Weekly customer feedback insights AI agent
Every Monday, the agent reads last week's feedback, finds the themes, and posts a report with three action items.
How it works
Trigger: Schedule — Mondays at 9:00
- Read the Feedback sheet and keep the last 7 days.
- The agent groups comments into themes and counts them.
- It saves the top 3 problems with a suggested fix as action items.
- It posts a short report to Slack.
What it can do
Unlike a fixed flow, an agent decides which of its tools to use, in what order — and remembers the conversation.
- Understands Arabic and English feedback
- Finds themes and average rating
- Suggests practical fixes
- Tracks action items in a sheet
System prompt
The instructions that shape the agent. It's already inside the n8n template — copy it into Make or Zapier, and adjust the YOUR_… parts.
You analyse customer feedback for YOUR_BUSINESS every week. You get the last 7 days of feedback (date, channel, rating 1–5, comment), in Arabic and English.
Do this:
1. Count the feedback and work out the average rating.
2. Group the comments into 3–6 themes (e.g. delivery speed, product quality, staff, price, app). For each: how many mentions, positive or negative, and one short example quote (translated to English if needed).
3. Pick the 3 most important problems and suggest one practical fix for each.
4. Use "Save action item" once for each of those 3 problems.
5. Use "Post report" once, at the end, with the full report: totals, themes, top problems, and 2 things customers loved.
Rules:
- Use only the feedback you were given. Never invent numbers or quotes.
- Don't include customers' names, phone numbers or emails in the report.
- Keep the report under 250 words.Safety rules
Built into the prompt and the tools:
- Uses only the feedback it was given
- Leaves customers' names and numbers out of the report
Uses your own AI key (Claude, OpenAI or another model). Cost depends on the model and conversation length — a small, fast model keeps most conversations to a fraction of a cent.
Diagram
Trigger App Logic Code AI Agent tool
Code
JavaScript used in the Code step — already included in the template below.
// Feedback sheet columns: date (YYYY-MM-DD) | customer | channel | rating (1-5) | comment
// Collects the last 7 days of feedback into one list for the agent.
const since = new Date(Date.now() - 7 * 24 * 60 * 60 * 1000).toISOString().slice(0, 10);
const items = $input.all()
.map((item) => item.json)
.filter((f) => String(f.date).slice(0, 10) >= since && (f.comment || f.rating))
.map((f) => ({ date: f.date, channel: f.channel, rating: Number(f.rating) || null, comment: f.comment }));
return [{ json: { count: items.length, feedback: items } }];Build it
Pick your tool — same automation, step by step.
Copy and paste straight into an n8n canvas (Ctrl+V), or download and use Import from File. Then connect your own credentials and replace every YOUR_… placeholder. AI steps use a Claude model node — swap it for OpenAI, Gemini or Ollama if you prefer.
Show code (289 lines)
{
"name": "Weekly customer feedback insights AI agent (Flowngn)",
"nodes": [
{
"parameters": {
"content": "## Weekly customer feedback insights AI agent\nEvery Monday, the agent reads last week's feedback, finds the themes, and posts a report with three action items.\n\n**Setup**\n1. Open each node and connect your own credentials.\n2. Replace every YOUR_… placeholder.\n3. Test once, then activate.\n\nFree template from Flowngn",
"height": 260,
"width": 340
},
"id": "2799f66d-51be-489b-a647-55f3c16eff14",
"name": "Setup notes",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
-420,
-60
]
},
{
"parameters": {
"rule": {
"interval": [
{
"field": "cronExpression",
"expression": "0 9 * * 1"
}
]
}
},
"id": "2799f66d-51be-489b-a647-55f3c16eff14",
