Amazon Ads AI Optimization
工作流概述
这是一个包含22个节点的复杂工作流,主要用于自动化处理各种任务。
工作流源代码
{
"id": "Agn9dzf5YTqcmQGN",
"meta": {
"instanceId": "8029058e18ae4ed6081000c1270d96039ad05959052aa2034dd96a215849bcf7",
"templateCredsSetupCompleted": true
},
"name": "Amazon Ads AI Optimization",
"tags": [
{
"id": "vjZ7QzTW2i7StzqX",
"name": "AI Flow",
"createdAt": "2025-04-10T00:32:55.235Z",
"updatedAt": "2025-04-10T00:32:55.235Z"
}
],
"nodes": [
{
"id": "0286c917-d771-4835-a5f8-71f79a5e59e8",
"name": "List Files",
"type": "n8n-nodes-base.googleDrive",
"position": [
-100,
-800
],
"parameters": {
"filter": {
"folderId": {
"__rl": true,
"mode": "list",
"value": "",
"cachedResultUrl": "",
"cachedResultName": "<choose report folder>"
}
},
"options": {},
"resource": "fileFolder",
"searchMethod": "query"
},
"credentials": {
"googleDriveOAuth2Api": {
"id": "UPKjIF2z8RkkmP21",
"name": "Google Drive account"
}
},
"typeVersion": 3
},
{
"id": "7d9b0c0a-86ee-4aae-8d73-66f409b0a57f",
"name": "OpenAI Chat Model",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
1620,
-540
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-4o",
"cachedResultName": "gpt-4o"
},
"options": {}
},
"credentials": {
"openAiApi": {
"id": "qszlkCg3ypMJEWvt",
"name": "OpenAi account"
}
},
"typeVersion": 1.2
},
{
"id": "d3d58b0a-3107-4525-92a8-d54332e9a8a5",
"name": "is XLSX",
"type": "n8n-nodes-base.if",
"position": [
540,
-800
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "820b48a1-676d-400b-894f-3b3a5203eca7",
"operator": {
"type": "string",
"operation": "contains"
},
"leftValue": "={{ $json.name }}",
"rightValue": ".xlsx"
}
]
}
},
"typeVersion": 2.2
},
{
"id": "884e4a08-3b19-4485-aba7-c69887607b82",
"name": "Get File",
"type": "n8n-nodes-base.googleDrive",
"position": [
100,
-800
],
"parameters": {
"fileId": {
"__rl": true,
"mode": "id",
"value": "={{ $json.id }}"
},
"options": {
"binaryPropertyName": "data",
"googleFileConversion": {
"conversion": {}
}
},
"operation": "download"
},
"credentials": {
"googleDriveOAuth2Api": {
"id": "UPKjIF2z8RkkmP21",
"name": "Google Drive account"
}
},
"typeVersion": 3
},
{
"id": "c72fde38-de38-4734-a7e8-aa70e8638cad",
"name": "Merge XLSX and CSV",
"type": "n8n-nodes-base.merge",
"position": [
1200,
-800
],
"parameters": {},
"typeVersion": 3.1
},
{
"id": "cd23e23c-9bb7-4b8d-90ab-8917783cf1ab",
"name": "Format Data",
"type": "n8n-nodes-base.code",
"position": [
1420,
-800
],
"parameters": {
"jsCode": "const result = {};
for (const item of items) {
const fileName = item.json.fileName || item.json.name || 'unknown_file';
const baseName = fileName
.split('.')[0]
.replace(/\s+/g, '_')
.toLowerCase()
.replace(/\s*\(\d+\)$/, '')
.replace(/_+$/, '')
.trim();
// regex → result key
const map = [
{ key: 'search_terms', regex: /search_term/ },
{ key: 'campaigns', regex: /campaign/ },
{ key: 'targeting', regex: /targeting/ },
{ key: 'placement', regex: /placement/ },
{ key: 'budgets', regex: /budget/ },
];
const entry = map.find(m => m.regex.test(baseName));
const mappedKey = entry ? entry.key : null;
console.log('fileName:', fileName);
console.log('baseName:', baseName);
console.log('mappedKey:', mappedKey);
if (!mappedKey) {
throw new Error(`${fileName} → ${baseName} → Unrecognized file name structure`);
}
result[mappedKey] = result[mappedKey] || [];
result[mappedKey].push(item.json);
}
return [{ json: result }];
"
},
"typeVersion": 2
},
{
"id": "02172577-d867-45a4-96ea-eb105169deff",
"name": "Set fileName",
"type": "n8n-nodes-base.set",
"position": [
320,
-800
],
"parameters": {
"options": {
"dotNotation": true,
"ignoreConversionErrors": false
},
"assignments": {
"assignments": [
{
"id": "a467fabb-d7d0-482d-8a6a-afcd97cc0d8c",
"name": "fileName",
"type": "string",
"value": "={{ $json.name }}"
}
]
},
"includeOtherFields": true
},
"typeVersion": 3.4
},
{
"id": "31db008f-20e4-4fe3-a9d0-1815b3802690",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"position": [
-140,
-1040
],
"parameters": {
"color": 3,
"width": 180,
"height": 200,
"content": "## Change
Choose the \"folder\" in the filter options to the folder containing your Ad reports
"
},
"typeVersion": 1
},
{
"id": "0ba8c273-8369-4009-9b93-b0fb243a3c85",
"name": "Sticky Note3",
"type": "n8n-nodes-base.stickyNote",
"position": [
1640,
-1000
],
"parameters": {
"width": 260,
"content": "## AI Analysis
Uses GPT-4o to process the bundled reports and generate optimization instructions.
