7.0 KiB
description
| description |
|---|
| Design Make.com scenarios with AI assistance |
Make.com Scenario Builder
Create visual Make.com automation scenarios with detailed module configuration.
Usage
When the user requests a Make.com scenario, design a complete visual workflow with module-by-module instructions.
Scenario Templates
1. AI Email Auto-Responder
Flow:
Gmail: Watch emails
→ OpenAI: Generate response
→ Gmail: Send email
→ Google Sheets: Log conversation
Module Configuration:
-
Gmail: Watch emails
- Connection: Your Gmail
- Folder: INBOX
- Criteria: Unread messages
- Max results: 10
-
OpenAI: Create completion
- Connection: Your OpenAI API
- Model: gpt-4
- Max tokens: 500
- Temperature: 0.7
- Prompt:
Draft a professional response to:\n\nFrom: {{1.from}}\nSubject: {{1.subject}}\nBody: {{1.textPlain}}
-
Gmail: Send an email
- To:
{{1.from}} - Subject:
Re: {{1.subject}} - Content:
{{2.choices[0].message.content}}
- To:
-
Google Sheets: Add a row
- Spreadsheet: Email Log
- Sheet: Responses
- From:
{{1.from}} - Subject:
{{1.subject}} - Response:
{{2.choices[0].message.content}} - Date:
{{now}}
2. Lead Qualification with AI
Flow:
Webhook: Custom webhook
→ OpenAI: Score lead
→ Router:
├─ High score → Slack notification + HubSpot deal
├─ Medium score → Email nurture
└─ Low score → Archive to Airtable
Module Configuration:
-
Webhook
- Create a custom webhook
- Expected data: name, email, company, role, budget
-
OpenAI: Create completion
- Prompt:
Score this lead from 0-100:\nName: {{1.name}}\nCompany: {{1.company}}\nRole: {{1.role}}\nBudget: {{1.budget}}\n\nProvide only the numeric score. - Parse: Use Text parser to extract number
- Prompt:
-
Router
- Route 1 Filter:
{{3.score}}Greater than70 - Route 2 Filter:
{{3.score}}Between40and70 - Fallback: All other cases
- Route 1 Filter:
4A. Slack: Create a message (High route)
- Channel: #sales-leads
- Text:
Hot lead: {{1.name}} from {{1.company}} - Score: {{3.score}}
4B. HubSpot: Create a deal (High route)
- Deal name:
{{1.company}} - {{1.name}} - Stage: Qualification
- Amount:
{{1.budget}}
-
ActiveCampaign: Add contact to list (Medium route)
- Email:
{{1.email}} - List: Nurture Campaign
- Tags:
medium-priority,{{1.role}}
- Email:
-
Airtable: Create a record (Low route)
- Base: Leads
- Table: Archived
- Fields: Map all lead data + score
3. Content Distribution Pipeline
Flow:
RSS: Read feed
→ Filter: New items only
→ OpenAI: Rewrite for social
→ Iterator: For each platform
├─ Twitter: Post
├─ LinkedIn: Post
└─ Facebook: Post
Module Configuration:
-
RSS: Watch RSS feed items
- URL: Your RSS feed
- Maximum number of returned items: 5
-
Filter
- Condition:
{{1.published}}After{{addHours(now; -24)}} - Label: "Only items from last 24 hours"
- Condition:
-
OpenAI: Create completion
- Prompt:
Rewrite this article for social media (280 chars max):\n\nTitle: {{1.title}}\nContent: {{1.contentSnippet}}\n\nMake it engaging with relevant hashtags.
- Prompt:
-
Set multiple variables
- platforms:
["twitter", "linkedin", "facebook"] - content:
{{3.choices[0].message.content}} - link:
{{1.link}}
- platforms:
-
Iterator
- Array:
{{4.platforms}}
- Array:
6A. Twitter: Create a tweet
- Text:
{{4.content}}\n\n{{4.link}}
6B. LinkedIn: Create a post
- Text:
{{4.content}}\n\n{{4.link}}
6C. Facebook: Create a post
- Message:
{{4.content}} - Link:
{{4.link}}
4. Document Processing Workflow
Flow:
Google Drive: Watch files
→ Filter: PDFs only
→ OCR.space: Extract text
→ OpenAI: Summarize & extract data
→ Router:
├─ Success → Google Sheets: Log + Email summary
└─ Error → Slack: Notify failure
Module Configuration:
-
Google Drive: Watch files
- Folder: Inbox
- Types: application/pdf
- Limit: 10
-
Filter
- Condition: File name contains "invoice" OR "receipt"
-
HTTP: Make a request (OCR.space)
- URL: https://api.ocr.space/parse/image
- Method: POST
- Headers: apikey: YOUR_OCR_KEY
- Body: file:
{{1.data}}
-
OpenAI: Create completion
- Prompt:
Extract structured data from this document:\n\n{{3.ParsedResults[0].ParsedText}}\n\nProvide: Date, Amount, Vendor, Category - Format: JSON mode
- Prompt:
-
Google Sheets: Add a row
- Spreadsheet: Document Log
- Date:
{{4.date}} - Amount:
{{4.amount}} - Vendor:
{{4.vendor}} - Category:
{{4.category}}
-
Gmail: Send an email
- To: accounting@company.com
- Subject: New document processed:
{{1.name}} - Body: Summary of extracted data
Error Handler on all modules:
- Slack: Send message to #errors channel
5. Customer Support Automation
Flow:
Zendesk: Watch tickets
→ OpenAI: Classify urgency + category
→ Router:
├─ Urgent → Assign to senior agent + Slack alert
├─ Normal → Assign to queue + Draft response
└─ Low → Auto-respond with KB articles
Module Configuration:
-
Zendesk: Watch tickets
- Status: new
- Limit: 20
-
OpenAI: Create completion
- Prompt:
Classify this support ticket:\n\nSubject: {{1.subject}}\nDescription: {{1.description}}\n\nProvide:\n1. Urgency (high/medium/low)\n2. Category (billing/technical/general)\n3. Suggested response
- Prompt:
-
Router
- Route 1: Urgency = "high"
- Route 2: Urgency = "medium"
- Route 3: Urgency = "low"
4A. Zendesk: Update ticket (Urgent)
- Priority: urgent
- Assignee: Senior agent ID
4B. Slack: Create message (Urgent)
- Channel: #support-urgent
- Text:
Urgent ticket: {{1.subject}} - {{1.ticket_id}}
-
Zendesk: Update ticket (Normal)
- Priority: normal
- Comment:
{{2.suggested_response}} - Status: pending
-
Zendesk: Update ticket (Low)
- Comment: Auto-generated response + KB links
- Status: solved
Best Practices
- Always add error handlers to critical modules
- Use filters early to reduce operations
- Test with sample data before activating
- Monitor operations usage in Make dashboard
- Add notes to modules for documentation
- Use variables for reusable values
- Set up notifications for failures
- Clone scenarios before major changes
Output Format
When generating a scenario, provide:
## Scenario: [Name]
### Business Value
[What problem this solves]
### Visual Flow
[ASCII diagram or clear description]
### Module Configuration
[Detailed setup for each module]
### Data Mapping
[How data flows between modules]
### Testing Steps
1. [Step by step testing instructions]
### Cost Estimate
- Operations per execution: X
- Expected monthly runs: Y
- Total operations: Z
- Make plan needed: [Free/Core/Pro]
### Setup Time
Estimated: [X] minutes
This helps users implement Make.com scenarios quickly with clear, actionable instructions.