"name": "Mondays 9:00",
"type": "n8n-nodes-base.scheduleTrigger",
"typeVersion": 1.2,
"position": [
0,
0
]
},
{
"parameters": {
"operation": "read",
"documentId": {
"__rl": true,
"value": "YOUR_SPREADSHEET_ID",
"mode": "id"
},
"sheetName": {
"__rl": true,
"value": "Feedback",
"mode": "name"
},
"options": {}
},
"id": "2799f66d-51be-489b-a647-55f3c16eff14",
"name": "Read feedback",
"type": "n8n-nodes-base.googleSheets",
"typeVersion": 4.5,
"position": [
300,
0
]
},
{
"parameters": {
"jsCode": "// Feedback sheet columns: date (YYYY-MM-DD) | customer | channel | rating (1-5) | comment\n// Collects the last 7 days of feedback into one list for the agent.\nconst since = new Date(Date.now() - 7 * 24 * 60 * 60 * 1000).toISOString().slice(0, 10);\n\nconst items = $input.all()\n .map((item) => item.json)\n .filter((f) => String(f.date).slice(0, 10) >= since && (f.comment || f.rating))\n .map((f) => ({ date: f.date, channel: f.channel, rating: Number(f.rating) || null, comment: f.comment }));\n\nreturn [{ json: { count: items.length, feedback: items } }];"
},
"id": "2799f66d-51be-489b-a647-55f3c16eff14",
"name": "Last 7 days",
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
600,
0
]
},
{
"parameters": {
"conditions": {
"options": {
"caseSensitive": true,
"leftValue": "",
"typeValidation": "loose"
},
"conditions": [
{
"id": "c1",
"leftValue": "={{ $json.count }}",
"rightValue": 0,
"operator": {
"type": "number",
"operation": "gt"
}
}
],
"combinator": "and"
},
"options": {}
},
"id": "2799f66d-51be-489b-a647-55f3c16eff14",
"name": "Any feedback?",
"type": "n8n-nodes-base.if",
"typeVersion": 2.2,
"position": [
900,
0
]
},
{
"parameters": {
"promptType": "define",
"text": "=This week's feedback ({{ $json.count }} items):\n{{ JSON.stringify($json.feedback, null, 2) }}",
"options": {
"systemMessage": "=You analyse customer feedback for YOUR_BUSINESS every week. You get the last 7 days of feedback (date, channel, rating 1–5, comment), in Arabic and English.\n\nDo this:\n1. Count the feedback and work out the average rating.\n2. Group the comments into 3–6 themes (e.g. delivery speed, product quality, staff, price, app). For each: how many mentions, positive or negative, and one short example quote (translated to English if needed).\n3. Pick the 3 most important problems and suggest one practical fix for each.\n4. Use \"Save action item\" once for each of those 3 problems.\n5. Use \"Post report\" once, at the end, with the full report: totals, themes, top problems, and 2 things customers loved.\n\nRules:\n- Use only the feedback you were given. Never invent numbers or quotes.\n- Don't include customers' names, phone numbers or emails in the report.\n- Keep the report under 250 words."
}
},
"id": "2799f66d-51be-489b-a647-55f3c16eff14",
"name": "Insights agent",
"type": "@n8n/n8n-nodes-langchain.agent",
"typeVersion": 1.9,
"position": [
1200,
0
]
},
{
"parameters": {
"model": {
"__rl": true,
"value": "claude-sonnet-5",
"mode": "list"
},
"options": {}
},
"id": "2799f66d-51be-489b-a647-55f3c16eff14",
"name": "Claude model",
"type": "@n8n/n8n-nodes-langchain.lmChatAnthropic",
"typeVersion": 1.3,
"position": [
1030,
220
]
},
{
"parameters": {
"descriptionType": "manual",
"toolDescription": "Saves one problem and its suggested fix to the Action items sheet.",
"operation": "append",
"documentId": {
"__rl": true,
"value": "YOUR_SPREADSHEET_ID",
"mode": "id"
},
"sheetName": {
"__rl": true,
"value": "Action items",
"mode": "name"
},
"columns": {
"mappingMode": "defineBelow",
"value": {
"week": "={{ $now.toISODate() }}",
"problem": "={{ $fromAI('problem', \"The problem customers mentioned\", 'string') }}",
"mentions": "={{ $fromAI('mentions', \"How many times it was mentioned\", 'number') }}",
"suggested_fix": "={{ $fromAI('fix', \"One practical fix\", 'string') }}",