Passes system instructions and cleaned data as input."
},
"typeVersion": 1
},
{
"id": "451bb016-1766-4688-aafc-75937e0d5c3f",
"name": "Sticky Note5",
"type": "n8n-nodes-base.stickyNote",
"position": [
-660,
-580
],
"parameters": {
"width": 540,
"height": 700,
"content": "## Amazon Ads Report Scheduling Instructions
To run this workflow, schedule the following Sponsored Products reports in the Amazon Ads Console:
Use \"Detailed\" for:
Search Term Report → Sponsored_Products_Search_Term_Detailed_L30
Targeting Report → Sponsored_Products_Targeting_Detailed_L30
Use \"Summary\" for:
Campaign Report → Sponsored_Products_Campaign_L30
Placement Report → Sponsored_Products_Placement_L30
Budget Report → Sponsored_Products_Budget_L30
Shared settings for all reports:
Date Range: Last 30 Days
Frequency: Daily
Format: .xlsx or .csv
Delivery: Email + Console Download
Make sure filenames match expectations so the workflow can route them correctly."
},
"typeVersion": 1
},
{
"id": "a671a4f1-05b0-4d7c-9cc1-8c2838593e34",
"name": "Sticky Note6",
"type": "n8n-nodes-base.stickyNote",
"position": [
-60,
-580
],
"parameters": {
"width": 400,
"height": 520,
"content": "## Report Delivery
How to get reports into Google Drive
Use one of the following:
📥 Manual Upload – Download emailed reports and move them to your Drive folder
🤖 Automation – Use n8n to watch Gmail for no-reply@amazon.com, extract attachments, and upload to Drive
💻 Drive Sync Folder – Use a local folder synced to Google Drive with rules for report types
Reports must match expected filenames so the flow can identify and classify them."
},
"typeVersion": 1
},
{
"id": "63a7f391-2bc7-41f9-a53f-e742950c60bf",
"name": "Sticky Note7",
"type": "n8n-nodes-base.stickyNote",
"position": [
360,
-580
],
"parameters": {
"width": 360,
"height": 520,
"content": "## Upgrade! 🚀
Apply for an Amazon Advertising API developer account to unlock full automation:
Generate reports programmatically via the Reports API
Fetch report files directly into n8n using HTTP or custom nodes
Eliminate email + Drive dependency entirely
🔗 https://advertising.amazon.com/API/docs/en-us/
Once approved, you can schedule report generation and download all required data securely and automatically.
