Convert to Question Answer Pairs": {
"main": [
[
{
"node": "QA Pairs to List",
"type": "main",
"index": 0
}
]
]
},
"Recursive Character Text Splitter": {
"ai_textSplitter": [
[
{
"node": "Default Data Loader",
"type": "ai_textSplitter",
"index": 0
}
]
]
},
"When clicking ‘Test workflow’": {
"main": [
[
{
"node": "Get Survey Results",
"type": "main",
"index": 0
}
]
]
},
"Apply K-means Clustering Algorithm": {
"main": [
[
{
"node": "Has Clusters?",
"type": "main",
"index": 0
}
]
]
}
}
}