"status": "open"
}
},
"options": {}
},
"id": "2799f66d-51be-489b-a647-55f3c16eff14",
"name": "Save action item",
"type": "n8n-nodes-base.googleSheetsTool",
"typeVersion": 4.5,
"position": [
1200,
220
]
},
{
"parameters": {
"descriptionType": "manual",
"toolDescription": "Posts the weekly feedback report.",
"select": "channel",
"channelId": {
"__rl": true,
"value": "#customer-feedback",
"mode": "name"
},
"text": "={{ $fromAI('report', \"The full weekly report\", 'string') }}",
"otherOptions": {}
},
"id": "2799f66d-51be-489b-a647-55f3c16eff14",
"name": "Post report",
"type": "n8n-nodes-base.slackTool",
"typeVersion": 2.2,
"position": [
1370,
220
]
}
],
"connections": {
"Mondays 9:00": {
"main": [
[
{
"node": "Read feedback",
"type": "main",
"index": 0
}
]
]
},
"Read feedback": {
"main": [
[
{
"node": "Last 7 days",
"type": "main",
"index": 0
}
]
]
},
"Last 7 days": {
"main": [
[
{
"node": "Any feedback?",
"type": "main",
"index": 0
}
]
]
},
"Any feedback?": {
"main": [
[
{
"node": "Insights agent",
"type": "main",
"index": 0
}
]
]
},
"Claude model": {
"ai_languageModel": [
[
{
"node": "Insights agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Save action item": {
"ai_tool": [
[
{
"node": "Insights agent",
"type": "ai_tool",
"index": 0
}
]
]
},
"Post report": {
"ai_tool": [
[
{
"node": "Insights agent",
"type": "ai_tool",
"index": 0
}
]
]
}
},
"pinData": {},
"settings": {
"executionOrder": "v1"
},
"meta": {
"templateCredsSetupCompleted": false
},
"tags": []
}In Make, create a new scenario and add these modules in order:
- Schedule the scenario: Mondays 9:00
- Google Sheets › Search Rows
- Tools › Set variable / Iterator + Array aggregator (same logic as the Code step)
- Filter on the next route: “Any feedback?”
- If yesMake AI Agents › Run an agent — give it these tools: “Save action item” (Google Sheets), “Post report” (Slack); paste the system prompt as the agent's instructions
- Every module run uses Make operations — the free plan includes a monthly quota.
- Make's built-in functions (map, sum, formatDate…) can replace most Code steps; for complex logic use a code module.
- Module and step names can differ slightly as Make and Zapier update their apps.
In Zapier, create a new Zap and add these steps in order:
- Schedule by Zapier — Mondays 9:00
- Google Sheets — Lookup Spreadsheet Rows (or Code by Zapier)
- Code by Zapier — Run JavaScript (paste the code from the Code section)
- Filter by Zapier: “Any feedback?”
- If yesZapier Agents — create an agent (or AI by Zapier for a single step) — give it these tools: “Save action item” (Google Sheets), “Post report” (Slack); paste the system prompt as the agent's instructions
- Zaps with more than two steps need a paid Zapier plan.
- Webhooks, Filters/Paths and Code by Zapier are premium features on some Zapier plans.
- Code by Zapier runs JavaScript: reuse the code shown above, reading inputs from inputData instead of $json.
- Module and step names can differ slightly as Make and Zapier update their apps.
Tips
- Send all feedback (forms, reviews, WhatsApp) into one Feedback sheet so the agent sees everything.
Related flows
AI AgentsAdvanced
Recruiting screening AI agent
Reads every application and CV, scores it against the job's must-haves, and flags strong candidates.
Webhook → AI model → Google Sheets → Slack15h/mo
AI AgentsAdvanced
Unpaid invoices collections AI agent
Every morning, the agent reviews overdue invoices, drafts polite reminders in the right tone, and flags the ones that need a call.
AI model → Google Sheets → Gmail → Slack10h/mo
AI AgentsAdvanced
IT helpdesk AI agent
Staff describe their IT problem in a chat; the agent walks them through known fixes and logs a ticket when needed.
AI model → Google Sheets → Slack15h/mo