**Double click** to edit me. [Guide](https://docs.n8n.io/workflows/sticky-notes/)"
},
"typeVersion": 1
},
{
"id": "e5a24705-0ad5-4629-b183-d279bdca8b29",
"name": "Preserve File Name",
"type": "n8n-nodes-base.set",
"position": [
980,
-900
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "d6883fe9-d04f-4c86-bc9a-f4dd526afca2",
"name": "fileName",
"type": "string",
"value": "={{ $('is XLSX').item.json.fileName }}"
}
]
},
"includeOtherFields": true
},
"typeVersion": 3.4
},
{
"id": "3c315a0c-a89e-490a-9a82-e3d96d2b94c7",
"name": "Email Optimizations",
"type": "n8n-nodes-base.gmail",
"position": [
2016,
-800
],
"webhookId": "b9d7c1a9-a1a3-4b97-97c9-a272f0e97127",
"parameters": {
"sendTo": "={{ $('Email Options').first().json.send_to }}",
"message": "={{
(() => {
let raw = $node[\"AI Analyze\"].json[\"text\"];
// 🔧 Remove triple backticks and optional \"json\" tag
raw = raw.replace(/^```json\s*/i, \"\").replace(/```$/, \"\").trim();
let data;
try {
data = JSON.parse(raw);
} catch (err) {
return `<p><strong>❌ Failed to parse AI output.</strong><br>${err.message}</p>`;
}
let msg = \"<h2>Amazon Ads Optimization Instructions</h2>\";
// Optional Summary Totals
const totalSpend = (data.campaign_adjustments || []).reduce((sum, c) => sum + (c.projected_daily_spend_usd || 0), 0);
const totalSales = (data.campaign_adjustments || []).reduce((sum, c) => sum + (c.projected_daily_sales_usd || 0), 0);
msg += `<p><strong>Total Budget Increase Recommended:</strong><br>`;
msg += `Estimated daily spend: <strong>$${totalSpend.toFixed(2)}</strong><br>`;
msg += `Estimated daily sales: <strong>$${totalSales.toFixed(2)}</strong></p>`;
// Campaign Adjustments
msg += \"<h3>Campaign Adjustments:</h3><ul>\";
(data.campaign_adjustments || []).forEach(c => {
msg += `<li><strong>${c.campaign_name}</strong><ul>`;
if (c.default_bid_multiplier !== undefined) {
const percent = Math.round((1 - c.default_bid_multiplier) * 100);
msg += `<li>Default bid × ${c.default_bid_multiplier} (<em>–${percent}%</em>)</li>`;
}
if (c.bid_adjustments) {
msg += \"<li>Bid adjustments:<ul>\";
msg += `<li>Top of Search: ${c.bid_adjustments.top_of_search ?? 0}%</li>`;
msg += `<li>Rest of Search: ${c.bid_adjustments.rest_of_search ?? 0}%</li>`;
msg += `<li>Product pages: ${c.bid_adjustments.product_pages ?? 0}%</li>`;
msg += \"</ul></li>\";
}
if (c.budget_change?.action !== \"none\") {
msg += `<li>Budget: ${c.budget_change.action} by ${c.budget_change.percent}%</li>`;
}
if (c.projected_daily_spend_usd && c.projected_daily_sales_usd) {
msg += `<li>Est. daily spend: $${c.projected_daily_spend_usd.toFixed(2)}</li>`;
msg += `<li>Est. daily sales: $${c.projected_daily_sales_usd.toFixed(2)}</li>`;
if (c.estimated_acos_percent !== undefined) {
msg += `<li>ACoS: ${c.estimated_acos_percent}%</li>`;
}
if (c.estimated_roas_multiple !== undefined) {
const color = c.estimated_roas_multiple < 1.0 ? 'red' : 'green';
msg += `<li>ROAS: <span style=\"color:${color}\">${c.estimated_roas_multiple.toFixed(2)}x</span></li>`;
}
}
msg += \"</ul></li>\";
});
msg += \"</ul>\";
// Keyword Recommendations
if ((data.keyword_recommendations?.add_exact?.length || 0) > 0 ||
(data.keyword_recommendations?.negative?.length || 0) > 0) {
msg += \"<h3>Keyword Recommendations:</h3><ul>\";
(data.keyword_recommendations.add_exact || []).forEach(k => {
msg += `<li>Add exact: \"<strong>${k.term}</strong>\" in <em>${k.campaign_name} / ${k.ad_group_name}</em> at <strong>$${k.suggested_bid}</strong></li>`;
});
(data.keyword_recommendations.negative || []).forEach(n => {
if (typeof n === 'string') {
msg += `<li>Negative: \"<strong>${n}</strong>\"</li>`;
} else {
msg += `<li>Negative: \"<strong>${n.term}</strong>\" in <em>${n.campaign_name || 'Unspecified Campaign'}</em></li>`;
}
});
msg += \"</ul>\";
}
// Targeting Recommendations
if ((data.targeting_recommendations || []).length > 0) {
msg += \"<h3>Targeting Recommendations:</h3><ul>\";
data.targeting_recommendations.forEach(t => {
const valueText = t.value ? ` by ${t.value}` : \"\";
msg += `<li>${t.target} in <em>${t.campaign_name} / ${t.ad_group_name}</em>: <strong>${t.action}</strong>${valueText}</li>`;
});
msg += \"</ul>\";
}
return msg;
})()
}}
",
"options": {},
"subject": "={{ $('Email Options').first().json.subject }}"
},
"credentials": {
"gmailOAuth2": {
"id": "6m7O3IpXy4mCRogW",
"name": "Brian Gmail"
}
},
"typeVersion": 2.1
},
{
"id": "f4fc0a70-2df9-4b7b-b60c-856b1b74ead7",
"name": "Extract XLSX Data",
"type": "n8n-nodes-base.extractFromFile",
"position": [
760,
-900
],
"parameters": {
"options": {},
"operation": "xlsx"
},
"typeVersion": 1
},
{
"id": "d0618a5b-1995-474d-a969-38e856b1b91a",
"name": "Extract CSV Data",
"type": "n8n-nodes-base.extractFromFile",
"position": [
760,
-700
],
"parameters": {
"options": {},
"binaryPropertyName": "=data"
},
"typeVersion": 1
},
{
"id": "67f9d0a2-2f34-416a-bc11-ef776e6e4ab3",
"name": "Preserve CSV File Name",
"type": "n8n-nodes-base.set",
"position": [
980,
-700
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "d6883fe9-d04f-4c86-bc9a-f4dd526afca2",
"name": "fileName",
"type": "string",
"value": "={{ $('is XLSX').item.json.fileName }}"
}
]
},
"includeOtherFields": true
},
"typeVersion": 3.4
},
{
"id": "818205c9-0fe9-4fe6-8556-657f087ba7b9",
"name": "When clicking ‘Test workflow’",
"type": "n8n-nodes-base.manualTrigger",
"position": [
-500,
-800
],
"parameters": {},
"typeVersion": 1
},
{
"id": "1612753d-0b7f-4ae5-9ec0-8ad39f1003b1",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
-580,
-1040
],
"parameters": {
"width": 220,
"content": "## Trigger
You may replace this with a scheduled event or poll the folder for changes."
},
"typeVersion": 1
},
{
"id": "158da856-b682-4f98-afcc-4fa12b978db0",
"name": "Email Options",
"type": "n8n-nodes-base.set",
"position": [
-300,
-800
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "60c2189a-2ca3-43ac-bffc-371bbc3c123b",
"name": "send_to",
"type": "string",
"value": "<enter send to email address>"
},
{
"id": "c6f588b3-b8b9-4a83-817b-a68de36d2570",
"name": "subject",
"type": "string",
"value": "<enter the email subject for report emails>"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "4f1f251e-5cfb-468d-9531-9c2ba2c875f6",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
-320,
-1040
],
"parameters": {
"color": 3,
"width": 160,
"content": "## Change!
Edit these email options."
},
"typeVersion": 1
},
{
"id": "ca2f4a7c-5aa9-4f6a-bc04-aedce5e0aaed",
"name": "AI Analyze",
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"position": [
1640,
-800
],
"parameters": {
"text": "={{JSON.stringify($json)}}",
"messages": {
"messageValues": [
{
"message": "You are an Amazon Ads Optimization Assistant. You will receive five structured datasets from Sponsored Products reports:
- search_terms
- campaigns
- targeting
- placement
- budgets
Your goal is to generate precise performance recommendations for bid strategy, targeting, and budget scaling.
---
1. Campaign Adjustments:
For each campaign, return:
- campaign_name (string)
- default_bid_multiplier (float, optional — only if bid should change)
- bid_adjustments: { top_of_search, rest_of_search, product_pages } (percentages)
- budget_change: { action: increase | decrease | none, percent: float }
- projected_daily_spend_usd (float)
- projected_daily_sales_usd (float)
- estimated_acos_percent (float)
- estimated_roas_multiple (float)
Base projections on historical 30-day data. If a budget increase is recommended, scale projected spend and sales proportionally. Return NaN only if data is insufficient.
---
2. Keyword Recommendations:
Recommend at least 5 exact-match keywords to add. Each must include:
- term
- campaign_name
- ad_group_name
- suggested_bid (USD)
Also return at least 3 negative keywords:
- { term: \"...\", campaign_name?: \"...\" }
Do not return keyword recommendations that lack campaign and ad group names.
---
3. Targeting Recommendations:
Recommend at least 3 targets to pause or increase bids. Return:
- target (ASIN, keyword, or match group)
- campaign_name
- ad_group_name
- action: \"pause\" or \"increase_bid\"
- value: float (if increasing bid)
---
Respond ONLY with a JSON object in this exact format. Do NOT include backticks, code blocks, or explanations:
{
\"campaign_adjustments\": [...],
\"keyword_recommendations\": {
\"add_exact\": [...],
\"negative\": [...]
},
\"targeting_recommendations\": [...]
}
"
}
]
},
"promptType": "define"
},
"typeVersion": 1.6
}
],
"active": false,
"pinData": {},
"settings": {
"executionOrder": "v1"
},
"versionId": "286aae2a-f8df-489d-9f03-89d0b50b1800",
"connections": {
"is XLSX": {
"main": [
[
{
"node": "Extract XLSX Data",
"type": "main",
"index": 0
}
],
[
{
"node": "Extract CSV Data",
"type": "main",
"index": 0
}
]
]
},
"Get File": {
"main": [
[
{
"node": "Set fileName",
"type": "main",
"index": 0
}
]
]
},
"AI Analyze": {
"main": [
[
{
"node": "Email Optimizations",
"type": "main",
"index": 0
}
]
]
},
"List Files": {
"main": [
[
{
"node": "Get File",
"type": "main",
"index": 0
}
]
]
},
"Format Data": {
"main": [
[
{
"node": "AI Analyze",
"type": "main",
"index": 0
}
]
]
},
"Set fileName": {
"main": [
[
{
"node": "is XLSX",
"type": "main",
"index": 0
}
]
]
},
"Email Options": {
"main": [
[
{
"node": "List Files",
"type": "main",
"index": 0
}
]
]
},
"Extract CSV Data": {
"main": [
[
{
"node": "Preserve CSV File Name",
"type": "main",
"index": 0
}
]
]
},
"Extract XLSX Data": {
"main": [
[
{
"node": "Preserve File Name",
"type": "main",
"index": 0
}
]
]
},
"OpenAI Chat Model": {
"ai_languageModel": [
[
{
"node": "AI Analyze",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Merge XLSX and CSV": {
"main": [
[
{
"node": "Format Data",
"type": "main",
"index": 0
}
]
]
},
"Preserve File Name": {
"main": [
[
{
"node": "Merge XLSX and CSV",
"type": "main",
"index": 0
}
]
]
},
"Preserve CSV File Name": {
"main": [
[
{
"node": "Merge XLSX and CSV",
"type": "main",
"index": 1
}
]
]
},
"When clicking ‘Test workflow’": {
"main": [
[
{
"node": "Email Options",
"type": "main",
"index": 0
}
]
]
}
}
}
功能特点
- 自动检测新邮件
- AI智能内容分析
- 自定义分类规则
- 批量处理能力
- 详细的处理日志
技术分析
节点类型及作用
- Googledrive
- @N8N/N8N Nodes Langchain.Lmchatopenai
- If
- Merge
- Code
复杂度评估
配置难度:
维护难度:
扩展性:
实施指南
前置条件
- 有效的Gmail账户
- n8n平台访问权限
- Google API凭证
- AI分类服务订阅
配置步骤
- 在n8n中导入工作流JSON文件
- 配置Gmail节点的认证信息
- 设置AI分类器的API密钥
- 自定义分类规则和标签映射
- 测试工作流执行
- 配置定时触发器(可选)
关键参数
| 参数名称 | 默认值 | 说明 |
|---|---|---|
| maxEmails | 50 | 单次处理的最大邮件数量 |
| confidenceThreshold | 0.8 | 分类置信度阈值 |
| autoLabel | true | 是否自动添加标签 |
最佳实践
优化建议
- 定期更新AI分类模型以提高准确性
- 根据邮件量调整处理批次大小
- 设置合理的分类置信度阈值
- 定期清理过期的分类规则
安全注意事项
- 妥善保管API密钥和认证信息
- 限制工作流的访问权限
- 定期审查处理日志
- 启用双因素认证保护Gmail账户
性能优化
- 使用增量处理减少重复工作
- 缓存频繁访问的数据
- 并行处理多个邮件分类任务
- 监控系统资源使用情况
故障排除
常见问题
邮件未被正确分类
检查AI分类器的置信度阈值设置,适当降低阈值或更新训练数据。
Gmail认证失败
确认Google API凭证有效且具有正确的权限范围,重新进行OAuth授权。
调试技巧
- 启用详细日志记录查看每个步骤的执行情况
- 使用测试邮件验证分类逻辑
- 检查网络连接和API服务状态
- 逐步执行工作流定位问题节点
错误处理
工作流包含以下错误处理机制:
- 网络超时自动重试(最多3次)
- API错误记录和告警
- 处理失败邮件的隔离机制
- 异常情况下的回滚操作